introduction to scientific method research paper

Writing an Introduction for a Scientific Paper

Dr. michelle harris, dr. janet batzli, biocore.

This section provides guidelines on how to construct a solid introduction to a scientific paper including background information, study question , biological rationale, hypothesis , and general approach . If the Introduction is done well, there should be no question in the reader’s mind why and on what basis you have posed a specific hypothesis.

Broad Question : based on an initial observation (e.g., “I see a lot of guppies close to the shore. Do guppies like living in shallow water?”). This observation of the natural world may inspire you to investigate background literature or your observation could be based on previous research by others or your own pilot study. Broad questions are not always included in your written text, but are essential for establishing the direction of your research.

Background Information : key issues, concepts, terminology, and definitions needed to understand the biological rationale for the experiment. It often includes a summary of findings from previous, relevant studies. Remember to cite references, be concise, and only include relevant information given your audience and your experimental design. Concisely summarized background information leads to the identification of specific scientific knowledge gaps that still exist. (e.g., “No studies to date have examined whether guppies do indeed spend more time in shallow water.”)

Testable Question : these questions are much more focused than the initial broad question, are specific to the knowledge gap identified, and can be addressed with data. (e.g., “Do guppies spend different amounts of time in water <1 meter deep as compared to their time in water that is >1 meter deep?”)

Biological Rationale : describes the purpose of your experiment distilling what is known and what is not known that defines the knowledge gap that you are addressing. The “BR” provides the logic for your hypothesis and experimental approach, describing the biological mechanism and assumptions that explain why your hypothesis should be true.

The biological rationale is based on your interpretation of the scientific literature, your personal observations, and the underlying assumptions you are making about how you think the system works. If you have written your biological rationale, your reader should see your hypothesis in your introduction section and say to themselves, “Of course, this hypothesis seems very logical based on the rationale presented.”

• A thorough rationale defines your assumptions about the system that have not been revealed in scientific literature or from previous systematic observation. These assumptions drive the direction of your specific hypothesis or general predictions.
• Defining the rationale is probably the most critical task for a writer, as it tells your reader why your research is biologically meaningful. It may help to think about the rationale as an answer to the questions— how is this investigation related to what we know, what assumptions am I making about what we don’t yet know, AND how will this experiment add to our knowledge? *There may or may not be broader implications for your study; be careful not to overstate these (see note on social justifications below).
• Expect to spend time and mental effort on this. You may have to do considerable digging into the scientific literature to define how your experiment fits into what is already known and why it is relevant to pursue.
• Be open to the possibility that as you work with and think about your data, you may develop a deeper, more accurate understanding of the experimental system. You may find the original rationale needs to be revised to reflect your new, more sophisticated understanding.
• As you progress through Biocore and upper level biology courses, your rationale should become more focused and matched with the level of study e ., cellular, biochemical, or physiological mechanisms that underlie the rationale. Achieving this type of understanding takes effort, but it will lead to better communication of your science.

***Special note on avoiding social justifications: You should not overemphasize the relevance of your experiment and the possible connections to large-scale processes. Be realistic and logical —do not overgeneralize or state grand implications that are not sensible given the structure of your experimental system. Not all science is easily applied to improving the human condition. Performing an investigation just for the sake of adding to our scientific knowledge (“pure or basic science”) is just as important as applied science. In fact, basic science often provides the foundation for applied studies.

Hypothesis / Predictions : specific prediction(s) that you will test during your experiment. For manipulative experiments, the hypothesis should include the independent variable (what you manipulate), the dependent variable(s) (what you measure), the organism or system , the direction of your results, and comparison to be made.

If you are doing a systematic observation , your hypothesis presents a variable or set of variables that you predict are important for helping you characterize the system as a whole, or predict differences between components/areas of the system that help you explain how the system functions or changes over time.

Experimental Approach : Briefly gives the reader a general sense of the experiment, the type of data it will yield, and the kind of conclusions you expect to obtain from the data. Do not confuse the experimental approach with the experimental protocol . The experimental protocol consists of the detailed step-by-step procedures and techniques used during the experiment that are to be reported in the Methods and Materials section.

Some Final Tips on Writing an Introduction

• As you progress through the Biocore sequence, for instance, from organismal level of Biocore 301/302 to the cellular level in Biocore 303/304, we expect the contents of your “Introduction” paragraphs to reflect the level of your coursework and previous writing experience. For example, in Biocore 304 (Cell Biology Lab) biological rationale should draw upon assumptions we are making about cellular and biochemical processes.
• Be Concise yet Specific: Remember to be concise and only include relevant information given your audience and your experimental design. As you write, keep asking, “Is this necessary information or is this irrelevant detail?” For example, if you are writing a paper claiming that a certain compound is a competitive inhibitor to the enzyme alkaline phosphatase and acts by binding to the active site, you need to explain (briefly) Michaelis-Menton kinetics and the meaning and significance of Km and Vmax. This explanation is not necessary if you are reporting the dependence of enzyme activity on pH because you do not need to measure Km and Vmax to get an estimate of enzyme activity.
• Another example: if you are writing a paper reporting an increase in Daphnia magna heart rate upon exposure to caffeine you need not describe the reproductive cycle of magna unless it is germane to your results and discussion. Be specific and concrete, especially when making introductory or summary statements.

Where Do You Discuss Pilot Studies? Many times it is important to do pilot studies to help you get familiar with your experimental system or to improve your experimental design. If your pilot study influences your biological rationale or hypothesis, you need to describe it in your Introduction. If your pilot study simply informs the logistics or techniques, but does not influence your rationale, then the description of your pilot study belongs in the Materials and Methods section.  

How will introductions be evaluated? The following is part of the rubric we will be using to evaluate your papers.

How to Write a Research Paper Introduction (with Examples)

The research paper introduction section, along with the Title and Abstract, can be considered the face of any research paper. The following article is intended to guide you in organizing and writing the research paper introduction for a quality academic article or dissertation.

The research paper introduction aims to present the topic to the reader. A study will only be accepted for publishing if you can ascertain that the available literature cannot answer your research question. So it is important to ensure that you have read important studies on that particular topic, especially those within the last five to ten years, and that they are properly referenced in this section. 1 What should be included in the research paper introduction is decided by what you want to tell readers about the reason behind the research and how you plan to fill the knowledge gap. The best research paper introduction provides a systemic review of existing work and demonstrates additional work that needs to be done. It needs to be brief, captivating, and well-referenced; a well-drafted research paper introduction will help the researcher win half the battle.

The introduction for a research paper is where you set up your topic and approach for the reader. It has several key goals:

• Present your research topic
• Capture reader interest
• Summarize existing research
• Position your own approach
• Define your specific research problem and problem statement
• Highlight the novelty and contributions of the study
• Give an overview of the paper’s structure

The research paper introduction can vary in size and structure depending on whether your paper presents the results of original empirical research or is a review paper. Some research paper introduction examples are only half a page while others are a few pages long. In many cases, the introduction will be shorter than all of the other sections of your paper; its length depends on the size of your paper as a whole.

Break through writer’s block. Write your research paper introduction with Paperpal’s Generative AI features

Table of Contents

What is the introduction for a research paper, what are the parts of introduction in the research, 1. introduce the research topic:, 2. determine a research niche:, 3. place your research within the research niche:, how to use paperpal to write the introduction section, frequently asked questions on research paper introduction, key points to remember.

The introduction in a research paper is placed at the beginning to guide the reader from a broad subject area to the specific topic that your research addresses. They present the following information to the reader

• Scope: The topic covered in the research paper
• Context: Background of your topic
• Importance: Why your research matters in that particular area of research and the industry problem that can be targeted

Why is the introduction important in a research paper?

The research paper introduction conveys a lot of information and can be considered an essential roadmap for the rest of your paper. A good introduction for a research paper is important for the following reasons:

• It stimulates your reader’s interest: A good introduction section can make your readers want to read your paper by capturing their interest. It informs the reader what they are going to learn and helps determine if the topic is of interest to them.
• It helps the reader understand the research background: Without a clear introduction, your readers may feel confused and even struggle when reading your paper. A good research paper introduction will prepare them for the in-depth research to come. It provides you the opportunity to engage with the readers and demonstrate your knowledge and authority on the specific topic.
• It explains why your research paper is worth reading: Your introduction can convey a lot of information to your readers. It introduces the topic, why the topic is important, and how you plan to proceed with your research.
• It helps guide the reader through the rest of the paper: The research paper introduction gives the reader a sense of the nature of the information that will support your arguments and the general organization of the paragraphs that will follow. It offers an overview of what to expect when reading the main body of your paper.

A good research paper introduction section should comprise three main elements: 2

• What is known: This sets the stage for your research. It informs the readers of what is known on the subject.
• What is lacking: This is aimed at justifying the reason for carrying out your research. This could involve investigating a new concept or method or building upon previous research.
• What you aim to do: This part briefly states the objectives of your research and its major contributions. Your detailed hypothesis will also form a part of this section.

Craft a compelling Introduction section with Paperpal. Try Now!

How to write a research paper introduction?

The first step in writing the research paper introduction is to inform the reader what your topic is and why it’s interesting or important. This is generally accomplished with a strong opening statement. The second step involves establishing the kinds of research that have been done and ending with limitations or gaps in the research that you intend to address. Finally, the research paper introduction clarifies how your own research fits in and what problem it addresses. If your research involved testing hypotheses, these should be stated along with your research question. The hypothesis should be presented in the past tense since it will have been tested by the time you are writing the research paper introduction.

The following key points, with examples, can guide you when writing the research paper introduction section:

• Highlight the importance of the research field or topic
• Describe the background of the topic
• Present an overview of current research on the topic

Example: The inclusion of experiential and competency-based learning has benefitted electronics engineering education. Industry partnerships provide an excellent alternative for students wanting to engage in solving real-world challenges. Industry-academia participation has grown in recent years due to the need for skilled engineers with practical training and specialized expertise. However, from the educational perspective, many activities are needed to incorporate sustainable development goals into the university curricula and consolidate learning innovation in universities.

• Reveal a gap in existing research or oppose an existing assumption
• Formulate the research question

Example: There have been plausible efforts to integrate educational activities in higher education electronics engineering programs. However, very few studies have considered using educational research methods for performance evaluation of competency-based higher engineering education, with a focus on technical and or transversal skills. To remedy the current need for evaluating competencies in STEM fields and providing sustainable development goals in engineering education, in this study, a comparison was drawn between study groups without and with industry partners.

• State the purpose of your study
• Highlight the key characteristics of your study
• Describe important results
• Highlight the novelty of the study.
• Offer a brief overview of the structure of the paper.

Example: The study evaluates the main competency needed in the applied electronics course, which is a fundamental core subject for many electronics engineering undergraduate programs. We compared two groups, without and with an industrial partner, that offered real-world projects to solve during the semester. This comparison can help determine significant differences in both groups in terms of developing subject competency and achieving sustainable development goals.

Write a Research Paper Introduction in Minutes with Paperpal

Paperpal is a generative AI-powered academic writing assistant. It’s trained on millions of published scholarly articles and over 20 years of STM experience. Paperpal helps authors write better and faster with:

• Real-time writing suggestions
• In-depth checks for language and grammar correction
• Paraphrasing to add variety, ensure academic tone, and trim text to meet journal limits

With Paperpal, create a research paper introduction effortlessly. In this step-by-step guide, we’ll walk you through how Paperpal transforms your initial ideas into a polished and publication-ready introduction.

Step 1: Sign up on Paperpal and click on the Copilot feature, under this choose Outlines > Research Article > Introduction

Step 2: Add your unstructured notes or initial draft, whether in English or another language, to Paperpal, which is to be used as the base for your content.

Step 3: Fill in the specifics, such as your field of study, brief description or details you want to include, which will help the AI generate the outline for your Introduction.

Step 4: Use this outline and sentence suggestions to develop your content, adding citations where needed and modifying it to align with your specific research focus.

Step 5: Turn to Paperpal’s granular language checks to refine your content, tailor it to reflect your personal writing style, and ensure it effectively conveys your message.

You can use the same process to develop each section of your article, and finally your research paper in half the time and without any of the stress.

Craft accurate research paper introductions with Paperpal. Start writing now!

The purpose of the research paper introduction is to introduce the reader to the problem definition, justify the need for the study, and describe the main theme of the study. The aim is to gain the reader’s attention by providing them with necessary background information and establishing the main purpose and direction of the research.

The length of the research paper introduction can vary across journals and disciplines. While there are no strict word limits for writing the research paper introduction, an ideal length would be one page, with a maximum of 400 words over 1-4 paragraphs. Generally, it is one of the shorter sections of the paper as the reader is assumed to have at least a reasonable knowledge about the topic. 2 For example, for a study evaluating the role of building design in ensuring fire safety, there is no need to discuss definitions and nature of fire in the introduction; you could start by commenting upon the existing practices for fire safety and how your study will add to the existing knowledge and practice.

When deciding what to include in the research paper introduction, the rest of the paper should also be considered. The aim is to introduce the reader smoothly to the topic and facilitate an easy read without much dependency on external sources. 3 Below is a list of elements you can include to prepare a research paper introduction outline and follow it when you are writing the research paper introduction. Topic introduction: This can include key definitions and a brief history of the topic. Research context and background: Offer the readers some general information and then narrow it down to specific aspects. Details of the research you conducted: A brief literature review can be included to support your arguments or line of thought. Rationale for the study: This establishes the relevance of your study and establishes its importance. Importance of your research: The main contributions are highlighted to help establish the novelty of your study Research hypothesis: Introduce your research question and propose an expected outcome. Organization of the paper: Include a short paragraph of 3-4 sentences that highlights your plan for the entire paper

Cite only works that are most relevant to your topic; as a general rule, you can include one to three. Note that readers want to see evidence of original thinking. So it is better to avoid using too many references as it does not leave much room for your personal standpoint to shine through. Citations in your research paper introduction support the key points, and the number of citations depend on the subject matter and the point discussed. If the research paper introduction is too long or overflowing with citations, it is better to cite a few review articles rather than the individual articles summarized in the review. A good point to remember when citing research papers in the introduction section is to include at least one-third of the references in the introduction.

The literature review plays a significant role in the research paper introduction section. A good literature review accomplishes the following: Introduces the topic – Establishes the study’s significance – Provides an overview of the relevant literature – Provides context for the study using literature – Identifies knowledge gaps However, remember to avoid making the following mistakes when writing a research paper introduction: Do not use studies from the literature review to aggressively support your research Avoid direct quoting Do not allow literature review to be the focus of this section. Instead, the literature review should only aid in setting a foundation for the manuscript.

Remember the following key points for writing a good research paper introduction: 4

• Avoid stuffing too much general information: Avoid including what an average reader would know and include only that information related to the problem being addressed in the research paper introduction. For example, when describing a comparative study of non-traditional methods for mechanical design optimization, information related to the traditional methods and differences between traditional and non-traditional methods would not be relevant. In this case, the introduction for the research paper should begin with the state-of-the-art non-traditional methods and methods to evaluate the efficiency of newly developed algorithms.
• Avoid packing too many references: Cite only the required works in your research paper introduction. The other works can be included in the discussion section to strengthen your findings.
• Avoid extensive criticism of previous studies: Avoid being overly critical of earlier studies while setting the rationale for your study. A better place for this would be the Discussion section, where you can highlight the advantages of your method.
• Avoid describing conclusions of the study: When writing a research paper introduction remember not to include the findings of your study. The aim is to let the readers know what question is being answered. The actual answer should only be given in the Results and Discussion section.

To summarize, the research paper introduction section should be brief yet informative. It should convince the reader the need to conduct the study and motivate him to read further. If you’re feeling stuck or unsure, choose trusted AI academic writing assistants like Paperpal to effortlessly craft your research paper introduction and other sections of your research article.

1. Jawaid, S. A., & Jawaid, M. (2019). How to write introduction and discussion. Saudi Journal of Anaesthesia, 13(Suppl 1), S18.

2. Dewan, P., & Gupta, P. (2016). Writing the title, abstract and introduction: Looks matter!. Indian pediatrics, 53, 235-241.

3. Cetin, S., & Hackam, D. J. (2005). An approach to the writing of a scientific Manuscript1. Journal of Surgical Research, 128(2), 165-167.

4. Bavdekar, S. B. (2015). Writing introduction: Laying the foundations of a research paper. Journal of the Association of Physicians of India, 63(7), 44-6.

Paperpal is a comprehensive AI writing toolkit that helps students and researchers achieve 2x the writing in half the time. It leverages 21+ years of STM experience and insights from millions of research articles to provide in-depth academic writing, language editing, and submission readiness support to help you write better, faster.  

Get accurate academic translations, rewriting support, grammar checks, vocabulary suggestions, and generative AI assistance that delivers human precision at machine speed. Try for free or upgrade to Paperpal Prime starting at US$19 a month to access premium features, including consistency, plagiarism, and 30+ submission readiness checks to help you succeed.  

Experience the future of academic writing – Sign up to Paperpal and start writing for free!  

Related Reads:

• Scientific Writing Style Guides Explained
• Ethical Research Practices For Research with Human Subjects
• 8 Most Effective Ways to Increase Motivation for Thesis Writing 
• 6 Tips for Post-Doc Researchers to Take Their Career to the Next Level

Practice vs. Practise: Learn the Difference

Academic paraphrasing: why paperpal’s rewrite should be your first choice , you may also like, how to cite in apa format (7th edition):..., how to write your research paper in apa..., how to choose a dissertation topic, how to write a phd research proposal, how to write an academic paragraph (step-by-step guide), research funding basics: what should a grant proposal..., how to write an abstract in research papers..., how to write dissertation acknowledgements, how to write the first draft of a..., mla works cited page: format, template & examples.

• Langson Library
• Science Library
• Grunigen Medical Library
• Law Library
• Connect From Off-Campus
• Accessibility
• Gateway Study Center

Email this link

Writing a scientific paper.

• Writing a lab report
• INTRODUCTION

Writing a "good" methods section

"methods checklist" from: how to write a good scientific paper. chris a. mack. spie. 2018..

• LITERATURE CITED
• Bibliography of guides to scientific writing and presenting
• Peer Review
• Presentations
• Lab Report Writing Guides on the Web

The purpose is to provide enough detail that a competent worker could repeat the experiment. Many of your readers will skip this section because they already know from the Introduction the general methods you used. However careful writing of this section is important because for your results to be of scientific merit they must be reproducible. Otherwise your paper does not represent good science.

• Exact technical specifications and quantities and source or method of preparation
• Describe equipment used and provide illustrations where relevant.
• Chronological presentation (but related methods described together)
• Questions about "how" and "how much" are answered for the reader and not left for them to puzzle over
• Discuss statistical methods only if unusual or advanced
• When a large number of components are used prepare tables for the benefit of the reader
• Do not state the action without stating the agent of the action
• Describe how the results were generated with sufficient detail so that an independent researcher (working in the same field) could reproduce the results sufficiently to allow validation of the conclusions.
• Can the reader assess internal validity (conclusions are supported by the results presented)?
• Can the reader assess external validity (conclusions are properly generalized beyond these specific results)?
• Has the chosen method been justified?
• Are data analysis and statistical approaches justified, with assumptions and biases considered?
• Avoid: including results in the Method section; including extraneous details (unnecessary to enable reproducibility or judge validity); treating the method as a chronological history of events; unneeded references to commercial products; references to “proprietary” products or processes unavailable to the reader. 
• << Previous: INTRODUCTION
• Next: RESULTS >>
• Last Updated: Aug 4, 2023 9:33 AM
• URL: https://guides.lib.uci.edu/scientificwriting

Off-campus? Please use the Software VPN and choose the group UCIFull to access licensed content. For more information, please Click here

Software VPN is not available for guests, so they may not have access to some content when connecting from off-campus.

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• Prev Chronic Dis

Successful Scientific Writing and Publishing: A Step-by-Step Approach

John k. iskander.

1 Centers for Disease Control and Prevention, Atlanta, Georgia

Sara Beth Wolicki

2 Association of Schools and Programs of Public Health, Washington, District of Columbia

Rebecca T. Leeb

Paul z. siegel.

Scientific writing and publication are essential to advancing knowledge and practice in public health, but prospective authors face substantial challenges. Authors can overcome barriers, such as lack of understanding about scientific writing and the publishing process, with training and resources. The objective of this article is to provide guidance and practical recommendations to help both inexperienced and experienced authors working in public health settings to more efficiently publish the results of their work in the peer-reviewed literature. We include an overview of basic scientific writing principles, a detailed description of the sections of an original research article, and practical recommendations for selecting a journal and responding to peer review comments. The overall approach and strategies presented are intended to contribute to individual career development while also increasing the external validity of published literature and promoting quality public health science.

Introduction

Publishing in the peer-reviewed literature is essential to advancing science and its translation to practice in public health ( 1 , 2 ). The public health workforce is diverse and practices in a variety of settings ( 3 ). For some public health professionals, writing and publishing the results of their work is a requirement. Others, such as program managers, policy makers, or health educators, may see publishing as being outside the scope of their responsibilities ( 4 ).

Disseminating new knowledge via writing and publishing is vital both to authors and to the field of public health ( 5 ). On an individual level, publishing is associated with professional development and career advancement ( 6 ). Publications share new research, results, and methods in a trusted format and advance scientific knowledge and practice ( 1 , 7 ). As more public health professionals are empowered to publish, the science and practice of public health will advance ( 1 ).

Unfortunately, prospective authors face barriers to publishing their work, including navigating the process of scientific writing and publishing, which can be time-consuming and cumbersome. Often, public health professionals lack both training opportunities and understanding of the process ( 8 ). To address these barriers and encourage public health professionals to publish their findings, the senior author (P.Z.S.) and others developed Successful Scientific Writing (SSW), a course about scientific writing and publishing. Over the past 30 years, this course has been taught to thousands of public health professionals, as well as hundreds of students at multiple graduate schools of public health. An unpublished longitudinal survey of course participants indicated that two-thirds agreed that SSW had helped them to publish a scientific manuscript or have a conference abstract accepted. The course content has been translated into this manuscript. The objective of this article is to provide prospective authors with the tools needed to write original research articles of high quality that have a good chance of being published.

Basic Recommendations for Scientific Writing

Prospective authors need to know and tailor their writing to the audience. When writing for scientific journals, 4 fundamental recommendations are: clearly stating the usefulness of the study, formulating a key message, limiting unnecessary words, and using strategic sentence structure.

To demonstrate usefulness, focus on how the study addresses a meaningful gap in current knowledge or understanding. What critical piece of information does the study provide that will help solve an important public health problem? For example, if a particular group of people is at higher risk for a specific condition, but the magnitude of that risk is unknown, a study to quantify the risk could be important for measuring the population’s burden of disease.

Scientific articles should have a clear and concise take-home message. Typically, this is expressed in 1 to 2 sentences that summarize the main point of the paper. This message can be used to focus the presentation of background information, results, and discussion of findings. As an early step in the drafting of an article, we recommend writing out the take-home message and sharing it with co-authors for their review and comment. Authors who know their key point are better able to keep their writing within the scope of the article and present information more succinctly. Once an initial draft of the manuscript is complete, the take-home message can be used to review the content and remove needless words, sentences, or paragraphs.

Concise writing improves the clarity of an article. Including additional words or clauses can divert from the main message and confuse the reader. Additionally, journal articles are typically limited by word count. The most important words and phrases to eliminate are those that do not add meaning, or are duplicative. Often, cutting adjectives or parenthetical statements results in a more concise paper that is also easier to read.

Sentence structure strongly influences the readability and comprehension of journal articles. Twenty to 25 words is a reasonable range for maximum sentence length. Limit the number of clauses per sentence, and place the most important or relevant clause at the end of the sentence ( 9 ). Consider the sentences:

• By using these tips and tricks, an author may write and publish an additional 2 articles a year.
• An author may write and publish an additional 2 articles a year by using these tips and tricks.

The focus of the first sentence is on the impact of using the tips and tricks, that is, 2 more articles published per year. In contrast, the second sentence focuses on the tips and tricks themselves.

Authors should use the active voice whenever possible. Consider the following example:

• Active voice: Authors who use the active voice write more clearly.
• Passive voice: Clarity of writing is promoted by the use of the active voice.

The active voice specifies who is doing the action described in the sentence. Using the active voice improves clarity and understanding, and generally uses fewer words. Scientific writing includes both active and passive voice, but authors should be intentional with their use of either one.

Sections of an Original Research Article

Original research articles make up most of the peer-reviewed literature ( 10 ), follow a standardized format, and are the focus of this article. The 4 main sections are the introduction, methods, results, and discussion, sometimes referred to by the initialism, IMRAD. These 4 sections are referred to as the body of an article. Two additional components of all peer-reviewed articles are the title and the abstract. Each section’s purpose and key components, along with specific recommendations for writing each section, are listed below.

Title. The purpose of a title is twofold: to provide an accurate and informative summary and to attract the target audience. Both prospective readers and database search engines use the title to screen articles for relevance ( 2 ). All titles should clearly state the topic being studied. The topic includes the who, what, when, and where of the study. Along with the topic, select 1 or 2 of the following items to include within the title: methods, results, conclusions, or named data set or study. The items chosen should emphasize what is new and useful about the study. Some sources recommend limiting the title to less than 150 characters ( 2 ). Articles with shorter titles are more frequently cited than articles with longer titles ( 11 ). Several title options are possible for the same study ( Figure ).

Two examples of title options for a single study.

Abstract . The abstract serves 2 key functions. Journals may screen articles for potential publication by using the abstract alone ( 12 ), and readers may use the abstract to decide whether to read further. Therefore, it is critical to produce an accurate and clear abstract that highlights the major purpose of the study, basic procedures, main findings, and principal conclusions ( 12 ). Most abstracts have a word limit and can be either structured following IMRAD, or unstructured. The abstract needs to stand alone from the article and tell the most important parts of the scientific story up front.

Introduction . The purpose of the introduction is to explain how the study sought to create knowledge that is new and useful. The introduction section may often require only 3 paragraphs. First, describe the scope, nature, or magnitude of the problem being addressed. Next, clearly articulate why better understanding this problem is useful, including what is currently known and the limitations of relevant previous studies. Finally, explain what the present study adds to the knowledge base. Explicitly state whether data were collected in a unique way or obtained from a previously unstudied data set or population. Presenting both the usefulness and novelty of the approach taken will prepare the reader for the remaining sections of the article.

Methods . The methods section provides the information necessary to allow others, given the same data, to recreate the analysis. It describes exactly how data relevant to the study purpose were collected, organized, and analyzed. The methods section describes the process of conducting the study — from how the sample was selected to which statistical methods were used to analyze the data. Authors should clearly name, define, and describe each study variable. Some journals allow detailed methods to be included in an appendix or supplementary document. If the analysis involves a commonly used public health data set, such as the Behavioral Risk Factor Surveillance System ( 13 ), general aspects of the data set can be provided to readers by using references. Because what was done is typically more important than who did it, use of the passive voice is often appropriate when describing methods. For example, “The study was a group randomized, controlled trial. A coin was tossed to select an intervention group and a control group.”

Results . The results section describes the main outcomes of the study or analysis but does not interpret the findings or place them in the context of previous research. It is important that the results be logically organized. Suggested organization strategies include presenting results pertaining to the entire population first, and then subgroup analyses, or presenting results according to increasing complexity of analysis, starting with demographic results before proceeding to univariate and multivariate analyses. Authors wishing to draw special attention to novel or unexpected results can present them first.

One strategy for writing the results section is to start by first drafting the figures and tables. Figures, which typically show trends or relationships, and tables, which show specific data points, should each support a main outcome of the study. Identify the figures and tables that best describe the findings and relate to the study’s purpose, and then develop 1 to 2 sentences summarizing each one. Data not relevant to the study purpose may be excluded, summarized briefly in the text, or included in supplemental data sets. When finalizing figures, ensure that axes are labeled and that readers can understand figures without having to refer to accompanying text.

Discussion . In the discussion section, authors interpret the results of their study within the context of both the related literature and the specific scientific gap the study was intended to fill. The discussion does not introduce results that were not presented in the results section. One way authors can focus their discussion is to limit this section to 4 paragraphs: start by reinforcing the study’s take-home message(s), contextualize key results within the relevant literature, state the study limitations, and lastly, make recommendations for further research or policy and practice changes. Authors can support assertions made in the discussion with either their own findings or by referencing related research. By interpreting their own study results and comparing them to others in the literature, authors can emphasize findings that are unique, useful, and relevant. Present study limitations clearly and without apology. Finally, state the implications of the study and provide recommendations or next steps, for example, further research into remaining gaps or changes to practice or policy. Statements or recommendations regarding policy may use the passive voice, especially in instances where the action to be taken is more important than who will implement the action.

Beginning the Writing Process

The process of writing a scientific article occurs before, during, and after conducting the study or analyses. Conducting a literature review is crucial to confirm the existence of the evidence gap that the planned analysis seeks to fill. Because literature searches are often part of applying for research funding or developing a study protocol, the citations used in the grant application or study proposal can also be used in subsequent manuscripts. Full-text databases such as PubMed Central ( 14 ), NIH RePORT ( 15 ), and CDC Stacks ( 16 ) can be useful when performing literature reviews. Authors should familiarize themselves with databases that are accessible through their institution and any assistance that may be available from reference librarians or interlibrary loan systems. Using citation management software is one way to establish and maintain a working reference list. Authors should clearly understand the distinction between primary and secondary references, and ensure that they are knowledgeable about the content of any primary or secondary reference that they cite.

Review of the literature may continue while organizing the material and writing begins. One way to organize material is to create an outline for the paper. Another way is to begin drafting small sections of the article such as the introduction. Starting a preliminary draft forces authors to establish the scope of their analysis and clearly articulate what is new and novel about the study. Furthermore, using information from the study protocol or proposal allows authors to draft the methods and part of the results sections while the study is in progress. Planning potential data comparisons or drafting “table shells” will help to ensure that the study team has collected all the necessary data. Drafting these preliminary sections early during the writing process and seeking feedback from co-authors and colleagues may help authors avoid potential pitfalls, including misunderstandings about study objectives.

The next step is to conduct the study or analyses and use the resulting data to fill in the draft table shells. The initial results will most likely require secondary analyses, that is, exploring the data in ways in addition to those originally planned. Authors should ensure that they regularly update their methods section to describe all changes to data analysis.

After completing table shells, authors should summarize the key finding of each table or figure in a sentence or two. Presenting preliminary results at meetings, conferences, and internal seminars is an established way to solicit feedback. Authors should pay close attention to questions asked by the audience, treating them as an informal opportunity for peer review. On the basis of the questions and feedback received, authors can incorporate revisions and improvements into subsequent drafts of the manuscript.

The relevant literature should be revisited periodically while writing to ensure knowledge of the most recent publications about the manuscript topic. Authors should focus on content and key message during the process of writing the first draft and should not spend too much time on issues of grammar or style. Drafts, or portions of drafts, should be shared frequently with trusted colleagues. Their recommendations should be reviewed and incorporated when they will improve the manuscript’s overall clarity.

For most authors, revising drafts of the manuscript will be the most time-consuming task involved in writing a paper. By regularly checking in with coauthors and colleagues, authors can adopt a systematic approach to rewriting. When the author has completed a draft of the manuscript, he or she should revisit the key take-home message to ensure that it still matches the final data and analysis. At this point, final comments and approval of the manuscript by coauthors can be sought.

Authors should then seek to identify journals most likely to be interested in considering the study for publication. Initial questions to consider when selecting a journal include:

• Which audience is most interested in the paper’s message?
• Would clinicians, public health practitioners, policy makers, scientists, or a broader audience find this useful in their field or practice?
• Do colleagues have prior experience submitting a manuscript to this journal?
• Is the journal indexed and peer-reviewed?
• Is the journal subscription or open-access and are there any processing fees?
• How competitive is the journal?

Authors should seek to balance the desire to be published in a top-tier journal (eg, Journal of the American Medical Association, BMJ, or Lancet) against the statistical likelihood of rejection. Submitting the paper initially to a journal more focused on the paper’s target audience may result in a greater chance of acceptance, as well as more timely dissemination of findings that can be translated into practice. Most of the 50 to 75 manuscripts published each week by authors from the Centers for Disease Control and Prevention (CDC) are published in specialty and subspecialty journals, rather than in top-tier journals ( 17 ).

The target journal’s website will include author guidelines, which will contain specific information about format requirements (eg, font, line spacing, section order, reference style and limit, table and figure formatting), authorship criteria, article types, and word limits for articles and abstracts.

We recommend returning to the previously drafted abstract and ensuring that it complies with the journal’s format and word limit. Authors should also verify that any changes made to the methods or results sections during the article’s drafting are reflected in the final version of the abstract. The abstract should not be written hurriedly just before submitting the manuscript; it is often apparent to editors and reviewers when this has happened. A cover letter to accompany the submission should be drafted; new and useful findings and the key message should be included.

Before submitting the manuscript and cover letter, authors should perform a final check to ensure that their paper complies with all journal requirements. Journals may elect to reject certain submissions on the basis of review of the abstract, or may send them to peer reviewers (typically 2 or 3) for consultation. Occasionally, on the basis of peer reviews, the journal will request only minor changes before accepting the paper for publication. Much more frequently, authors will receive a request to revise and resubmit their manuscript, taking into account peer review comments. Authors should recognize that while revise-and-resubmit requests may state that the manuscript is not acceptable in its current form, this does not constitute a rejection of the article. Authors have several options in responding to peer review comments:

• Performing additional analyses and updating the article appropriately
• Declining to perform additional analyses, but providing an explanation (eg, because the requested analysis goes beyond the scope of the article)
• Providing updated references
• Acknowledging reviewer comments that are simply comments without making changes

In addition to submitting a revised manuscript, authors should include a cover letter in which they list peer reviewer comments, along with the revisions they have made to the manuscript and their reply to the comment. The tone of such letters should be thankful and polite, but authors should make clear areas of disagreement with peer reviewers, and explain why they disagree. During the peer review process, authors should continue to consult with colleagues, especially ones who have more experience with the specific journal or with the peer review process.

There is no secret to successful scientific writing and publishing. By adopting a systematic approach and by regularly seeking feedback from trusted colleagues throughout the study, writing, and article submission process, authors can increase their likelihood of not only publishing original research articles of high quality but also becoming more scientifically productive overall.

Acknowledgments

The authors acknowledge PCD ’s former Associate Editor, Richard A. Goodman, MD, MPH, who, while serving as Editor in Chief of CDC’s Morbidity and Mortality Weekly Report Series, initiated a curriculum on scientific writing for training CDC’s Epidemic Intelligence Service Officers and other CDC public health professionals, and with whom the senior author of this article (P.Z.S.) collaborated in expanding training methods and contents, some of which are contained in this article. The authors acknowledge Juan Carlos Zevallos, MD, for his thoughtful critique and careful editing of previous Successful Scientific Writing materials. We also thank Shira Eisenberg for editorial assistance with the manuscript. This publication was supported by the Cooperative Agreement no. 1U360E000002 from CDC and the Association of Schools and Programs of Public Health. The findings and conclusions of this article do not necessarily represent the official views of CDC or the Association of Schools and Programs of Public Health. Names of journals and citation databases are provided for identification purposes only and do not constitute any endorsement by CDC.

The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the U.S. Department of Health and Human Services, the Public Health Service, the Centers for Disease Control and Prevention, or the authors' affiliated institutions.

Suggested citation for this article: Iskander JK, Wolicki SB, Leeb RT, Siegel PZ. Successful Scientific Writing and Publishing: A Step-by-Step Approach. Prev Chronic Dis 2018;15:180085. DOI: https://doi.org/10.5888/pcd15.180085 .

Educational resources and simple solutions for your research journey

How to Write the Methods Section of a Research Paper

Writing a research paper is both an art and a skill, and knowing how to write the methods section of a research paper is the first crucial step in mastering scientific writing. If, like the majority of early career researchers, you believe that the methods section is the simplest to write and needs little in the way of careful consideration or thought, this article will help you understand it is not 1 .

We have all probably asked our supervisors, coworkers, or search engines “ how to write a methods section of a research paper ” at some point in our scientific careers, so you are not alone if that’s how you ended up here.  Even for seasoned researchers, selecting what to include in the methods section from a wealth of experimental information can occasionally be a source of distress and perplexity.   

Additionally, journal specifications, in some cases, may make it more of a requirement rather than a choice to provide a selective yet descriptive account of the experimental procedure. Hence, knowing these nuances of how to write the methods section of a research paper is critical to its success. The methods section of the research paper is not supposed to be a detailed heavy, dull section that some researchers tend to write; rather, it should be the central component of the study that justifies the validity and reliability of the research.

Are you still unsure of how the methods section of a research paper forms the basis of every investigation? Consider the last article you read but ignore the methods section and concentrate on the other parts of the paper . Now think whether you could repeat the study and be sure of the credibility of the findings despite knowing the literature review and even having the data in front of you. You have the answer!   

Having established the importance of the methods section , the next question is how to write the methods section of a research paper that unifies the overall study. The purpose of the methods section , which was earlier called as Materials and Methods , is to describe how the authors went about answering the “research question” at hand. Here, the objective is to tell a coherent story that gives a detailed account of how the study was conducted, the rationale behind specific experimental procedures, the experimental setup, objects (variables) involved, the research protocol employed, tools utilized to measure, calculations and measurements, and the analysis of the collected data 2 .

In this article, we will take a deep dive into this topic and provide a detailed overview of how to write the methods section of a research paper . For the sake of clarity, we have separated the subject into various sections with corresponding subheadings.  

Table of Contents

What is the methods section of a research paper ?  

The methods section is a fundamental section of any paper since it typically discusses the ‘ what ’, ‘ how ’, ‘ which ’, and ‘ why ’ of the study, which is necessary to arrive at the final conclusions. In a research article, the introduction, which serves to set the foundation for comprehending the background and results is usually followed by the methods section, which precedes the result and discussion sections. The methods section must explicitly state what was done, how it was done, which equipment, tools and techniques were utilized, how were the measurements/calculations taken, and why specific research protocols, software, and analytical methods were employed.  

Why is the methods section important?  

The primary goal of the methods section is to provide pertinent details about the experimental approach so that the reader may put the results in perspective and, if necessary, replicate the findings 3 .  This section offers readers the chance to evaluate the reliability and validity of any study. In short, it also serves as the study’s blueprint, assisting researchers who might be unsure about any other portion in establishing the study’s context and validity. The methods plays a rather crucial role in determining the fate of the article; an incomplete and unreliable methods section can frequently result in early rejections and may lead to numerous rounds of modifications during the publication process. This means that the reviewers also often use methods section to assess the reliability and validity of the research protocol and the data analysis employed to address the research topic. In other words, the purpose of the methods section is to demonstrate the research acumen and subject-matter expertise of the author(s) in their field.  

Structure of methods section of a research paper  

Similar to the research paper, the methods section also follows a defined structure; this may be dictated by the guidelines of a specific journal or can be presented in a chronological or thematic manner based on the study type. When writing the methods section , authors should keep in mind that they are telling a story about how the research was conducted. They should only report relevant information to avoid confusing the reader and include details that would aid in connecting various aspects of the entire research activity together. It is generally advisable to present experiments in the order in which they were conducted. This facilitates the logical flow of the research and allows readers to follow the progression of the study design.   

It is also essential to clearly state the rationale behind each experiment and how the findings of earlier experiments informed the design or interpretation of later experiments. This allows the readers to understand the overall purpose of the study design and the significance of each experiment within that context. However, depending on the particular research question and method, it may make sense to present information in a different order; therefore, authors must select the best structure and strategy for their individual studies.   

In cases where there is a lot of information, divide the sections into subheadings to cover the pertinent details. If the journal guidelines pose restrictions on the word limit , additional important information can be supplied in the supplementary files. A simple rule of thumb for sectioning the method section is to begin by explaining the methodological approach ( what was done ), describing the data collection methods ( how it was done ), providing the analysis method ( how the data was analyzed ), and explaining the rationale for choosing the methodological strategy. This is described in detail in the upcoming sections.    

How to write the methods section of a research paper  

Contrary to widespread assumption, the methods section of a research paper should be prepared once the study is complete to prevent missing any key parameter. Hence, please make sure that all relevant experiments are done before you start writing a methods section . The next step for authors is to look up any applicable academic style manuals or journal-specific standards to ensure that the methods section is formatted correctly. The methods section of a research paper typically constitutes materials and methods; while writing this section, authors usually arrange the information under each category.

The materials category describes the samples, materials, treatments, and instruments, while experimental design, sample preparation, data collection, and data analysis are a part of the method category. According to the nature of the study, authors should include additional subsections within the methods section, such as ethical considerations like the declaration of Helsinki (for studies involving human subjects), demographic information of the participants, and any other crucial information that can affect the output of the study. Simply put, the methods section has two major components: content and format. Here is an easy checklist for you to consider if you are struggling with how to write the methods section of a research paper .   

• Explain the research design, subjects, and sample details  
• Include information on inclusion and exclusion criteria  
• Mention ethical or any other permission required for the study  
• Include information about materials, experimental setup, tools, and software  
• Add details of data collection and analysis methods  
• Incorporate how research biases were avoided or confounding variables were controlled  
• Evaluate and justify the experimental procedure selected to address the research question  
• Provide precise and clear details of each experiment  
• Flowcharts, infographics, or tables can be used to present complex information     
• Use past tense to show that the experiments have been done   
• Follow academic style guides (such as APA or MLA ) to structure the content  
• Citations should be included as per standard protocols in the field  

Now that you know how to write the methods section of a research paper , let’s address another challenge researchers face while writing the methods section —what to include in the methods section .  How much information is too much is not always obvious when it comes to trying to include data in the methods section of a paper. In the next section, we examine this issue and explore potential solutions.   

What to include in the methods section of a research paper  

The technical nature of the methods section occasionally makes it harder to present the information clearly and concisely while staying within the study context. Many young researchers tend to veer off subject significantly, and they frequently commit the sin of becoming bogged down in itty bitty details, making the text harder to read and impairing its overall flow. However, the best way to write the methods section is to start with crucial components of the experiments. If you have trouble deciding which elements are essential, think about leaving out those that would make it more challenging to comprehend the context or replicate the results. The top-down approach helps to ensure all relevant information is incorporated and vital information is not lost in technicalities. Next, remember to add details that are significant to assess the validity and reliability of the study. Here is a simple checklist for you to follow ( bonus tip: you can also make a checklist for your own study to avoid missing any critical information while writing the methods section ).  

• Structuring the methods section : Authors should diligently follow journal guidelines and adhere to the specific author instructions provided when writing the methods section . Journals typically have specific guidelines for formatting the methods section ; for example, Frontiers in Plant Sciences advises arranging the materials and methods section by subheading and citing relevant literature. There are several standardized checklists available for different study types in the biomedical field, including CONSORT (Consolidated Standards of Reporting Trials) for randomized clinical trials, PRISMA (Preferred Reporting Items for Systematic reviews and Meta-Analysis) for systematic reviews and meta-analysis, and STROBE (STrengthening the Reporting of OBservational studies in Epidemiology) for cohort, case-control, cross-sectional studies. Before starting the methods section , check the checklist available in your field that can function as a guide.     
• Organizing different sections to tell a story : Once you are sure of the format required for structuring the methods section , the next is to present the sections in a logical manner; as mentioned earlier, the sections can be organized according to the chronology or themes. In the chronological arrangement, you should discuss the methods in accordance with how the experiments were carried out. An example of the method section of a research paper of an animal study should first ideally include information about the species, weight, sex, strain, and age. Next, the number of animals, their initial conditions, and their living and housing conditions should also be mentioned. Second, how the groups are assigned and the intervention (drug treatment, stress, or other) given to each group, and finally, the details of tools and techniques used to measure, collect, and analyze the data. Experiments involving animal or human subjects should additionally state an ethics approval statement. It is best to arrange the section using the thematic approach when discussing distinct experiments not following a sequential order.  
• Define and explain the objects and procedure: Experimental procedure should clearly be stated in the methods section . Samples, necessary preparations (samples, treatment, and drug), and methods for manipulation need to be included. All variables (control, dependent, independent, and confounding) must be clearly defined, particularly if the confounding variables can affect the outcome of the study.  
• Match the order of the methods section with the order of results: Though not mandatory, organizing the manuscript in a logical and coherent manner can improve the readability and clarity of the paper. This can be done by following a consistent structure throughout the manuscript; readers can easily navigate through the different sections and understand the methods and results in relation to each other. Using experiment names as headings for both the methods and results sections can also make it simpler for readers to locate specific information and corroborate it if needed.   
• Relevant information must always be included: The methods section should have information on all experiments conducted and their details clearly mentioned. Ask the journal whether there is a way to offer more information in the supplemental files or external repositories if your target journal has strict word limitations. For example, Nature communications encourages authors to deposit their step-by-step protocols in an open-resource depository, Protocol Exchange which allows the protocols to be linked with the manuscript upon publication. Providing access to detailed protocols also helps to increase the transparency and reproducibility of the research.  
• It’s all in the details: The methods section should meticulously list all the materials, tools, instruments, and software used for different experiments. Specify the testing equipment on which data was obtained, together with its manufacturer’s information, location, city, and state or any other stimuli used to manipulate the variables. Provide specifics on the research process you employed; if it was a standard protocol, cite previous studies that also used the protocol.  Include any protocol modifications that were made, as well as any other factors that were taken into account when planning the study or gathering data. Any new or modified techniques should be explained by the authors. Typically, readers evaluate the reliability and validity of the procedures using the cited literature, and a widely accepted checklist helps to support the credibility of the methodology. Note: Authors should include a statement on sample size estimation (if applicable), which is often missed. It enables the reader to determine how many subjects will be required to detect the expected change in the outcome variables within a given confidence interval.  
• Write for the audience: While explaining the details in the methods section , authors should be mindful of their target audience, as some of the rationale or assumptions on which specific procedures are based might not always be obvious to the audience, particularly for a general audience. Therefore, when in doubt, the objective of a procedure should be specified either in relation to the research question or to the entire protocol.  
• Data interpretation and analysis : Information on data processing, statistical testing, levels of significance, and analysis tools and software should be added. Mention if the recommendations and expertise of an experienced statistician were followed. Also, evaluate and justify the preferred statistical method used in the study and its significance.  

What NOT to include in the methods section of a research paper  

To address “ how to write the methods section of a research paper ”, authors should not only pay careful attention to what to include but also what not to include in the methods section of a research paper . Here is a list of do not’s when writing the methods section :  

• Do not elaborate on specifics of standard methods/procedures: You should refrain from adding unnecessary details of experiments and practices that are well established and cited previously.  Instead, simply cite relevant literature or mention if the manufacturer’s protocol was followed.  
• Do not add unnecessary details : Do not include minute details of the experimental procedure and materials/instruments used that are not significant for the outcome of the experiment. For example, there is no need to mention the brand name of the water bath used for incubation.    
• Do not discuss the results: The methods section is not to discuss the results or refer to the tables and figures; save it for the results and discussion section. Also, focus on the methods selected to conduct the study and avoid diverting to other methods or commenting on their pros or cons.  
• Do not make the section bulky : For extensive methods and protocols, provide the essential details and share the rest of the information in the supplemental files. The writing should be clear yet concise to maintain the flow of the section.  

We hope that by this point, you understand how crucial it is to write a thoughtful and precise methods section and the ins and outs of how to write the methods section of a research paper . To restate, the entire purpose of the methods section is to enable others to reproduce the results or verify the research. We sincerely hope that this post has cleared up any confusion and given you a fresh perspective on the methods section .

As a parting gift, we’re leaving you with a handy checklist that will help you understand how to write the methods section of a research paper . Feel free to download this checklist and use or share this with those who you think may benefit from it.  

References  

• Bhattacharya, D. How to write the Methods section of a research paper. Editage Insights, 2018. https://www.editage.com/insights/how-to-write-the-methods-section-of-a-research-paper (2018).
• Kallet, R. H. How to Write the Methods Section of a Research Paper. Respiratory Care 49, 1229–1232 (2004). https://pubmed.ncbi.nlm.nih.gov/15447808/
• Grindstaff, T. L. & Saliba, S. A. AVOIDING MANUSCRIPT MISTAKES. Int J Sports Phys Ther 7, 518–524 (2012). https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3474299/

Editage All Access is a subscription-based platform that unifies the best AI tools and services designed to speed up, simplify, and streamline every step of a researcher’s journey. The Editage All Access Pack is a one-of-a-kind subscription that unlocks full access to an AI writing assistant, literature recommender, journal finder, scientific illustration tool, and exclusive discounts on professional publication services from Editage.  

Based on 22+ years of experience in academia, Editage All Access empowers researchers to put their best research forward and move closer to success. Explore our top AI Tools pack, AI Tools + Publication Services pack, or Build Your Own Plan. Find everything a researcher needs to succeed, all in one place –  Get All Access now starting at just $14 a month !    

Related Posts

Back to School – Lock-in All Access Pack for a Year at the Best Price

Journal Turnaround Time: Researcher.Life and Scholarly Intelligence Join Hands to Empower Researchers with Publication Time Insights 

• Research Guides

BSCI 1510L Literature and Stats Guide: 3.2 Components of a scientific paper

• 1 What is a scientific paper?
• 2 Referencing and accessing papers
• 2.1 Literature Cited
• 2.2 Accessing Scientific Papers
• 2.3 Traversing the web of citations
• 2.4 Keyword Searches
• 3 Style of scientific writing
• 3.1 Specific details regarding scientific writing

3.2 Components of a scientific paper

• 4 Summary of the Writing Guide and Further Information
• Appendix A: Calculation Final Concentrations
• 1 Formulas in Excel
• 2 Basic operations in Excel
• 3 Measurement and Variation
• 3.1 Describing Quantities and Their Variation
• 3.2 Samples Versus Populations
• 3.3 Calculating Descriptive Statistics using Excel
• 4 Variation and differences
• 5 Differences in Experimental Science
• 5.1 Aside: Commuting to Nashville
• 5.2 P and Detecting Differences in Variable Quantities
• 5.3 Statistical significance
• 5.4 A test for differences of sample means: 95% Confidence Intervals
• 5.5 Error bars in figures
• 5.6 Discussing statistics in your scientific writing
• 6 Scatter plot, trendline, and linear regression
• 7 The t-test of Means
• 8 Paired t-test
• 9 Two-Tailed and One-Tailed Tests
• 10 Variation on t-tests: ANOVA
• 11 Reporting the Results of a Statistical Test
• 12 Summary of statistical tests
• 1 Objectives
• 2 Project timeline
• 3 Background
• 4 Previous work in the BSCI 111 class
• 5 General notes about the project
• 6 About the paper
• 7 References

Nearly all journal articles are divided into the following major sections: abstract, introduction, methods, results, discussion, and references or literature cited.   Usually the sections are labeled as such, although often the introduction (and sometimes the abstract) is not labeled.  Sometimes alternative section titles are used.  The abstract is sometimes called the "summary", the methods are sometimes called "materials and methods", and the discussion is sometimes called "conclusions".   Some journals also include the minor sections of "key words" following the abstract, and "acknowledgments" following the discussion.  In some journals, the sections may be divided into subsections that are given descriptive titles.  However, the general division into the six major sections is nearly universal.

3.2.1 Abstract

The abstract is a short summary (150-200 words or less) of the important points of the paper.  It does not generally include background information.  There may be a very brief statement of the rationale for conducting the study.  It describes what was done, but without details.  It also describes the results in a summarized way that usually includes whether or not the statistical tests were significant.  It usually concludes with a brief statement of the importance of the results.  Abstracts do not include references.  When writing a paper, the abstract is always the last part to be written.

The purpose of the abstract is to allow potential readers of a paper to find out the important points of the paper without having to actually read the paper.  It should be a self-contained unit capable of being understood without the benefit of the text of the article . It essentially serves as an "advertisement" for the paper that readers use to determine whether or not they actually want to wade through the entire paper or not.  Abstracts are generally freely available in electronic form and are often presented in the results of an electronic search.  If searchers do not have electronic access to the journal in which the article is published, the abstract is the only means that they have to decide whether to go through the effort (going to the library to look up the paper journal, requesting a reprint from the author, buying a copy of the article from a service, requesting the article by Interlibrary Loan) of acquiring the article.  Therefore it is important that the abstract accurately and succinctly presents the most important information in the article.

3.2.2 Introduction

The introduction section of a paper provides the background information necessary to understand why the described experiment was conducted.  The introduction should describe previous research on the topic that has led to the unanswered questions being addressed by the experiment and should cite important previous papers that form the background for the experiment.  The introduction should also state in an organized fashion the goals of the research, i.e. the particular, specific questions that will be tested in the experiments.  There should be a one-to-one correspondence between questions raised in the introduction and points discussed in the conclusion section of the paper.  In other words, do not raise questions in the introduction unless you are going to have some kind of answer to the question that you intend to discuss at the end of the paper. 

You may have been told that every paper must have a hypothesis that can be clearly stated.  That is often true, but not always.  If your experiment involves a manipulation which tests a specific hypothesis, then you should clearly state that hypothesis.  On the other hand, if your experiment was primarily exploratory, descriptive, or measurative, then you probably did not have an  a priori  hypothesis, so don't pretend that you did and make one up.  (See the discussion in the introduction to Experiment 5 for more on this.)  If you state a hypothesis in the introduction, it should be a general hypothesis and not a null or alternative hypothesis for a statistical test.  If it is necessary to explain how a statistical test will help you evaluate your general hypothesis, explain that in the methods section. 

A good introduction should be fairly heavy with citations.  This indicates to the reader that the authors are informed about previous work on the topic and are not working in a vacuum.  Citations also provide jumping-off points to allow the reader to explore other tangents to the subject that are not directly addressed in the paper.  If the paper supports or refutes previous work, readers can look up the citations and make a comparison for themselves. 

"Do not get lost in reviewing background information. Remember that the Introduction is meant to introduce the reader to your research, not summarize and evaluate all past literature on the subject (which is the purpose of a review paper). Many of the other studies you may be tempted to discuss in your Introduction are better saved for the Discussion, where they become a powerful tool for comparing and interpreting your results. Include only enough background information to allow your reader to understand why you are asking the questions you are and why your hypotheses are reasonable ones. Often, a brief explanation of the theory involved is sufficient.

Write this section in the past or present tense, never in the future. " (Steingraber et al. 1985)

In other words, the introduction section relates what the topic being investigated is, why it is important, what research (if any) has been done prior that is relevant to what you are trying to do, and in what ways you will be looking into this topic.

An example to think about:

This is an example of a student-derived introduction.  Read the paragraph and before you go beyond, think about the paragraph first.

"Hand-washing is one of the most effective and simplest of ways to reduce infection and disease, and thereby causing less death.  When examining the effects of soap on hands, it was the work of Sickbert-Bennett and colleagues (2005) that showed that using soap or an alcohol on the hands during hand-washing was a significant effect in removing bacteria from the human hand.  Based on the work of this, the team led by Larsen (1991) then showed that the use of computer imaging could be a more effective way to compare the amount of bacteria on a hand."

There are several aspects within this "introduction" that could use improvement.  A group of any random 4 of you could easily come up with at 10 different things to reword, revise, expand upon.

In specific, there should be one thing addressed that more than likely you did not catch when you were reading it.

The citations: Not the format, but the logical use of them.

Look again. "...the work of Sickbert-Bennett...(2005)" and then "Based on the work of this, the team led by Larsen (1991)..."

How can someone in 1991 use or base their work on something from 2005?

They cannot.  You can spend an entire hour using spellcheck and reading through and through again to find all the little things to "give it more oomph", but at the core, you still must present a clear and concise and logical thought-process.

3.2.3 Methods (taken mostly verbatim from Steingraber et al. 1985, until the version A, B,C portion)

The function of the methods section is to describe all experimental procedures, including controls.  The description should be complete enough to enable someone else to repeat your work.  If there is more than one part to the experiment, it is a good idea to describe your methods and present your results in the same order in each section. This may not be the same order in which the experiments were performed -it is up to you to decide what order of presentation will make the most sense to your reader.

1.  Explain why each procedure was done, i.e., what variable were you measuring and why? Example:

Difficult to understand :  First, I removed the frog muscle and then I poured Ringer’s solution on it. Next, I attached it to the kymograph.

Improved:   I removed the frog muscle and poured Ringer’s solution on it to prevent it from drying out. I then attached the muscle to the kymograph in order to determine the minimum voltage required for contraction.

Better:   Frog muscle was excised between attachment points to the bone. Ringer's solution was added to the excised section to prevent drying out. The muscle was attached to the kymograph in order to determine the minimum voltage required for contraction.

2.  Experimental procedures and results are narrated in the past tense (what you did, what you found, etc.) whereas conclusions from your results are given in the present tense.

3.  Mathematical equations and statistical tests are considered mathematical methods and should be described in this section along with the actual experimental work. (Show a sample calculation, state the type of test(s) performed and program used)

4.  Use active rather than passive voice when possible.  [Note: see Section 3.1.4 for more about this.]  Always use the singular "I" rather than the plural "we" when you are the only author of the paper (Methods section only).  Throughout the paper, avoid contractions, e.g. did not vs. didn’t.

5.  If any of your methods is fully described in a previous publication (yours or someone else’s), you can cite work that instead of describing the procedure again.

Example:  The chromosomes were counted at meiosis in the anthers with the standard acetocarmine technique of Snow (1955).

Below is a PARTIAL and incomplete version of a "method".  Without getting into the details of why, Version A and B are bad.  A is missing too many details and B is giving some extra details but not giving some important ones, such as the volumes used.  Version C is still not complete, but it is at least a viable method. Notice that C is also not the longest....it is possible to be detailed without being long-winded.

In other words, the methods section is what you did in the experiment and has enough details that someone else can repeat your experiment.  If the methods section has excluded one or more important detail(s) such that the reader of the method does not know what happened, it is a 'poor' methods section.  Similarly, by giving out too many useless details a methods section can be 'poor'.

You may have multiple sub-sections within your methods (i.e., a section for media preparation, a section for where the chemicals came from, a section for the basic physical process that occurred, etc.,).  A methods section is  NEVER  a list of numbered steps.

3.2.4 Results (with excerpts from Steingraber et al. 1985)

The function of this section is to summarize general trends in the data without comment, bias, or interpretation. The results of statistical tests applied to your data are reported in this section although conclusions about your original hypotheses are saved for the Discussion section. In other words, you state "the P-value" in Results and whether below/above 0.05 and thus significant/not significant while in the Discussion you restate the P-value and then formally state what that means beyond "significant/not significant".

Tables and figures  should be used  when they are a more efficient way to convey information than verbal description. They must be independent units, accompanied by explanatory captions that allow them to be understood by someone who has not read the text. Do not repeat in the text the information in tables and figures, but do cite them, with a summary statement when that is appropriate.  Example:

Incorrect:   The results are given in Figure 1.

Correct:   Temperature was directly proportional to metabolic rate (Fig. 1).

Please note that the entire word "Figure" is almost never written in an article.  It is nearly always abbreviated as "Fig." and capitalized.  Tables are cited in the same way, although Table is not abbreviated.

Whenever possible, use a figure instead of a table. Relationships between numbers are more readily grasped when they are presented graphically rather than as columns in a table.

Data may be presented in figures and tables, but this may not substitute for a verbal summary of the findings. The text should be  understandable  by someone who has not seen your figures and tables.

1.  All results should be presented, including those that do not support the hypothesis.

2.  Statements made in the text must be supported by the results contained in figures and tables.

3.  The results of statistical tests can be presented in parentheses following a verbal description.

Example: Fruit size was significantly greater in trees growing alone (t = 3.65, df = 2, p < 0.05).

Simple results of statistical tests may be reported in the text as shown in the preceding example.  The results of multiple tests may be reported in a table if that increases clarity. (See Section 11 of the Statistics Manual for more details about reporting the results of statistical tests.)  It is not necessary to provide a citation for a simple t-test of means, paired t-test, or linear regression.  If you use other more complex (or less well-known) tests, you should cite the text or reference you followed to do the test.  In your materials and methods section, you should report how you did the test (e.g. using the statistical analysis package of Excel). 

It is NEVER appropriate to simply paste the results from statistical software into the results section of your paper.   The output generally reports more information than is required and it is not in an appropriate format for a paper. Similar, do NOT place a screenshot.  

Should you include every data point or not in the paper?  Prior to 2010 or so, most papers would probably not present the actual raw data collected, but rather show the "descriptive statistics" about their data (mean, SD, SE, CI, etc.). Often, people could simply contact the author(s) for the data and go from there.  As many journals have a significant on-line footprint now, it has become increasingly more common that the entire data could be included in the paper.  And realize why the entire raw data may not have been included in a publication. Prior to about 2010, your publication had limited  paper space  to be seen on.  If you have a sample of size of 10 or 50, you probably could show the entire data set easily in one table/figure and it not take up too much printed space. If your sample size was 500 or 5,000 or more, the size of the data alone would take pages of printed text.  Given how much the Internet and on-line publications have improved/increased in storage space, often now there will be either an embedded file to access or the author(s) will place the file on-line somewhere with an address link, such as GitHub.  Videos of the experiment are also shown as well now.

3.2.4.1 Tables

• Do not repeat information in a table that you are depicting in a graph or histogram; include a table only if it presents new information.
• It is easier to compare numbers by reading down a column rather than across a row. Therefore, list sets of data you want your reader to compare in vertical form.
• Provide each table with a number (Table 1, Table 2, etc.) and a title. The numbered title is placed above the table .
• Please see Section 11 of the Excel Reference and Statistics Manual for further information on reporting the results of statistical tests.

3.2.4.2. Figures

• These comprise graphs, histograms, and illustrations, both drawings and photographs. Provide each figure with a number (Fig. 1, Fig. 2, etc.) and a caption (or "legend") that explains what the figure shows. The numbered caption is placed below the figure .  Figure legend = Figure caption.
• Figures submitted for publication must be "photo ready," i.e., they will appear just as you submit them, or photographically reduced. Therefore, when you graduate from student papers to publishable manuscripts, you must learn to prepare figures that will not embarrass you. At the present time, virtually all journals require manuscripts to be submitted electronically and it is generally assumed that all graphs and maps will be created using software rather than being created by hand.  Nearly all journals have specific guidelines for the file types, resolution, and physical widths required for figures.  Only in a few cases (e.g. sketched diagrams) would figures still be created by hand using ink and those figures would be scanned and labeled using graphics software.  Proportions must be the same as those of the page in the journal to which the paper will be submitted. 
• Graphs and Histograms: Both can be used to compare two variables. However, graphs show continuous change, whereas histograms show discrete variables only.  You can compare groups of data by plotting two or even three lines on one graph, but avoid cluttered graphs that are hard to read, and do not plot unrelated trends on the same graph. For both graphs, and histograms, plot the independent variable on the horizontal (x) axis and the dependent variable on the vertical (y) axis. Label both axes, including units of measurement except in the few cases where variables are unitless, such as absorbance.
• Drawings and Photographs: These are used to illustrate organisms, experimental apparatus, models of structures, cellular and subcellular structure, and results of procedures like electrophoresis. Preparing such figures well is a lot of work and can be very expensive, so each figure must add enough to justify its preparation and publication, but good figures can greatly enhance a professional article, as your reading in biological journals has already shown.

3.2.5 Discussion (modified; taken from Steingraber et al. 1985)

The function of this section is to analyze the data and relate them to other studies. To "analyze" means to evaluate the meaning of your results in terms of the original question or hypothesis and point out their biological significance.

1. The Discussion should contain at least:

• the relationship between the results and the original hypothesis, i.e., whether they support the hypothesis, or cause it to be rejected or modified
• an integration of your results with those of previous studies in order to arrive at explanations for the observed phenomena
• possible explanations for unexpected results and observations, phrased as hypotheses that can be tested by realistic experimental procedures, which you should describe

2. Trends that are not statistically significant can still be discussed if they are suggestive or interesting, but cannot be made the basis for conclusions as if they were significant.

3. Avoid redundancy between the Results and the Discussion section. Do not repeat detailed descriptions of the data and results in the Discussion. In some journals, Results and Discussions are joined in a single section, in order to permit a single integrated treatment with minimal repetition. This is more appropriate for short, simple articles than for longer, more complicated ones.

4.  End the Discussion with a summary of the principal points you want the reader to remember. This is also the appropriate place to propose specific further study if that will serve some purpose,  but do not end with the tired cliché  that "this problem needs more study." All problems in biology need more study. Do not close on what you wish you had done, rather finish stating your conclusions and contributions.

5.  Conclusion section.  Primarily dependent upon the complexity and depth of an experiment, there may be a formal conclusion section after the discussion section. In general, the last line or so of the discussion section should be a more or less summary statement of the overall finding of the experiment.  IF the experiment was large enough/complex enough/multiple findings uncovered, a distinct paragraph (or two) may be needed to help clarify the findings.  Again, only if the experiment scale/findings warrant a separate conclusion section.

3.2.6 Title

The title of the paper should be the last thing that you write.  That is because it should distill the essence of the paper even more than the abstract (the next to last thing that you write). 

The title should contain three elements:

1. the name of the organism studied;

2. the particular aspect or system studied;

3. the variable(s) manipulated.

Do not be afraid to be grammatically creative. Here are some variations on a theme, all suitable as titles:

THE EFFECT OF TEMPERATURE ON GERMINATION OF ZEA MAYS

DOES TEMPERATURE AFFECT GERMINATION OF ZEA MAYS?

TEMPERATURE AND ZEA MAYS GERMINATION: IMPLICATIONS FOR AGRICULTURE

Sometimes it is possible to include the principal result or conclusion in the title:

HIGH TEMPERATURES REDUCE GERMINATION OF ZEA MAYS

Note for the BSCI 1510L class: to make your paper look more like a real paper, you can list all of the other group members as co-authors.  However, if you do that, you should list you name first so that we know that you wrote it.

Structure of a Research Paper

Structure of a Research Paper: IMRaD Format

I. The Title Page

• Title: Tells the reader what to expect in the paper.
• Author(s): Most papers are written by one or two primary authors. The remaining authors have reviewed the work and/or aided in study design or data analysis (International Committee of Medical Editors, 1997). Check the Instructions to Authors for the target journal for specifics about authorship.
• Keywords [according to the journal]
• Corresponding Author: Full name and affiliation for the primary contact author for persons who have questions about the research.
• Financial & Equipment Support [if needed]: Specific information about organizations, agencies, or companies that supported the research.
• Conflicts of Interest [if needed]: List and explain any conflicts of interest.

II. Abstract: “Structured abstract” has become the standard for research papers (introduction, objective, methods, results and conclusions), while reviews, case reports and other articles have non-structured abstracts. The abstract should be a summary/synopsis of the paper.

III. Introduction: The “why did you do the study”; setting the scene or laying the foundation or background for the paper.

IV. Methods: The “how did you do the study.” Describe the --

• Context and setting of the study
• Specify the study design
• Population (patients, etc. if applicable)
• Sampling strategy
• Intervention (if applicable)
• Identify the main study variables
• Data collection instruments and procedures
• Outline analysis methods

V. Results: The “what did you find” --

• Report on data collection and/or recruitment
• Participants (demographic, clinical condition, etc.)
• Present key findings with respect to the central research question
• Secondary findings (secondary outcomes, subgroup analyses, etc.)

VI. Discussion: Place for interpreting the results

• Main findings of the study
• Discuss the main results with reference to previous research
• Policy and practice implications of the results
• Strengths and limitations of the study

VII. Conclusions: [occasionally optional or not required]. Do not reiterate the data or discussion. Can state hunches, inferences or speculations. Offer perspectives for future work.

VIII. Acknowledgements: Names people who contributed to the work, but did not contribute sufficiently to earn authorship. You must have permission from any individuals mentioned in the acknowledgements sections. 

IX. References:  Complete citations for any articles or other materials referenced in the text of the article.

• IMRD Cheatsheet (Carnegie Mellon) pdf.
• Adewasi, D. (2021 June 14).  What Is IMRaD? IMRaD Format in Simple Terms! . Scientific-editing.info. 
• Nair, P.K.R., Nair, V.D. (2014). Organization of a Research Paper: The IMRAD Format. In: Scientific Writing and Communication in Agriculture and Natural Resources. Springer, Cham. https://doi.org/10.1007/978-3-319-03101-9_2
• Sollaci, L. B., & Pereira, M. G. (2004). The introduction, methods, results, and discussion (IMRAD) structure: a fifty-year survey.   Journal of the Medical Library Association : JMLA ,  92 (3), 364–367.
• Cuschieri, S., Grech, V., & Savona-Ventura, C. (2019). WASP (Write a Scientific Paper): Structuring a scientific paper.   Early human development ,  128 , 114–117. https://doi.org/10.1016/j.earlhumdev.2018.09.011

Module 1: Introduction: What is Research?

Learning Objectives

By the end of this module, you will be able to:

• Explain how the scientific method is used to develop new knowledge
• Describe why it is important to follow a research plan

The Scientific Method consists of observing the world around you and creating a  hypothesis  about relationships in the world. A hypothesis is an informed and educated prediction or explanation about something. Part of the research process involves testing the  hypothesis , and then examining the results of these tests as they relate to both the hypothesis and the world around you. When a researcher forms a hypothesis, this acts like a map through the research study. It tells the researcher which factors are important to study and how they might be related to each other or caused by a  manipulation  that the researcher introduces (e.g. a program, treatment or change in the environment). With this map, the researcher can interpret the information he/she collects and can make sound conclusions about the results.

Research can be done with human beings, animals, plants, other organisms and inorganic matter. When research is done with human beings and animals, it must follow specific rules about the treatment of humans and animals that have been created by the U.S. Federal Government. This ensures that humans and animals are treated with dignity and respect, and that the research causes minimal harm.

No matter what topic is being studied, the value of the research depends on how well it is designed and done. Therefore, one of the most important considerations in doing good research is to follow the design or plan that is developed by an experienced researcher who is called the  Principal Investigator  (PI). The PI is in charge of all aspects of the research and creates what is called a  protocol  (the research plan) that all people doing the research must follow. By doing so, the PI and the public can be sure that the results of the research are real and useful to other scientists.

Module 1: Discussion Questions

• How is a hypothesis like a road map?
• Who is ultimately responsible for the design and conduct of a research study?
• How does following the research protocol contribute to informing public health practices?

Email Updates

University Libraries

• Find Books & More
• Find Articles
• Find E-Journals
• Find Course Reserves
• E-Book Collections
• Get Stuff from Other Libraries
• Campus Delivery
• Recommend a Purchase
• Get Help from a Librarian
• Library Workshops
• Interlibrary Loan
• Individual Help
• Place an Item on Reserve
• Request an Instruction Session
• Actuarial Science
• Environmental Sciences
• General Sciences
• Mathematics
• Check My Account
• Renew My Books
• My Interlibrary Loan
• EndNote Basic
• Sciences Library Staff
• Location & Hours
• Calendar of Events
• Borrowing Policies
• Library Policies & Guidelines
• Subject Guides
• UI Sciences Library blog and RSS feed
• UI Sciences Library Twitter
• UI Sciences Library Facebook

Reading Science: Navigating Scientific Articles

The organization of a scientific article.

Primary research articles are typically organized into sections: introduction, materials and methods, results, and discussion (called IMRD).

Identify key elements

You may need to read an article several times in order to gain an understanding of it, but you can start by identifying key elements in a quick survey before you read.

Can you find?

• What was the purpose of the study? (in the introduction)
• Was the hypothesis supported? (in the discussion)
• What can you learn from the figures? Do you see trends? (in the results)
• How might the results be used in the future? What comes next? (in the discussion/conclusion)
• What were the limitations of the study? (in the discussion/conclusion)
• How was the experiment conducted? (in the materials and methods)
• How does this study build on previous research? (in the introduction)

Examples of key elements in a scientific paper

Files and links

• Scientific articles with Learning Lens annotations
• NPR: Her incredible sense of smell is helping scientists find new ways to diagnose disease
• Discovery of volatile biomarkers of Parkinson’s disease from sebum
• Worksheet Activity

Contact me.

• Last Updated: Sep 3, 2024 2:22 PM
• URL: https://guides.lib.uiowa.edu/ReadingScience

Have a language expert improve your writing

Run a free plagiarism check in 10 minutes, generate accurate citations for free.

• Knowledge Base
• Dissertation
• What Is a Research Methodology? | Steps & Tips

What Is a Research Methodology? | Steps & Tips

Published on August 25, 2022 by Shona McCombes and Tegan George. Revised on November 20, 2023.

Your research methodology discusses and explains the data collection and analysis methods you used in your research. A key part of your thesis, dissertation , or research paper , the methodology chapter explains what you did and how you did it, allowing readers to evaluate the reliability and validity of your research and your dissertation topic .

It should include:

• The type of research you conducted
• How you collected and analyzed your data
• Any tools or materials you used in the research
• How you mitigated or avoided research biases
• Why you chose these methods
• Your methodology section should generally be written in the past tense .
• Academic style guides in your field may provide detailed guidelines on what to include for different types of studies.
• Your citation style might provide guidelines for your methodology section (e.g., an APA Style methods section ).

Instantly correct all language mistakes in your text

Upload your document to correct all your mistakes in minutes

Table of contents

How to write a research methodology, why is a methods section important, step 1: explain your methodological approach, step 2: describe your data collection methods, step 3: describe your analysis method, step 4: evaluate and justify the methodological choices you made, tips for writing a strong methodology chapter, other interesting articles, frequently asked questions about methodology.

Prevent plagiarism. Run a free check.

Your methods section is your opportunity to share how you conducted your research and why you chose the methods you chose. It’s also the place to show that your research was rigorously conducted and can be replicated .

It gives your research legitimacy and situates it within your field, and also gives your readers a place to refer to if they have any questions or critiques in other sections.

You can start by introducing your overall approach to your research. You have two options here.

Option 1: Start with your “what”

What research problem or question did you investigate?

• Aim to describe the characteristics of something?
• Explore an under-researched topic?
• Establish a causal relationship?

And what type of data did you need to achieve this aim?

• Quantitative data , qualitative data , or a mix of both?
• Primary data collected yourself, or secondary data collected by someone else?
• Experimental data gathered by controlling and manipulating variables, or descriptive data gathered via observations?

Option 2: Start with your “why”

Depending on your discipline, you can also start with a discussion of the rationale and assumptions underpinning your methodology. In other words, why did you choose these methods for your study?

• Why is this the best way to answer your research question?
• Is this a standard methodology in your field, or does it require justification?
• Were there any ethical considerations involved in your choices?
• What are the criteria for validity and reliability in this type of research ? How did you prevent bias from affecting your data?

Once you have introduced your reader to your methodological approach, you should share full details about your data collection methods .

Quantitative methods

In order to be considered generalizable, you should describe quantitative research methods in enough detail for another researcher to replicate your study.

Here, explain how you operationalized your concepts and measured your variables. Discuss your sampling method or inclusion and exclusion criteria , as well as any tools, procedures, and materials you used to gather your data.

Surveys Describe where, when, and how the survey was conducted.

• How did you design the questionnaire?
• What form did your questions take (e.g., multiple choice, Likert scale )?
• Were your surveys conducted in-person or virtually?
• What sampling method did you use to select participants?
• What was your sample size and response rate?

Experiments Share full details of the tools, techniques, and procedures you used to conduct your experiment.

• How did you design the experiment ?
• How did you recruit participants?
• How did you manipulate and measure the variables ?
• What tools did you use?

Existing data Explain how you gathered and selected the material (such as datasets or archival data) that you used in your analysis.

• Where did you source the material?
• How was the data originally produced?
• What criteria did you use to select material (e.g., date range)?

The survey consisted of 5 multiple-choice questions and 10 questions measured on a 7-point Likert scale.

The goal was to collect survey responses from 350 customers visiting the fitness apparel company’s brick-and-mortar location in Boston on July 4–8, 2022, between 11:00 and 15:00.

Here, a customer was defined as a person who had purchased a product from the company on the day they took the survey. Participants were given 5 minutes to fill in the survey anonymously. In total, 408 customers responded, but not all surveys were fully completed. Due to this, 371 survey results were included in the analysis.

• Information bias
• Omitted variable bias
• Regression to the mean
• Survivorship bias
• Undercoverage bias
• Sampling bias

Qualitative methods

In qualitative research , methods are often more flexible and subjective. For this reason, it’s crucial to robustly explain the methodology choices you made.

Be sure to discuss the criteria you used to select your data, the context in which your research was conducted, and the role you played in collecting your data (e.g., were you an active participant, or a passive observer?)

Interviews or focus groups Describe where, when, and how the interviews were conducted.

• How did you find and select participants?
• How many participants took part?
• What form did the interviews take ( structured , semi-structured , or unstructured )?
• How long were the interviews?
• How were they recorded?

Participant observation Describe where, when, and how you conducted the observation or ethnography .

• What group or community did you observe? How long did you spend there?
• How did you gain access to this group? What role did you play in the community?
• How long did you spend conducting the research? Where was it located?
• How did you record your data (e.g., audiovisual recordings, note-taking)?

Existing data Explain how you selected case study materials for your analysis.

• What type of materials did you analyze?
• How did you select them?

In order to gain better insight into possibilities for future improvement of the fitness store’s product range, semi-structured interviews were conducted with 8 returning customers.

Here, a returning customer was defined as someone who usually bought products at least twice a week from the store.

Surveys were used to select participants. Interviews were conducted in a small office next to the cash register and lasted approximately 20 minutes each. Answers were recorded by note-taking, and seven interviews were also filmed with consent. One interviewee preferred not to be filmed.

• The Hawthorne effect
• Observer bias
• The placebo effect
• Response bias and Nonresponse bias
• The Pygmalion effect
• Recall bias
• Social desirability bias
• Self-selection bias

Mixed methods

Mixed methods research combines quantitative and qualitative approaches. If a standalone quantitative or qualitative study is insufficient to answer your research question, mixed methods may be a good fit for you.

Mixed methods are less common than standalone analyses, largely because they require a great deal of effort to pull off successfully. If you choose to pursue mixed methods, it’s especially important to robustly justify your methods.

Here's why students love Scribbr's proofreading services

Discover proofreading & editing

Next, you should indicate how you processed and analyzed your data. Avoid going into too much detail: you should not start introducing or discussing any of your results at this stage.

In quantitative research , your analysis will be based on numbers. In your methods section, you can include:

• How you prepared the data before analyzing it (e.g., checking for missing data , removing outliers , transforming variables)
• Which software you used (e.g., SPSS, Stata or R)
• Which statistical tests you used (e.g., two-tailed t test , simple linear regression )

In qualitative research, your analysis will be based on language, images, and observations (often involving some form of textual analysis ).

Specific methods might include:

• Content analysis : Categorizing and discussing the meaning of words, phrases and sentences
• Thematic analysis : Coding and closely examining the data to identify broad themes and patterns
• Discourse analysis : Studying communication and meaning in relation to their social context

Mixed methods combine the above two research methods, integrating both qualitative and quantitative approaches into one coherent analytical process.

Above all, your methodology section should clearly make the case for why you chose the methods you did. This is especially true if you did not take the most standard approach to your topic. In this case, discuss why other methods were not suitable for your objectives, and show how this approach contributes new knowledge or understanding.

In any case, it should be overwhelmingly clear to your reader that you set yourself up for success in terms of your methodology’s design. Show how your methods should lead to results that are valid and reliable, while leaving the analysis of the meaning, importance, and relevance of your results for your discussion section .

• Quantitative: Lab-based experiments cannot always accurately simulate real-life situations and behaviors, but they are effective for testing causal relationships between variables .
• Qualitative: Unstructured interviews usually produce results that cannot be generalized beyond the sample group , but they provide a more in-depth understanding of participants’ perceptions, motivations, and emotions.
• Mixed methods: Despite issues systematically comparing differing types of data, a solely quantitative study would not sufficiently incorporate the lived experience of each participant, while a solely qualitative study would be insufficiently generalizable.

Remember that your aim is not just to describe your methods, but to show how and why you applied them. Again, it’s critical to demonstrate that your research was rigorously conducted and can be replicated.

1. Focus on your objectives and research questions

The methodology section should clearly show why your methods suit your objectives and convince the reader that you chose the best possible approach to answering your problem statement and research questions .

2. Cite relevant sources

Your methodology can be strengthened by referencing existing research in your field. This can help you to:

• Show that you followed established practice for your type of research
• Discuss how you decided on your approach by evaluating existing research
• Present a novel methodological approach to address a gap in the literature

3. Write for your audience

Consider how much information you need to give, and avoid getting too lengthy. If you are using methods that are standard for your discipline, you probably don’t need to give a lot of background or justification.

Regardless, your methodology should be a clear, well-structured text that makes an argument for your approach, not just a list of technical details and procedures.

If you want to know more about statistics , methodology , or research bias , make sure to check out some of our other articles with explanations and examples.

• Normal distribution
• Measures of central tendency
• Chi square tests
• Confidence interval
• Quartiles & Quantiles

Methodology

• Cluster sampling
• Stratified sampling
• Thematic analysis
• Cohort study
• Peer review
• Ethnography

Research bias

• Implicit bias
• Cognitive bias
• Conformity bias
• Hawthorne effect
• Availability heuristic
• Attrition bias

Methodology refers to the overarching strategy and rationale of your research project . It involves studying the methods used in your field and the theories or principles behind them, in order to develop an approach that matches your objectives.

Methods are the specific tools and procedures you use to collect and analyze data (for example, experiments, surveys , and statistical tests ).

In shorter scientific papers, where the aim is to report the findings of a specific study, you might simply describe what you did in a methods section .

In a longer or more complex research project, such as a thesis or dissertation , you will probably include a methodology section , where you explain your approach to answering the research questions and cite relevant sources to support your choice of methods.

In a scientific paper, the methodology always comes after the introduction and before the results , discussion and conclusion . The same basic structure also applies to a thesis, dissertation , or research proposal .

Depending on the length and type of document, you might also include a literature review or theoretical framework before the methodology.

Quantitative research deals with numbers and statistics, while qualitative research deals with words and meanings.

Quantitative methods allow you to systematically measure variables and test hypotheses . Qualitative methods allow you to explore concepts and experiences in more detail.

Reliability and validity are both about how well a method measures something:

• Reliability refers to the  consistency of a measure (whether the results can be reproduced under the same conditions).
• Validity   refers to the  accuracy of a measure (whether the results really do represent what they are supposed to measure).

If you are doing experimental research, you also have to consider the internal and external validity of your experiment.

A sample is a subset of individuals from a larger population . Sampling means selecting the group that you will actually collect data from in your research. For example, if you are researching the opinions of students in your university, you could survey a sample of 100 students.

In statistics, sampling allows you to test a hypothesis about the characteristics of a population.

Cite this Scribbr article

If you want to cite this source, you can copy and paste the citation or click the “Cite this Scribbr article” button to automatically add the citation to our free Citation Generator.

McCombes, S. & George, T. (2023, November 20). What Is a Research Methodology? | Steps & Tips. Scribbr. Retrieved September 5, 2024, from https://www.scribbr.com/dissertation/methodology/

Is this article helpful?

Shona McCombes

Other students also liked, what is a theoretical framework | guide to organizing, what is a research design | types, guide & examples, qualitative vs. quantitative research | differences, examples & methods, what is your plagiarism score.

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Open access
• Published: 04 September 2024

A two-stage defect detection method for unevenly illuminated self-adhesive printed materials

• Guifeng Peng 1 ,
• Tao Song 1 ,
• Songxiao Cao 1 ,
• Bin Zhou 1 &
• Qing Jiang 1  

Scientific Reports volume  14 , Article number:  20547 ( 2024 ) Cite this article

Metrics details

• Applied optics
• Computer science
• Electrical and electronic engineering
• Information technology
• Mechanical engineering
• Optical techniques

The process of printing defect detection usually suffers from challenges such as inaccurate defect extraction and localization, caused by uneven illumination and complex textures. Moreover, image difference-based defect detection methods often result in numerous small-scale pseudo defects. To address these challenges, this paper proposes a comprehensive defect detection approach that integrates brightness correction and a two-stage defect detection strategy for self-adhesive printed materials. Concretely, a joint bilateral filter coupled with brightness correction corrects uneven brightness properly, meanwhile smoothing the grid-like texture in complex printed material images. Then, in the first detection stage, an image difference method based on a bright–dark difference template group is designed to effectively locate printing defects despite slight brightness fluctuations. Afterward, a discriminative method based on feature similarity is employed to filter out small-scale pseudo-defects in the second detection stage. The experimental results show that the improved difference method achieves an average precision of 99.1% in defect localization on five different printing pattern samples. Furthermore, the second stage reduces the false detection rate to under 0.5% while maintaining the low missed rate.

Similar content being viewed by others

AI-assisted reliability assessment for gravure offset printing system

Defect detection of printed circuit board assembly based on YOLOv5

Printing accuracy tracking with 2D optical microscopy and super-resolution metamaterial-assisted 1D terahertz spectroscopy

Introduction.

Self-adhesive label printed materials have been widely used in daily life, such as daily chemicals, medical, and electronics. The production process of these printed materials is susceptible to various factors that may result in printing defects, such as environmental interference, material variations, and mechanical faults. Real-time detection of defective products during the production process is crucial for guaranteeing the quality of printed materials. Given the inherent difficulties in observing printing defects, manual inspection methods have proven to be inefficient. Consequently, automated rapid optical quality inspection for printed materials holds immense promise. With the rapid development of industrial automation, machine vision and image processing technology have been widely used in surface defect detection 1 , 2 , 3 , 4 , 5 , 6 , including printing defects 7 .

Among traditional methods for printing defect detection, template matching is a commonly used approach. It directly assesses printing quality based on the similarity between the test image and the template image 8 , 9 , making it simple, intuitive, and easy to implement. Luo et al. 10 set up multi-templates with different rotation deviations to match the target image and assess the printing quality. Despite this approach decreased the false detection rate due to image rotate variance, the process of multiple template matching was time-consuming and affected the real-time performance. Ma et al. 11 distinguished the foreground and background of printed materials with simple text content and then detected defects in the foreground using template matching. Although this method had low-time–cost image matching, achieving precise separation of foreground and background for printed materials with complex graphics remained challenging. Liu et al. 12 created a template by fusing edge features of multiple images and identified defects using feature similarity matching based on Euclidean distance, yet this method was only applied to detect edge defects and not suitable for objects with complex printing graphics. Liu et al. 13 aligned the test image using adaptive template matching and utilized the Structural Similarity (SSIM) to measure the differences between the test image and the template image. Overall, template matching mainly focused on the global pixel correlations between images and had high requirements for lighting stability, making it difficult to detect small defects. These limitations render template matching insufficient for meeting actual industrial requirements.

Another popular approach is to perform difference operation between a test image and a reference image after image alignment. Li et al. 14 employed image difference method to extract cigarette label printing defects and applied the minimum external rectangle in analyzing defect shapes. However, the simple image difference method introduced pseudo defects, which reduced the detection accuracy. Guan et al. 15 generated the grayscale threshold image using the Gaussian distribution principle, which was then subjected to second difference processing with the difference image to eliminate the pseudo defects. Wang et al. 16 employed twice grayscale difference to segment non-edge defects and used gradient difference to segment edge defects. Despite this method eliminated the pseudo defects by fusing the binary images of the two types of defects, it was difficult to detect defects with low gray levels and small size. Li et al. 17 proposed the RGB subimage block sliding method to address the pseudo defects caused by local deformation of printed materials, then used the improved cosine similarity measure to perform twice gradient matching on the defect candidate regions. Although existing methods have achieved decent results in eliminating pseudo defects, they have shown particular sensitivity to lighting and limited effectiveness for objects with complex printing graphics. Zhang et al. 18 constructed a bright image and dark image as reference templates and performed defect detection through the difference operation between the test image and the two templates, which mitigated the issue of changes in brightness to a certain extent. However, its detection accuracy remained limited when facing printed materials with complex graphics.

In recent years, numerous studies have emerged focusing on detection through deep learning methods 6 , 19 , 20 . Unlike traditional methods, deep learning-based ones primarily utilize convolutional neural networks (CNN) to extract image defect features 21 , leveraging their unique advantages in mining defect information and demonstrating more robust recognition performance for surface defects 22 , 23 , 24 , 25 , 26 , 27 . Among them, You Only Look Once (YOLO) 28 , as a mainstream of object detection, is often considered for defect detection tasks. Liu et al. 29 achieved accurate detection of printing defects by inserting a Coordinate Attention mechanism into YOLOv5s. Li et al. 30 enhanced the model's capability to detect defects in printed paper by incorporating diverse attention mechanisms into YOLOv7. However, the success of supervised deep learning methods in achieving impressive detection results heavily relies on a large number of training samples and the balance between samples 31 . Li et al. 32 addressed the issue of imbalanced defect samples by augmenting the existing small amount of defect sample types, thereby improving the model’s classification ability for defects. For printed materials, printing defects include incomplete text, white specks, ink splatters, stains, misprints, and missed print, as shown in Fig.  1 . Due to the diversity and different shape characteristics of the same defect type, the need for a large annotated defect dataset is a significant challenge, as manually collecting and annotating data can be both expensive and time-consuming. Although there are some unsupervised zero-shot anomaly detection methods 33 , 34 , they still struggle to avoid the time-consuming nature of network training, which is unacceptable in practical self-adhesive printing production.

Representative printing defect samples. Standard images(top). Defect images(bottom).

During the image acquisition process of printed materials, irregular brightness fluctuations may occur locally due to factors such as mechanical vibration and inconsistent positioning and orientation of the printed materials. In addition, as shown in Fig.  2 , the complex graphic regions in images of printed materials often exhibit textures with diverse shapes and sizes, making the detection of printing defects more difficult. Consequently, mitigating the impact of brightness fluctuations and texture complexity constitutes a significant aspect. This paper aims to address the challenge of a high false detection rate in printing defect detection caused by uneven brightness, complex textures, and pseudo defects introduced by the image difference method. It solves this problem by delving into two core aspects: texture filtering with brightness correction, and two-stage defect discrimination.

Examples of image texture.

In summary, the main contributions of this paper are as follows:

To alleviate the interference caused by complex textures and uneven illumination, a joint bilateral filter that incorporates a nonlinear brightness correction factor is introduced to preprocess the images of self-adhesive printed materials.

An improved image difference method based on a dual difference template group is proposed for the rapid localization of printing defects, serving as the first stage of detection.

In the second stage of detection, a discriminative method utilizing multi-feature similarity is employed to filter out small-scale pseudo-defective areas to minimize the false detection rate.

The structure of this paper is as follows. Section “ Related work ” briefly reviews the relevant work, including brightness correction, image filtering, detection model based on difference templates, and image feature description algorithms. Section “ Methodology ” provides an in-depth analysis of the proposed method. To verify its effectiveness, Section “ Experiments and analysis ” provides the results of relevant experiments and visual analysis, and Section “ Conclusions ” concludes this article.

Related work

Image brightness correction algorithm.

At present, in the field of image brightness or color calibration, the main methods focus on enhancing or balancing the brightness of low-light or unevenly illuminated images 35 , 36 , 37 , 38 , 39 . However, these methods were insufficient to ensure the similar brightness or color distribution between the two images. Niu et al. 40 , 41 used the reference image to calibrate the color of the target image and align its color distribution. However, this calibration method tended to completely eliminate defect areas, making it unsuitable for surface defect detection.

Image filtering algorithm

In the field of image processing, a series of filters are adopted to smooth noise or texture. Early filters such as the mean filter 42 , median filter 43 , and Gaussian filter 44 were commonly used and demonstrated satisfactory smoothing effects on scattered noise. However, their performance was limited for textures presented in area blocks. Subsequently, the bilateral filter 45 , guided filter 46 , and joint bilateral filter 47 which extends from the bilateral filter were proposed.

Peak Signal-to-Noise Ratio (PSNR) is a widely used metric to quantify the error between the original image and the processed image based on pixel-level differences before and after image processing. It serves as a prevalent indicator to evaluate the effect of image texture smoothing, a high PSNR value indicates that the quality difference between the two images is small. As shown in Fig.  3 , the PSNR between the image after joint bilateral filtering and the texture-free image is the largest, which means that it is closer to the texture free state. Obviously, the joint bilateral filter exhibits a superior texture-smoothing effect.

The comparative effects of image filtering.

Difference template detection algorithm

The defect detection model based on difference template 18 computed the mean and standard deviation from multiple grayscale defect-free images, then constructed the bright and dark templates by combining the mean and deviation images with a predefined threshold. Pixels in the test image were deemed abnormal if their values exceeded those of the bright template or fell below those of the dark template.

The mean image and deviation image are calculated as follows:

where I i ( i  = 1, 2, …, n ) represents the defect-free samples, I r and I σ represent the mean image and the deviation image, respectively, and n denotes the number of samples.

The bright and dark templates are calculated as follows:

where a 1 , b 1 , and a 2 , b 2 are the absolute and relative thresholds for constructing bright and dark templates, respectively, and are four empirical values that need to be determined experimentally, usually a 1  =  a 2 and b 1  =  b 2 .

Image feature description algorithm

Image features are data or symbolic representations used to describe structural information in an image. Common image features include texture and shape descriptors such as Histogram of Oriented Gradient (HOG) 48 , 49 , 50 and Local Binary Pattern (LBP) 49 , 50 , pixel frequency distribution represented by Fast Fourier Transform (FFT) 51 , 52 , and geometric structure characterized by Hu moments 48 , 53 , 54 . Assessing the similarity of image features enables the evaluation of the matching between two objects 55 .

Methodology

The process of our proposed method is divided into three parts: image filtering with brightness correction, dual difference template group detection model as the first stage, and multi-feature fusion and comparison as the second stage, as shown in Fig.  4 .

Overall framework of the proposed method.

All input images are registered and cropped before image rectification and detection to ensure consistency in position, rotation angle, and size. In the brightness correction and filtering part, a nonlinear brightness correction function is introduced to properly adjust local brightness and eliminate complex textures, ensuring that the brightness distribution of the rectified test image is closer to the standard reference image. In the first stage, a dual difference template group is used to perform an image difference operation with the test image to locate defects. The relatively small pixel blocks obtained are regarded as defect candidates. In the second stage, subimages are extracted from the templates and the test image based on the regions of these small defect candidates. The HOG and FFT feature distances are calculated between the subimages, and a comparison operation is then performed after the distances are adaptively and dynamically weighted, accurately identifying the actual small defects.

Texture filtering with brightness correction

As shown in Fig.  5 , the spatial distance between neighborhood pixels and the central pixel determines corresponding spatial domain weights, the weights decrease as the distance increases. In the reference image used for guiding filtering, pixel domain weights are determined by evaluating the difference between the pixel value at the central position and those of the neighboring pixels, the weights decrease as the differences between pixel values increase. Similarly, the brightness correction weights are determined by the pixel value difference between the original image and the reference image at the same pixel position. The reference image is the mean image computed according to formula ( 1 ). To reduce the computational overhead, weight lists are generated. This allows for rapid access to the brightness correction weights and filtering weights through indexing. The weight lists are established as follows:

where List is the weight table, m ( m  = 0, 1, …, N )denotes the pixel value difference or the spatial distance length of the pixels in the x and y directions, and N is the maximum length of the List . The maximum length of the weight list in the pixel domain and correction domain is the maximum pixel value of the image, while the maximum length of the weight list in the spatial domain is the filter kernel size. σ serves as the weight adjustment parameter.

The principle of joint bilateral filtering with brightness correction. The List b , List r , and List s represent the weight lists of the brightness correction, pixel domain, and spatial domain, where the number codes denote the index in lists.

The formulas as follow depict the method for calculating filtering with a brightness correction factor:

where p and k represent the central pixel position and the neighborhood pixel position, respectively. I and J represent the pixel of the original image and the reference image, respectively. b k is the brightness correction weight, ε is a minimum value, which is used to prevent counting errors, and C k is the brightness correction factor. s k , x , and s k , y represent the components of spatial domain weights in the x and y directions, respectively. r k denoted the pixel domain weight, and I p ' is the pixel after rectification.

By incorporating the pixel values ratio between the original image and reference images, the original image pixels perform a nonlinear brightness transform when filtering. As previously mentioned, σ serves as the weight adjustment parameter, thus the adjustment parameters for brightness correction can be denoted as σ b , while those for filtering weight include σ r and σ s . By adjusting these parameters, the extent of brightness correction and smoothing of the image could be changed.

First stage detection with dual difference template group

Although the difference template model 18 has achieved satisfactory performance for printed materials with simple graphical content, using the same deviation image for both bright and dark templates will lead to over description of brightness variation.

As shown in Fig.  6 a, the bright and dark templates are constructed by statistically analyzing the maximum and minimum pixel values between the defect-free image samples in the R, G, and B channels, respectively. The computation method can be represented as follows:

where I b and I d denote the bright template and dark template, respectively.

Detailed illustration of the dual difference template group detection model. ( a ) The construction process of the template group. ( b ) The defects detection process, including abnormal pixels extraction, defect segmentation and localization.

The template represents the extreme value of brightness fluctuation in the defect-free samples, thus describing the change of illumination in the shooting environment to a certain extent. Refer to formula ( 2 ), the bright deviation template is calculated by the bright template and the mean image. Similarly, the dark deviation template is obtained by the dark template and the mean image. The template and deviation template together form a difference template group.

As shown in Fig.  6 b, based on the characteristics of a single-channel image, any defect could be classified as a bright or dark defect. After the subtraction operation is performed between the test image and the bright difference template group, pixels exceeding the bright threshold are classified as bright abnormal pixels. Similarly, pixels exceeding the dark threshold after subtracting the tested image from the dark difference template group are categorized as dark abnormal pixels. The abnormal pixel is distinguished by:

where I ( x , y ) represents the pixel value of the image to be detected. V b and V d are the bright and dark deviation templates. T b and T d denote the discrimination thresholds of bright and dark defects, respectively, usually T b  =  T d .

If the area of a connected abnormal pixel region, obtained from the binarized defect segmentation map that combines abnormal pixel mark maps from the R, G, and B channels, exceeds 100 pixels, the region is deemed a defect, while the small pixel blocks with 20–100 area obtained are regarded as defect candidates.

Second stage detection with local feature fusion

In the first stage of detection, there may be small false abnormal areas caused by the pixel value near the critical threshold, leading to false detection. So, we perform feature comparison on the defect candidate regions for further judgment.

Due to significant differences in contour and shape between defects and non-defective regions, HOG features excel in describing target contours and shapes, enabling effective discrimination between genuine defects and false ones. However, HOG features may misclassify complex image regions lacking distinct contours and gradient direction information as defect areas. On the other hand, since the similarity in pixel distribution structure among similar graphical regions, leveraging the advantages of FFT features in describing the overall image structure and pixel frequency information can partially compensate for the shortcomings of HOG features. Therefore, a second stage detection is performed on the candidate regions using HOG and FFT features.

The flowchart of the feature extraction and distance fusion is shown in Fig.  7 . Subimages are extracted from the templates and test image according to the candidate regions, then converted to grayscale. These grayscale subimages are divided into several cells, which are further grouped into blocks that could overlap each other. The gradient directions and amplitudes of all pixels within each cell are counted to generate histograms. In these histograms, the horizontal axis represents the gradient direction, while the vertical axis represents the cumulative amplitude for each gradient direction interval. Generally, the gradient direction ranges from 0° to 180°, and is divided into nine equal bins. The eigenvectors of each cell unit in the block are connected in a series to obtain the block descriptor, then the final HOG feature of the subimage can be obtained by cascading all feature vectors in the blocks. The FFT feature is computed for the entire subimage, with the feature vector obtained by flattening the transform result, including both magnitude and phase information.

The detail of the features extraction and distance fusion.

Since the acquired candidate regions vary in size, the dimensions of the divided cells and blocks for HOG feature extraction are not predetermined, but dynamically adjusted. The specific division process is shown in Fig.  8 . The default initial size of cells and blocks is assumed to be 2 × 2. To reduce the dispersion of the feature data, which affects the descriptive effectiveness, the cell number is limited to 30. When the data count does not meet the requirements, the cell size is incrementally enlarged along the width and height axis of the subimage. In addition, if the dimensions of cells fail to form a 2 × 2 block, the block size adjusts to 1 × 1.

The size setting process of cell and block.

In feature-based defect detection methods, image similarity is usually evaluated by calculating the feature distance between the test image and reference image to distinguish defects. However, the importance of information described by different features varies for different regions. Appropriate weights should be assigned to different types of features so that defects can be reliably detected based on the weighted distance.

The chi-square distance is used to calculate the HOG feature distance between the subimages of the defect candidate region in the test image, bright template, and dark template. Similarly, Euclidean distance is used to calculate FFT feature distance.

where H and F represent the HOG feature vectors and FFT feature vectors, respectively. D hog and D fft denote the feature distance between two subimages. I and i refer to the indices of the columns in the respective feature row vectors.

Feature distances are computed for three pairs of subimages: bright template and dark template, bright template and test image, and dark template and test image, resulting in a total of six distance data for the two feature categories.

Drawing from the preceding distance data, the coefficient weights for the HOG and FFT feature distances are deduced through the employment of the coefficient of variation method, with the calculation formulas presented as follows:

where A j and S j are the mean and the standard deviation of the distance data of each feature category, respectively. V j denotes the coefficient of variation of the data, and w j is the weight corresponding to the feature distance of each category.

Since the coefficient of variation method assigns greater weights to categories with larger internal data discrepancies, and our method prioritizes feature distance with smaller difference, the deduced weights are inversely proportional processed before appending to the corresponding distance.

where D fusion,bd , D fusion,bt , and D fusion,dt represent the weighted fused feature distance between the bright template and dark template, the bright template and test image, and the dark template and test image after nonlinear weight enhancement.

If both D fusion,bt and D fusion,dt are greater than D fusion,bd , the current subimage ( I t ) is deemed abnormal, indicating a defect in the corresponding region of the original image. The discrimination is as follows:

Experiments and analysis

Experimental preparation.

Dataset construction: As shown in Figure 9 , we use an array camera to sample images from self-adhesive printed materials of five different patterns. The dataset consists of 1772 RGB images with a resolution of 2448 × 2048 pixels, and sample counts from patterns 1 to 5 are 371, 474, 446, 249, and 232, with 45, 56, 54, 21, and 19 defect samples, respectively. For each pattern, 20 defect-free images are collected for constructing difference templates, and the remaining images are for testing.

Exemplary samples and image sampling platform.

Evaluation metrics: we use three standard quantitative indicators: precision (P), false detection rate (FDR), and missed rate (MR) to evaluate the defect detection results.

Brightness correction and filtering parameters setting

As shown in Fig. 10 , we conduct experiments and analysis separately on the settings of brightness correction and filtering parameters. Images with a size of 600*200 pixels divided into three intervals are used to verify the experimental effects. Each interval is assigned a fixed pixel value distribution range to simulate the brightness and noise distribution of the images.

The brightness correction and filtering effects under different parameters. ( a ) The pixel value distribution ranges of the three intervals for the original image and reference image are ([58,62][70,75][200,205]) and ([70,75][70,75][170,175]), respectively. ( b ) The pixel value distribution ranges of the three intervals for the images are ([55,65][75,85][120,170]). The horizontal axis in the graphs represents the pixel coordinates of the image, while the vertical axis represents the pixel value. The red profile line indicates the pixel value variation of the 100th row of pixels in the image.

The brightness correction effects of the image can be observed through the profile lines in Fig. 10 a. Increasing σ s does not affect the brightness correction effect while maintaining the σ b . Conversely, when σ s is unchanged, a larger σ b results in a more pronounced brightness correction effect. Moreover, the image to be corrected achieving the same brightness range as the reference image is only possible when σ b exceeds the brightness difference between the two images.

Fig. 10 b illustrates the image smoothing effects under different filtering parameters, the smoothness of the image profile lines, the slope of these lines at the boundaries of the three intervals, and the clarity of the image boundaries provide intuitive indications of the filtering effect. Observing the profile lines of the third interval reveals that only σ r exceeds the range of noise variation for achieving a better smoothing effect. In the image comparison in the third row, it is evident that larger σ r values result in greater blurring of the edges. Additionally, σ s contributes minimally to the smoothing effect, but larger values of this parameter actually lead to greater blurring of the image edges.

According to the analysis above and the brightness variation range as well as the texture pixel fluctuation range of the experimental samples, to ensure the preservation of edge details in images of printed materials while avoiding the elimination of defect regions during brightness correction, the optimal parameters are ultimately selected: σ b = 20, σ s = 3, and σ r = 20.

Brightness correction and filtering experiment

Figure 11 illustrates an example of the effects of texture smoothing and brightness correction respectively. The histograms in Fig. 11 a show the brightness correction results that the brightness distribution of the rectified image is close to the reference image. Meanwhile, Fig. 11 b–e enumerate pixel values of image profile lines in the second column, with Fig. 11 e demonstrating that our filtering with brightness correction method got the superior smooth effect. As shown in the third column figures, the texture in the smoothed image had been effectively eliminated.

The effects of brightness correction and filtering. ( a ) Gray histograms of the V-channel images in the HSV images corresponding to the original image, reference image, filtered image, and Rectified image. ( b )–( e ) The first column figures represent Locally enlarged subimages corresponding to the above images, respectively. The second column figures are the 250th row profile lines corresponding to the first column subimages. The third column figures are the local texture regions corresponding to the first column subimages.

First stage defect detection experiment

We employed both the detection model outlined in reference 18 and our improved version to conduct defects localization experiments on the samples of five printing patterns. The results of the experiments are delineated in Tables 1 and 2 , respectively, it becomes apparent that the reference method encounters challenges in effectively mitigating both the false detection rate (FDR) and missed rate (MR) simultaneously. Furthermore, the improved detection model necessitates the configuration of relatively fewer detection parameters.

Second stage defect detection experiment

Feature distance fusion experiment: Firstly, we independently validate the effect of HOG features and FFT features in representing images, with detailed comparisons illustrated in the first and second row graph of Fig. 12 . We respectively selected 12 typical normal and abnormal region subimages to illustrate the representational effect of the HOG and FFT features on these subimages. Referring to discrimination formula ( 16 ), a subimage is deemed abnormal if both the “Image-bright template” and the “Image-dark template” distances are greater than the “Bright–dark template distance” (both “orange triangle” and “green rectangle” are above “blue circle”), otherwise is deemed normal. It becomes evident that FFT features perform well in discriminating normal subimages by the distribution plots in Fig. 12 a. However, the HOG feature distance data incorrectly represents the normal subimage as an abnormal state. Conversely, the HOG features in Fig. 12 b exhibit strong representational capabilities for anomalous image subimages, yet the FFT feature distance data struggle to accurately identify all anomalous subimages.

Feature distance data distribution diagram of subimages. ( a ) Normal region subimages. ( b ) Abnormal region subimages.

Based on the above analysis, adaptive weight allocation is applied to the HOG feature distance and FFT feature distance according to formulas ( 10 )–( 12 ), subsequently calculating a comprehensive distance with superior representation capability. It is evident from the graph that the fusion feature distance adeptly distinguishes between normal and abnormal subimages.

Defect detection experiment: we employed a comparative approach of the above feature fusion distances to perform the second stage detection experiments on the five pattern samples, comparing the outcomes with those of the first stage detection, as delineated in Table  3 . The comparative results unveil that the pseudo-defective areas were effectively filtered out by employing image feature discrimination, resulting in a significant reduction in false detection rates. Fig. 13 demonstrates the final detection results of different defect types.

Results of printing defect detection. Binary mark maps of defect segmentation(top). The defect localization images(bottom).

Ablation study of brightness correction: To demonstrate the necessity of image brightness correction for defect detection, we compared the detection results under optimal detection thresholds without and with brightness correction on printed samples, as shown in Table 4 . Fig. 14 illustrates that the brightness unevenness of pattern 4 and 5 samples is relatively minor, leading to inconspicuous differences in FDR. However, even adjusting the threshold to achieve a lower FDR when without correction, a higher MR may still occur.

Analysis results of brightness unevenness in the samples of each pattern. The variance represents the degree of brightness variation.

Comparison experiment and time complexity

Table 5 illustrates the performance comparison with existing recent baselines. The proposed method yields the highest precision and the lowest FDR and MR, indicating significant detection superiority, especially for Liu et al. 13 , the average indicators improved by 34.3%, 34.5%, and 29.2%, respectively. Their method performed adaptive threshold based binarization before performing difference detection on the images, which was significantly affected by lighting, making it difficult to achieve satisfactory results. Li et al. 17 showed good detection precision for patterns 1 and 3 samples and a generally lower MR than Liu et al. 13 . However, their method was less compatible with other printing patterns, and the process of sliding subimage blocks was extremely time-consuming.

Defect detection experiment on additional datasets

To validate that our method is still effective in practical industrial datasets, we did further experiments using a text dataset 56 and the leather dataset from the MVTec 57 , with detailed detection results shown in Table 6 . It is evident that the proposed method results in a higher MR. The primary reason could be that the constructed detection templates have an overly broad tolerance range, which is a consequence of the considerable overall brightness variation because of the original samples in the text dataset having undergone artificial data augmentation. Additionally, the rough and irregular surface textures of the leather samples necessitate a larger smoothing parameter to eliminate texture interference, which significantly diminishes the prominence of many defect features. Nevertheless, our method still exhibits substantial detection effectiveness, with the detection effects illustrated in Fig. 15 .

Qualitative defect detection effects on additional datasets. Binary mark maps of defect segmentation(top). The defect localization images(bottom).

Conclusions

Few existing methods for printing defect detection pay attention to self-adhesive label printed materials with complex textures under uneven lighting conditions, and the common method of image difference often generates pseudo-defective regions. To solve these problems, we propose a two-stage printing defect detection method with brightness correction and texture filtering. The following conclusions in this work could be drawn:

A brightness correction factor is incorporated into the joint bilateral filter, which facilitates local brightness adjustments to a certain extent and texture smoothing in images of printed materials, initially reducing the interference of uneven brightness and complex textures in defect detection.

Compared with the traditional printing defect detection method, the proposed difference method based on bright–dark difference template groups accurately describes the actual brightness variations of the image through templates, enabling excellent extraction and localization of defects despite the interference of slight brightness fluctuations.

To overcome the challenge of pseudo defects generated by the image difference method, we conduct a second-stage detection on small defect candidate regions by leveraging the advantages of HOG in describing image contours and FFT features in describing image pixel frequency distributions, which further reduces the false detection rate.

The experimental results indicate that the proposed method achieves excellent performance, with false detection rate and missed rate both below 0.5%.

Data availability

The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.

Shankar, N. G., Ravi, N. & Zhong, Z. W. A real-time print-defect detection system for web offset printing. Measurement 42 , 645–652 (2009).

Article   ADS   Google Scholar  

Annaby, M. H., Fouda, Y. M. & Rushdi, M. A. Improved normalized cross-correlation for defect detection in printed-circuit boards. IEEE Trans. Semicond. Manuf. 32 , 199–211 (2019).

Article   Google Scholar  

Liu, A., Yang, E., Wu, J., Teng, Y. & Yu, L. Double sparse low rank decomposition for irregular printed fabric defect detection. Neurocomputing 482 , 287–297 (2022).

Jian, C., Gao, J. & Ao, Y. Automatic surface defect detection for mobile phone screen glass based on machine vision. Appl. Soft Comput. 52 , 348–358 (2017).

Staude, A. et al. Quantification of the capability of micro-CT to detect defects in castings using a new test piece and a voxel-based comparison method. NDT E Int. 44 , 531–536 (2011).

Article   CAS   Google Scholar  

Yixuan, L., Dongbo, W., Jiawei, L. & Hui, W. Aeroengine blade surface defect detection system based on improved faster RCNN. Int. J. Intell. Syst. 2023 , 1–14 (2023).

Zhang, F., Li, Y., Zeng, Q. & Lu, L. Application of printing defects detection based on visual saliency. J. Phys. Conf. Ser. 1920 , 012053 (2021).

Gong, W., Zhang, K., Yang, C., Yi, M. & Wu, J. Adaptive visual inspection method for transparent label defect detection of curved glass bottle. In 2020 International Conference on Computer Vision, Image and Deep Learning (CVIDL) 90–95 (IEEE, Chongqing, China, 2020). https://doi.org/10.1109/CVIDL51233.2020.00024 .

Wang, Y. Research on image matching in printing defects detection based on machine vision. In 2019 IEEE 19th International Conference on Communication Technology (ICCT) 1648–1652 (2019). https://doi.org/10.1109/ICCT46805.2019.8947055 .

Luo, B. & Guo, G. Fast printing defects inspection based on multi-matching. In 2016 12th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD) 1492–1496 (2016). https://doi.org/10.1109/FSKD.2016.7603397 .

Ma, B. et al. The defect detection of personalized print based on template matching. In 2017 IEEE International Conference on Unmanned Systems (ICUS) 266–271 (2017). https://doi.org/10.1109/ICUS.2017.8278352 .

Liu, B., Chen, Y., Xie, J. & Chen, B. Industrial printing image defect detection using multi-edge feature fusion algorithm. Complexity 2021 , 1–10 (2021).

Liu, X., Li, Y., Guo, Y. & Zhou, L. Printing defect detection based on scale-adaptive template matching and image alignment. Sensors 23 , 4414 (2023).

Article   ADS   PubMed   PubMed Central   Google Scholar  

Li, M. X., Liu, Q. X. & He, C. Y. Research and achievement on cigarette label printing defect detection algorithm. Appl. Mech. Mater. 200 , 689–693 (2012).

Guan, Y. Y. & Ye, Y. C. Printing defects detection based on two-times difference image method. Appl. Mech. Mater. 340 , 512–516 (2013).

Yangping, W., Shaowei, X., Zhengping, Z., Yue, S. & Zhenghai, Z. Real-time defect detection method for printed images based on grayscale and gradient differences. J. Eng. Sci. Technol. Rev. 11 , 180–188 (2018).

Li, D. et al. Printed label defect detection using twice gradient matching based on improved cosine similarity measure. Expert Syst. Appl. 204 , 117372 (2022).

Zhang, E., Chen, Y., Gao, M., Duan, J. & Jing, C. Automatic defect detection for web offset printing based on machine vision. Appl. Sci. 9 , 3598 (2019).

Prunella, M. et al. Deep learning for automatic vision-based recognition of industrial surface defects: A survey. IEEE Access 11 , 43370–43423 (2023).

Huang, L. & Gong, A. Surface defect detection for no-service rails with skeleton-aware accurate and fast network. IEEE Trans. Ind. Inform. 20 , 4571–4581 (2024).

Singh, S. A., Kumar, A. S. & Desai, K. A. Comparative assessment of common pre-trained CNNs for vision-based surface defect detection of machined components. Expert Syst. Appl. 218 , 119623 (2023).

Valente, A. C. et al. Print Defect Mapping with Semantic Segmentation. In 2020 IEEE Winter Conference on Applications of Computer Vision (WACV) 3540–3548 (IEEE, Snowmass Village, CO, USA, 2020). https://doi.org/10.1109/WACV45572.2020.9093470 .

Chen, Z. et al. Bi-deformation-UNet: Recombination of differential channels for printed surface defect detection. Vis. Comput. 39 , 3995–4013 (2023).

Li, D., Li, Y., Li, J. & Lu, G. Differential enhanced siamese segmentation network for printed label defect detection. In 2023 IEEE International Conference on Image Processing (ICIP) 530–534 (2023). https://doi.org/10.1109/ICIP49359.2023.10222759 .

Sun, Y., Wei, J., Li, J., Wei, Q. & Ye, W. An online color and shape integrated detection method for flexible packaging surface defects. Meas. Sci. Technol. 35 , 066207 (2024).

Qian, Y., Liu, X., Ma, X. & Kuang, H. DA-UNet: An improved UNet-based algorithm for segmenting defects in printed cards. In 2024 IEEE 6th Advanced Information Management, Communicates, Electronic and Automation Control Conference (IMCEC) vol. 6 1065–1069 (2024).

Chen, W. et al. Small target detection algorithm for printing defects detection based on context structure perception and multi-scale feature fusion. Signal Image Video Process. 18 , 657–667 (2024).

Hussain, M. YOLO-v1 to YOLO-v8, the rise of YOLO and its complementary nature toward digital manufacturing and industrial defect detection. Machines 11 , 677 (2023).

Liu, J. et al. An improved printing defect detection method based on YOLOv5s. In 2023 5th International Conference on Communications, Information System and Computer Engineering (CISCE) 253–257 (IEEE, Guangzhou, China, 2023). https://doi.org/10.1109/CISCE58541.2023.10142568 .

Li, X. et al. A novel hybrid YOLO approach for precise paper defect detection with a dual-layer template and an attention mechanism. IEEE Sens. J. 24 , 11651–11669 (2024).

Bai, D. et al. Surface defect detection methods for industrial products with imbalanced samples: A review of progress in the 2020s. Eng. Appl. Artif. Intell. 130 , 107697 (2024).

Li, J., Pan, J. & Zhang, Q. A Printing defect recognition method based on class-imbalanced learning. In 2023 IEEE 3rd International Conference on Power, Electronics and Computer Applications (ICPECA) 370–373 (IEEE, Shenyang, China, 2023). https://doi.org/10.1109/ICPECA56706.2023.10075748 .

Qiu, J., Shi, H., Hu, Y. & Yu, Z. Unraveling false positives in unsupervised defect detection models: A study on anomaly-free training datasets. Sensors 23 , 9360 (2023).

Yao, H. et al. Global-regularized neighborhood regression for efficient zero-shot texture anomaly detection. Preprint at https://doi.org/10.48550/arXiv.2406.07333 (2024).

Gao, Y., Xu, Y., Tian, S. & Chang, B. Transformation and analysis of pixels based on low-level-light image. In 2008 International Conference on Optical Instruments and Technology: Optoelectronic Measurement Technology and Applications vol. 7160 782–789 (SPIE, 2009).

Park, S. et al. Brightness and color correction for dual camera image registration. In 2016 IEEE International Conference on Consumer Electronics-Asia (ICCE-Asia) 1–2 (2016). https://doi.org/10.1109/ICCE-Asia.2016.7804811 .

Takahashi, K., Monno, Y., Tanaka, M. & Okutomi, M. Effective color correction pipeline for a noisy image. In 2016 IEEE International Conference on Image Processing (ICIP) 4002–4006 (IEEE, Phoenix, AZ, USA, 2016). https://doi.org/10.1109/ICIP.2016.7533111 .

Cao, G. et al. Contrast enhancement of brightness-distorted images by improved adaptive gamma correction. Comput. Electr. Eng. 66 , 569–582 (2018).

Su, Y., Wu, M. & Yan, Y. Image enhancement and brightness equalization algorithms in low illumination environment based on multiple frame sequences. IEEE Access 11 , 61535–61545 (2023).

Niu, Y., Zhang, H., Guo, W. & Ji, R. Image quality assessment for color correction based on color contrast similarity and color value difference. IEEE Trans. Circuits Syst. Video Technol. 28 , 849–862 (2018).

Niu, Y., Liu, P., Zhao, T. & Fan, Y. Matting-based residual optimization for structurally consistent image color correction. IEEE Trans. Circuits Syst. Video Technol. 30 , 3624–3636 (2020).

Pitas, I. & Venetsanopoulos, A. Nonlinear mean filters in image processing. IEEE Trans. Acoust. Speech Signal Process. 34 , 573–584 (1986).

Justusson, B. I. Median filtering: statistical properties. In Two-Dimensional Digital Signal Prcessing II: Transforms and Median Filters 161–196 (Springer, Berlin, Heidelberg, 1981). https://doi.org/10.1007/BFb0057597 .

Ito, K. & Xiong, K. Gaussian filters for nonlinear filtering problems. IEEE Trans. Autom. Control 45 , 910–927 (2000).

Article   MathSciNet   Google Scholar  

Gadde, A., Narang, S. K. & Ortega, A. Bilateral filter: Graph spectral interpretation and extensions. In 2013 IEEE International Conference on Image Processing 1222–1226 (IEEE, Melbourne, Australia, 2013). https://doi.org/10.1109/ICIP.2013.6738252 .

He, K., Sun, J. & Tang, X. Guided image filtering. IEEE Trans. Pattern Anal. Mach. Intell. 35 , 1397–1409 (2013).

Article   PubMed   Google Scholar  

Caraffa, L., Tarel, J.-P. & Charbonnier, P. The guided bilateral filter: when the joint/cross bilateral filter becomes robust. IEEE Trans. Image Process. 24 , 1199–1208 (2015).

Article   ADS   MathSciNet   PubMed   Google Scholar  

Jia, S., Zhao, X., Li, Y. & Wang, K. A particle filter human tracking method based on HOG and Hu moment. In 2014 IEEE International Conference on Mechatronics and Automation 1581–1586 (2014). https://doi.org/10.1109/ICMA.2014.6885936 .

Xiang, Z., Tan, H. & Ye, W. The excellent properties of a dense grid-based HOG feature on face recognition compared to gabor and LBP. IEEE Access 6 , 29306–29319 (2018).

Liu, H., Jia, X., Su, C., Yang, H. & Li, C. Tire appearance defect detection method via combining HOG and LBP features. Front. Phys. 10 , (2023).

Rangaswamy, Y., Raja, K. B. & Venugopal, K. R. FRDF: Face recognition using fusion of DTCWT and FFT features. Procedia Comput. Sci. 54 , 809–817 (2015).

Ishihara, S. et al. 2D FFT and AI-Based Analysis of Wallpaper Patterns and Relations Between Kansei. In Advances in Affective and Pleasurable Design (ed. Fukuda, S.) 329–338 (Springer International Publishing, Cham, 2020). https://doi.org/10.1007/978-3-030-20441-9_35 .

Zhang, L., Xiang, F., Pu, J. & Zhang, Z. Application of improved HU moments in object recognition. In 2012 IEEE International Conference on Automation and Logistics 554–558 (2012). https://doi.org/10.1109/ICAL.2012.6308139 .

Zhang, B., Zhang, Y., Liu, J. & Wang, B. FGFF descriptor and modified Hu moment-based hand gesture recognition. Sensors 21 , 6525 (2021).

Wu, X., Fu, K., Liu, Z. & Chen, W. A Brief survey of feature based image matching. In 2022 IEEE 17th Conference on Industrial Electronics and Applications (ICIEA) 1634–1639 (2022). https://doi.org/10.1109/ICIEA54703.2022.10006226 .

Malaysia, U. S. Label printing defect detection system dataset. Roboflow Universe (2023).

Bergmann, P., Fauser, M., Sattlegger, D. & Steger, C. MVTec AD—A comprehensive real-world dataset for unsupervised anomaly detection. In 9592–9600 (2019).

Download references

Acknowledgements

This work was supported by the State Administration for Market Regulation Science and Technology Foundation of China under Grant 2021MK90.

Author information

Authors and affiliations.

College of Metrology Measurement and Instrument, China Jiliang University, Hangzhou, 310018, China

Guifeng Peng, Tao Song, Songxiao Cao, Bin Zhou & Qing Jiang

You can also search for this author in PubMed   Google Scholar

Contributions

G.P. wrote the main manuscript text. T.S. was responsible for the overall direction and supervision of the paper. All authors reviewed the manuscript.

Corresponding author

Correspondence to Tao Song .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/ .

Reprints and permissions

About this article

Cite this article.

Peng, G., Song, T., Cao, S. et al. A two-stage defect detection method for unevenly illuminated self-adhesive printed materials. Sci Rep 14 , 20547 (2024). https://doi.org/10.1038/s41598-024-71514-z

Download citation

Received : 31 May 2024

Accepted : 28 August 2024

Published : 04 September 2024

DOI : https://doi.org/10.1038/s41598-024-71514-z

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Printing defect
• Image difference
• Brightness correction
• Defect detection

By submitting a comment you agree to abide by our Terms and Community Guidelines . If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing: AI and Robotics newsletter — what matters in AI and robotics research, free to your inbox weekly.

Writing the Five Principal Sections: Abstract, Introduction, Methods, Results and Discussion

• First Online: 12 October 2013

Cite this chapter

• Karen Englander 2  

Part of the book series: SpringerBriefs in Education ((BRIEFSEDUCAT))

4688 Accesses

The five-part structure of research papers (Introduction, Method, Results and Discussion plus Abstract) serves as the conceptual basis for the content. The structure can be considered an “hourglass” or a “conversation” with predictable elements. Each section of the paper has its own common internal structure and linguistic features. The structures and features are explained with examples from published papers in a range of scientific disciplines. Exceptions to the four-part structure are also discussed.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Subscribe and save.

• Get 10 units per month
• Download Article/Chapter or eBook
• 1 Unit = 1 Article or 1 Chapter
• Cancel anytime
• Available as PDF
• Read on any device
• Instant download
• Own it forever
• Available as EPUB and PDF
• Compact, lightweight edition
• Dispatched in 3 to 5 business days
• Free shipping worldwide - see info

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Cargill, M., & O’Connor, P. (2009). Writing scientific research articles . West Sussex: Wiley-Blackwell.

Google Scholar  

Cooper, E. A., Piazza, E. A., & Banks, M. S. (2012). The perceptual basis of common photographic practice. Journal of Vision, 12 (5), 1–14.

Article   Google Scholar  

Esquer Mendez, J. L., Hernández Rodríguez, M., & Bückle Ramirez, L. F. (2010). Thermal tolerance and compatibility zones as a tool to establish the optimum culture condition of the halibut Paralichthys californicus (Ayres, 1859). Aquaculture Research, 41 (7), 1015–1021.

Glassman-Deal, H. (2010). Science research writing: For non-native speakers of English . London: Imperial College Press.

Latour, B., & Woolgar, S. (1986). Laboratory life: The social construction of scientific facts . Princeton: Princeton University Press.

Penrose, A. M., & Katz, S. B. (2001). Writing in the sciences: Exploring conventions of scientific discourse . New York: Longman.

Pérez-Flores, M. A., Antonio-Carpio, R. G., Enrique Gómez-Treviño, E., & Ferguson, I. (2011). 3D inversion of control source LIN electromagnetic data. Unpublished manuscript.

Rodríguez-Santiago, M. A., & Rosales-Casián, J. A. (2011). Parasite structure of the Ocean Whitefish Caulolatilus princeps from Baja California, México (East Pacific). Helgoland Marine Research, 65 (2), 197–202.

Swales, J. (1990). Genre analysis: English in academic and research settings . Cambridge: Cambridge University Press.

Swales, J. M., & Feak, C. (2004). Academic writing for graduate students: Essential tasks and skills (2nd ed.). Ann Arbor: University of Michigan Press.

Download references

Author information

Authors and affiliations.

Department of Languages, Literatures and Linguistics, York University, Toronto, ON, Canada

Karen Englander

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Karen Englander .

Rights and permissions

Reprints and permissions

Copyright information

© 2014 The Author(s)

About this chapter

Englander, K. (2014). Writing the Five Principal Sections: Abstract, Introduction, Methods, Results and Discussion. In: Writing and Publishing Science Research Papers in English. SpringerBriefs in Education. Springer, Dordrecht. https://doi.org/10.1007/978-94-007-7714-9_8

Download citation

DOI : https://doi.org/10.1007/978-94-007-7714-9_8

Published : 12 October 2013

Publisher Name : Springer, Dordrecht

Print ISBN : 978-94-007-7713-2

Online ISBN : 978-94-007-7714-9

eBook Packages : Humanities, Social Sciences and Law Education (R0)

Share this chapter

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Publish with us

Policies and ethics

• Find a journal
• Track your research

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here .

Loading metrics

Open Access

Peer-reviewed

Research Article

In-shoe plantar shear stress sensor design, calibration and evaluation for the diabetic foot

Roles Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Software, Validation, Visualization, Writing – original draft, Writing – review & editing

Affiliation Department of Mechanical, Aerospace and Civil Engineering (MACE), University of Manchester, Manchester, United Kingdom

Roles Investigation, Methodology, Writing – review & editing

Roles Conceptualization, Data curation, Funding acquisition, Investigation, Resources, Supervision, Validation, Writing – review & editing

Affiliation Medical School, NIHR Exeter BRC, University of Exeter, Exeter, United Kingdom

Roles Investigation, Methodology, Software, Writing – review & editing

Roles Investigation, Software, Writing – review & editing

Roles Resources, Supervision, Writing – review & editing

Affiliation Musculoskeletal Biomechanics and Research in Science and Engineering faculty of Manchester Metropolitan University, Manchester, United Kingdom

Affiliation Manchester University NHS Foundation Trust within the Departments of Diabetes and Vascular Surgery, Manchester, United Kingdom

Roles Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing

Roles Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing

* E-mail: [email protected]

• Athia H. Haron, 
• Lutong Li, 
• Jiawei Shuang, 
• Chaofan Lin, 
• Helen Dawes, 
• Maedeh Mansoubi, 
• Damian Crosby, 
• Garry Massey, 
• Neil Reeves, 

• Published: September 4, 2024
• https://doi.org/10.1371/journal.pone.0309514
• Peer Review
• Reader Comments

Plantar shear stress may have an important role in the formation of a Diabetic Foot Ulcer, but its measurement is regarded as challenging and has limited research. This paper highlights the importance of anatomical specific shear sensor calibration and presents a feasibility study of a novel shear sensing system which has measured in-shoe shear stress from gait activity on both healthy and diabetic subjects. The sensing insole was based on a strain gauge array embedded in a silicone insole backed with a commercial normal pressure sensor. Sensor calibration factors were investigated using a custom mechanical test rig with indenter to exert both normal and shear forces. Indenter size and location were varied to investigate the importance of both loading area and position on measurement accuracy. The sensing insole, coupled with the calibration procedure, was tested one participant with diabetes and one healthy participant during two sessions of 15 minutes of treadmill walking. Calibration with different indenter areas (from 78.5 mm 2 to 707 mm 2 ) and different positions (up to 40 mm from sensor centre) showed variation in measurements of up to 80% and 90% respectively. Shear sensing results demonstrated high repeatability (>97%) and good accuracy (mean absolute error < ±18 kPa) in bench top mechanical tests and less than 21% variability within walking of 15-minutes duration. The results indicate the importance of mechanical coupling between embedded shear sensors and insole materials. It also highlights the importance of using an appropriate calibration method to ensure accurate shear stress measurement. The novel shear stress measurement system presented in this paper, demonstrates a viable method to measure accurate and repeatable in-shoe shear stress using the calibration procedure described. The validation and calibration methods outlined in this paper could be utilised as a standardised approach for the research community to develop and validate similar measurement technologies.

Citation: Haron AH, Li L, Shuang J, Lin C, Dawes H, Mansoubi M, et al. (2024) In-shoe plantar shear stress sensor design, calibration and evaluation for the diabetic foot. PLoS ONE 19(9): e0309514. https://doi.org/10.1371/journal.pone.0309514

Editor: Andrea Tigrini, Polytechnic University of Marche: Universita Politecnica delle Marche, ITALY

Received: January 16, 2024; Accepted: August 14, 2024; Published: September 4, 2024

Copyright: © 2024 Haron et al. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data are available from the Mendeley Data database (DOI 10.17632/pcggh2rzm3.1 ). Only raw anonymised data can be shared due to GDPR restrictions from HRC ethics committee.

Funding: This work was partially funded by Engineering and Physical Sciences Research Council (EPSRC) grant number EP/W00366X/1.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Diabetic foot ulceration (DFU) affects 15–25% of people with diabetes at some point in their lifetime [ 1 ] and has a high social and economic cost with countries like the UK spending approximately £1 billion annually [ 2 ]. Worldwide the prevalence of diabetes is rising, and it is predicted that 552 million people will have the condition by 2030 [ 3 ]. Measurement of plantar normal stress and plantar shear stress has shown the potential to predict DFU risk [ 4 , 5 ]. However, whilst commercial systems are available to measure normal plantar stress in-shoe there are no commercially available in-shoe plantar shear stress measurement systems. Shear stress has been directly measured during barefoot gait using mechanical sensor arrays coupled with resistive or capacitive sensors [ 6 – 8 ], utilising piezoelectric materials and their charge outputs [ 9 ] and through a variety of optical methods including polycarbonate arrays [ 6 ], optical bend loss [ 7 ] and laser interferometry of bi-refringent films [ 8 ]. Perry et al. [ 10 ] used an array-based device [ 11 ] to study bunching and stretching of adjacent plantar tissue and they found that tissue stretching from shear stress was the predominant mechanism. They report that peak shear stress and peak plantar pressure occur in the same place in 50% of cases, but actually occur at different times, which is contradictory to results reported by other researchers [ 12 ]. Contradictory results are typical from these studies using custom-built shear stress measurement devices due to the relatively low numbers of participants with diabetes tested in the trials, with typical sample sizes of ten. All these measurement methods are bespoke devices and only a handful of foot-to-floor shear stress measurement devices exist worldwide. Larger scale studies with matched control groups are required to provide firm conclusions on plantar surface shear stresses experienced by people with diabetes.

Shear stress measurement is further complicated as all diabetic patients are strongly advised to walk using footwear (and never barefoot), therefore, to understand the shear stresses induced on the plantar surface, in-shoe shear stress measurement must be taken. Although direct shear stress measurement is important in DFU risk management, future use of artificial intelligence methods [ 13 , 14 ] may enable risk management with current measurement technologies.

In-shoe plantar shear stress is difficult to measure and reported measurements vary widely, for example, measurements of shear stress on the 1st metatarsal head varied from 16 kPa [ 15 ] to 140 kPa [ 5 ] in healthy participants. Therefore, either there is widespread inter-participant variability and/or there are mechanisms which cause errors for in-shoe shear stress measurement. Measurement error has been widely reported for in-shoe normal stress systems with causation linked to sensor wear and calibration [ 16 , 17 ]. Specifically, calibrating with similar load ranges to those desired to be measured improved accuracy by up to 20 times [ 16 ] and accuracy was reduced when smaller areas of loading were applied [ 17 ]. It is likely that similar calibration issues will affect in-shoe shear stress sensor measurement accuracy. Researchers have made excellent progress in developing novel in-shoe plantar shear stress measurement systems; however, they have not yet fully considered the implications of calibration methods on measurement accuracy [ 4 , 5 ]. The choice of indenter area of loading, shape and location is also an important consideration for accurate and reliable sensor calibration; despite this, to the authors’ knowledge this has not been investigated and reported in the literature. A key principle in calibration is that the applied loading should be a good representation of the real-world scenario. In the context of plantar foot mechanics, and for example the metatarsal heads, there is variation in the magnitude of loading, area of loading, shape and potentially the location of the bone in relation to the sensor. This paper presents the design and evaluation of an in-shoe shear stress sensor and considers the impact of calibration on measurement accuracy.

This paper describes a sensor system design and conducts a performance investigation. Three investigations were conducted: calibration investigation, loading profile comparison and sensor validation. These investigations and how they relate to one another are shown in Fig 1 .

• PPT PowerPoint slide
• PNG larger image
• TIFF original image

https://doi.org/10.1371/journal.pone.0309514.g001

Sensor system design

Sensing principle..

Coulomb’s law of friction describes frictional force being proportional to reaction load. In the case of shear sensing insoles this means that there can be no shear stress (friction) without normal stress (reaction load) and that the magnitude of associated shear stress will always be less than that of normal stress. Like most other shear sensors in the literature [ 5 , 15 , 18 – 22 ] the shear sensor is embedded in a hyperelastic or viscoelastic, isotropic incompressible elastomer, as opposed to a discrete sensor placed on the insole or isolated from the main body of the insole material.

Fig 2A shows a cylindrical section of elastomer insole with cross-sectional area, A, containing a strain gauge orientated in the shear plane and a normal stress sensor with sensor readings in mV, S, and N, respectively. The material properties (stress-strain relationship) for the silicone are non-linear but can be approximated as three linear regions (low: ≤ ε 1 = 0.04 strain ; medium: ≤ ε 2 = 0.115 strain ; high: > ε 2 strain ); see Fig 2B . The strains for the three linear regions were determined from the stress-strain curve of the silicone under compressive loading at the target stresses of 14 kPa (low), 70 kPa (medium) and 140 kPa (high). Stress-strain relationships for normal compressive loading are given by Eq 1 , where C medium , and C high are negative intercepts in units of pascal ( C low = 0) and E is the gradient in Pascal.

[A] Cylindrical section of elastomer containing strain gauge and normal force sensor [B] Stress-strain curve of the elastomer under compression stress. Linear approximations for deformation were made for three regions of the curve (low, medium, and high stress magnitudes), sectioned by the compressive strains ε 1 and ε 2 , with corresponding gradient E used for calibration. [C] Cylindrical section deformed by normal force only. [D] Cylindrical section deformed by both normal and shear forces.

https://doi.org/10.1371/journal.pone.0309514.g002

Fig 2C shows the section being loaded with a normal force which creates a reduction in thickness but an increase in diameter described by Eq 2 (assuming constant volume) which gives sensor readings S N and N N , which are signal voltage measurements (mV) for the shear stress and normal stress respectively, described by Eqs 3 and 4 where k N and k S are constants (sensor gains) determined by experiment with units Pa/mV and strain/mV respectively (other equation parameters defined in Fig 2 with SI units).

Fig 2D shows applied loading from both normal and shear force giving a sensor reading S N + S and N N for the shear stress and normal stress respectively. The applied shear stress, σ S , can be determined from Eq 5 and Eq 1 (assuming an isotropic material) which requires measurements from the normal stress sensor, N N , to decouple the effect on the strain gauge from normal force (where i = low , medium or high ).

Shear stress sensor design.

The shear stress sensing system primarily consists of the strain gauge rosette, a normal stress sensor, and the flexion stiffener and load concentrator; here on in referred to as the ‘shear stress system sensor’ or ‘SSS sensor’. A 3-element strain measuring rosette (1-RY81-3/120, Hottinger Bruel & Kjaer UK Ltd, Royston, England) was chosen for the shear stress sensor ( Fig 3A ) arranged in rectangular 0°-45°-90° directions to allow for calculation of resultant shear in both the anterior-posterior (AP) and medial-lateral (ML) directions. The sensor was then embedded in silicone (Sil A50 Smooth- Sil Addition Cure silicone, Smooth-On Inc. Macungie, USA). To assemble the sensor, the first 2mm silicone base layer was poured into a custom 3D printed square mould with dimensions of 20 x 20 x 4 mm (width x length x height). After curing the surface was cleaned and the strain gauges were soldered to 2-core 2.8 mm 2 external diameter shielded wires (JY-1060, Pro-Power by Newark, Chicago, USA). The strain gauges were then placed on the surface of the silicone using a custom 3D printed jig with tabs and bolts to align the strain gauges in the correct angular position. A thin second layer of silicone (approximately 0.5 mm thick) was then poured and allowed to fully cure, the jig was then removed and a final layer of silicone was poured on top to give a total thickness of 4 mm. A 15 mm diameter, 0.8 mm thick phenolic sheet material flexion stiffener and load concentrator was placed at the center of the sensor assembly and the top layer of silicone was then allowed to cure. The full assembly of the sensor is shown in Fig 3B and 3C .

[A] Configuration of the strain rosette in the sensor with three strain gauges arranged at 0°– 45° - 90°, and the relationship between the local strain axes and the global applied shear direction axes (Medial-Lateral, ML, and Anterior-Posterior, AP). [B] Section view of the SSS sensor. [C] Top view of the SSS sensor and its dimensions. [D] Locations of SSS sensors in the sensing insole.

https://doi.org/10.1371/journal.pone.0309514.g003

As mentioned in the sensing principle, the shear stress is obtained from Eq 5 , however, for the SSS sensor to measure both AP and ML shear stress, orientation of the strain gauges needs to be considered. From the configuration shown in Fig 3A for stress measurements calculated from strain gauges A, B and C the shear stress is given by Eqs 6 and 7 .

Where θ AB and θ BC are the angles between the individual strain gauges in the rosette, which were at 45°.

Shear stress sensor number, placement, and integration.

Key DFU risk areas, accounting for at more than one-third of DFU cases are the calcaneus, first metatarsal head and hallux areas of the foot [ 23 – 26 ], so placement of the SSS sensors in the insole was in these three locations. To maximize accuracy of the measured sensing data, all sensors were anatomically matched to the participant. This was achieved through a ‘palpation and marking paper’ approach in which a healthcare professional identified the bony landmarks of the foot, marked these areas on the foot surface with skin-safe marker, and the participant stands on the paper to transfer the markings. These markings were then used to ensure SSS sensors were correctly located on the silicone insole, with the sensor x-axis aligned with the anterior posterior direction. The signal wires were laid out from the SSS sensor in the ML direction to reduce fatigue loading from flexion during gait. A 1–2 mm depth of silicone was then poured and cured before a further layer of silicone was poured and cured to make a total insole thickness of 5 mm to complete the insole, as shown in Fig 3D . Three normal stress sensors (A301 FlexiForce 0-44N, Tekscan Inc., Norwood, Massachusetts, USA) were then secured to the bottom of the insole with silicone glue (Permatex 80050 Clear RTV Silicone Adhesive Sealant, Permatex, Illinois Tool Works Inc., Solon, Ohio, USA) with their center coincident with the SSS sensors.

Data acquisition system (DAQ) and signal processing.

A Teensy 4.1 32-bit microcontroller (PJRC, Portland, Oregon, USA), ARM Cortex-M7 processor, with clock speed of 600 MHz and integrated SD storage card, was used to collect and store the voltage readings from the SSS sensors ( Fig 4B ). Flexiforce normal stress sensors were connected via a 10 kOhm circuit divider to analog inputs, whilst shear sensing strain gauges were amplified using a 24-bit high-precision analog-to-digital amplifier (HX711 ADC, HALJIA, Zhongai, China) then routed to digital inputs of the microcontroller. All signals were collected at a sampling rate of 80 Hz. Data was logged to the 16GB SD card and streamed via an ESP8266 UART WiFi adapter (Espressif Systems, Shanghai, China) to allow for continuous monitoring. Power was supplied to all components via 3V and 5V power rails from the microcontroller, sourced from an external 3.7V 3500mAh Lithium Polymer battery (LP104567, EEMB, Moscow, Russian Federation) that was regulated through a linear regulator (LDO, B08HQQ32M2, DollaTek, Hong Kong, China). For both left and right foot measurements, two identical systems were used to collect the measurements, and placed on a custom, adjustable neoprene fitness belt (Frienda, China), during walking trials ( Fig 4 ).

[A] Participant walking on a treadmill with the sensor insole system. The data acquisition system (DAQ) was attached to a belt, and each insole (left and right foot) has a separate but identical DAQ input. [B] Block diagram of the DAQ system, collecting data at 80 Hz.

https://doi.org/10.1371/journal.pone.0309514.g004

A custom MATLAB (The Mathworks Inc., Natick Massachusetts, USA) script was used to parse and analyse the data collected. The data was minimally pre-processed before finalized into calibrated stress measurements. This pre-processing stage included removing only obvious outliers (which accounted for up to 0.05% of the measurement data if present). This was made using the filloutlier function with the ‘quartile’ outlier detection option: ‘quartiles’ outliers which were elements more than 1.5 interquartile ranges above the upper quartile (75 percent) or below the lower quartile (25 percent)) and correcting DC offsets. Data from each foot were analyzed separately.

Calibration investigation: Bench top mechanical testing

Experimental setup and test method..

To investigate the effect of calibration on the sensor’s performance, both shear and normal force were applied to the SSS sensor insole (summarised in Fig 1A ). A uniaxial mechanical testing machine (Instron 5982K2680 100kN 350°C, 500N load cell, Instron ® Norwood, Massachusetts, USA) applied and measured shear force using a bespoke shear stress rig through an indenter of area, A, shown in Fig 5A . A normal reaction force was applied through a screw thread to the indenter to facilitate frictional shear stress application. Measurement of normal reaction force was through a load cell and ADC (‎ADN1903027, 196.2 N Weight Sensor Load Cell, Haljia, China) capturing data at 80Hz using an Arduino (Arduino Mega 2560 Rev3, Arduino, Somerville, MA, USA). For pure normal stress loading calibration, the insole was placed flat on a plate in the uniaxial testing machine fitted with a large compression platen on the bottom and an indenter with a specific area, A, applying compression force from the top, shown in Fig 5B .

[A] Custom shear stress rig made of rigid 10 mm acrylic sheet plates which applied the force of the mechanical testing machine as a shear force onto the insole. The shear stress was calculated using the applied force and area of the custom indenter. The indenter’s compressive stiffness was 30.1 MPa, ~12 times stiffer than the silicone sensor of 2.5 MPa. [B] Custom normal stress calibration setup where the insole was placed on a compression platen.

https://doi.org/10.1371/journal.pone.0309514.g005

Sensor loading area investigation.

To evaluate the effect of indenter area, A, five flat ended cylindrical indenters with diameters of, 10, 15, 20, 25, 30 mm were used to load the SSS sensor at its center. While studies have shown that there is a difference between various indenter shape loading profiles and the corresponding mechanical responses of the material [ 27 , 28 ], we determined that the normal stress distribution that was measured at the surface of the SSS sensor was similar for both flat and rounded indenter profiles. The only notable difference was the size of the normal stress distribution, as a flat indenter covered a larger area than the rounded indenter of the same diameter. Thus, choosing a flat indenter of a smaller size gave the same loading results as a larger rounded indenter.

The tests applied a cyclic shear force with a 1 Hz triangular waveform pattern ranging from 0 to 50 N in combination with a constant normal stress of 140 kPa through all the indenters. SSS Sensor output signal, S N + S , in mV was measured for each of the loading areas.

Sensor loading location investigation.

Ideally a sensor would be co-located with the anatomical part applying the load, however, this may not always be practically possible so an understanding of the relationship between the location of the SSS sensor, the location of the applied loading and the accuracy of measurement is required. To investigate the effect of loading location, twelve loading locations were chosen, six in the anterior direction and six in the lateral direction both measuring 0, 10, 15, 20, 30, 40 mm from the center of the shear stress sensor. Loads were applied in both the medial or posterior direction respectively. Cyclic loading was applied to the SSS sensor insole of the same characteristic as the area of loading investigation (see ‘Sensing loading area investigation’ section). SSS Sensor output signal, S N + S , in mV was measured for each of the loading locations.

Loading profile comparison: Human plantar loading specific sensor calibration

Comparison of normal stress profiles..

Shear loading application area and location affect strain measurements, so it is important to consider plantar stress loading from the human foot. During walking plantar stress is dependent on many factors including foot size and anatomy, weight, morbidity and walking patterns, all of which are different between participants. From the sensor calibration investigations in the results section, we can see that (i) loading location and (ii) loading area may affect the output of the SSS sensor so these must be considered during calibration.

• Loading location variation can be removed by placing the SSS sensors at personalised anatomical locations in the insole, which is the approach we have taken.
• Loading area variation can be controlled through calibration. This was determined through a comparison and matching of normal stress loading profiles of the specific participant’s foot anatomy with bench top mechanical test experiments involving various loading area sizes (flat cylindrical indenters).

To capture the plantar normal stress loading profiles of our participants, in the SSS sensor locations of the calcaneus, first metatarsal head and the hallux, we conducted measurements in-shoe during a two-minute treadmill walk using an F-scan insole (Tekscan Inc., Boston, USA) coupled with a non-instrumented insole of the same material properties and thickness as our designed insole. Then the test rig ( Fig 4B ) was used with 15, 20, 30 and 40 mm diameter indenter sizes to load the silicone insole from 0 to 250 N (to simulate a normal stress range up to 1400 kPa, which is comparable to the 1000–1900 kPa normal plantar stresses during gait reported in the literature [ 29 , 30 ].

Measurements of plantar normal stress distribution were captured with the same F-scan and insole used with the participants. To simulate the different foot structures, we adjusted the diameter of cylindrical indenters (15, 20, 30 and 40 mm), which were based on the ranges of average anatomical dimensions of the hallux, metatarsal head, and calcaneus bones [ 31 – 36 ], see results and discussion ‘Human plantar loading consideration for sensor calibration’ section. An illustrated summary of this investigation can also be found in Fig 1B .

Statistical analysis as a method for calibration indenter choice.

Comparisons were made between the participant’s mean normal stress profiles with the bench top test rig results (gait data averaged over 20 gait cycles from three different sensing locations hallux, first metatarsal head, and calcaneus, bench top test rig results for 15, 20, 30 and 40 mm indenter diameters). Magnitudes of both results were scaled to have a maximum unity magnitude to enable comparison. The normal stress profiles (normal stress vs displacement across anatomical location) were collected along a 2D cross section of 40 mm in length across the foot-width of loaded area (see results and discussion ‘Human plantar loading consideration for sensor calibration’ section). Calibration indenter diameters for the hallux, first metatarsal head and calcaneus locations were chosen based on either the highest R 2 value from a multiple linear regression between the gait measures and the test rig measures or the maximum measurement sensitivity area of the SSS sensor (see results and discussion ‘Sensor calibration’ section).

Sensor validation: Bench top mechanical testing

The following section describes the sensor validation, as summarised in Fig 1C . A 30 mm diameter indenter was used to calibrate the SSS sensor, as this was determined to be the maximum sensing area of the sensor (see results and discussion ‘Sensor calibration’ section). This was achieved through a series of mechanical tests detailed in Table 1 , with shear stresses applied in both ML and AP directions and conducted at 1Hz, to simulate average walking speed frequency.

https://doi.org/10.1371/journal.pone.0309514.t001

The shear stress magnitudes chosen for low, medium, and high levels were 10%, 50% and 100% of the 140 kPa maximum in-shoe plantar shear stress reported in the literature respectively [ 37 ]. This enabled calculation of the calibration parameters coefficients E low , E high , C medium and C high , according to Eq 1 .

To validate the calibrated SSS sensor, a shear stress of 70 kPa with a normal stress of 125 kPa was applied in both the ML and AP direction at 0.8 Hz. Additionally, a shear stress was also applied in the 45° direction (14 kPa shear stress, 28 kPa normal stress at 1Hz).

Two measurements of error were made. The first was an overall mean absolute error (MAE), which is the mean of the difference between the measurement from the test rig and the calibrated SSS sensor measurement (in kPa). The second was peak error, measured as the percentage error at peak loads between the applied measurement from the test rig and the calibrated SSS sensor measurement. Peak values of measured shear stress were taken from 10 cycles and a standard deviation was calculated. Repeatability was calculated from the SSS sensor measurements as the standard deviation of the peak plantar stresses divided by the mean of the peak plantar stress, presented as percentage (e.g. a mean peak measurement of 100 kPa and a standard deviation of those peak measurements at ± 10 kPa, would result in (10/100) x 100% = 10% deviation from the peak value, and thus 90% repeatability).

Sensor validation: Gait lab treadmill walking

To further validate the sensors, a gait laboratory treadmill walking test was performed on a single anthropometrically matched healthy participant and a single participant with diabetes (both male and 45 years old, weighing 88 kg and 75 kg, height of 1.75 and 1.66 m, EU shoe size 44 and 42, weight per insole area 32 kPa and 35 kPa, walking speed 0.92 ms -1 and 0.95 ms -1 for the healthy participant and participant with diabetes respectively). The study received approval from the NHS Health Research Authority and Health and Care Research Wales (HCRW) Ethics Committee (REC reference: 22/NW/0216), and all participants provided written consent. Trial Registration number: NCT05865353. Participants were recruited between 1 st November 2022 till 30 th May 2023. Data collection was conducted in two parts (1) baseline visit and (2) main data collection, Table 2 .

https://doi.org/10.1371/journal.pone.0309514.t002

Baseline visit.

Anthropometric data was collected, and anatomical landmarks determined using the ‘palpation and marking paper’ method described in the ‘Shear stress sensor number, placement and integration’ section. The participants conducted a 2 minute treadmill walk while wearing a pair of silicone insoles, made from the same materials and dimensions as the sensor insole but without active sensors, and a pair of F-Scan pressure sensing insoles, in a prophylactic shoe (Sponarind 97308, Finn Comfort Inc. Hassfurt, Bavaria, Germany), designed with shock-absorbing properties and a larger volume, ideal for people with diabetes. Normal stress data was collected using the F-scan insoles, at a self-selected gait speed to determine normal plantar stress profiles (results of which were used for the comparison of normal stress profiles, in ‘Human plantar loading specific sensor calibration’ section). Table 2 shows the participant data collected during the baseline visit.

Main data collection.

The participants returned for the main data collection where they were asked to wear the sensing insole in the specialist diabetic shoe. They then walked twice on a split belt treadmill with integrated force plates (M-Gait, Motek Medical BV, Amsterdam, Netherlands) for 15 minutes at their self-selected speed (see Table 2 ).

Data analysis: Shear stress gait measures and repeatability.

Mean and standard deviation of peak shear stress and peak normal stress measurements were extracted from 20 gait cycles measured by the sensing insole. Measurement repeatability was determined and comparisons, between the two walking periods within each individual walking session (start, middle, and end). We collected statistical data for both plantar shear stress and normal stress measurements to perform inter-participant comparisons. These included statistics for Plantar Stresses (Normal, AP Shear, and ML Shear) across all three sensor areas, encompassing mean values, standard deviations, peak stresses, and variability (or percentage difference) of measurements within the 15-minute treadmill walk (intra-walk) and between two treadmill walks (inter-walk).

Results and discussion

Sensor calibration.

Shear stress measurement accuracy is affected by the calibration method. Specifically, the shear stress sensor measured output signal decreases exponentially with both increasing loading application area, and increasing loading distance away from sensor center, see Fig 6 . The results in Fig 6A show that the measured output decreases by ~80% from 1.5 mV to 0.3 mV, for a calibration loading application area of 10 mm diameter to 30 mm diameter respectively. This means that if the sensor was calibrated for the smaller 10 mm area and a larger 30 mm diameter load was applied, the measurements would be underestimated by 80%. Likewise, calibrating for a larger area, and applying load for a small area will greatly overestimate the measurements. Increasing the loading application area increases the area over which the force is distributed over the sensor, thus more of the loading is applied away from the center of the shear stress sensor. From the results shown in Fig 6B and 6C the location of loading application also reduces sensor sensitivity. All this means that the shear stress sensor will only be able to measure accurately if the calibration loading area matches the desired measurement loading application area (or are reasonable similar areas).

Mean peak signal of shear stress (SSS sensor) total output (mV) from 10 cyclic triangular loading. [A]—Effect of area of loading on SSS sensor measured outputs. [B, C]- Effect of location of loading on SSS sensor output for medial and posterior respectively.

https://doi.org/10.1371/journal.pone.0309514.g006

Fig 6B and 6C show the influence of loading location on SSS sensor measurements for the same applied loading area (25 mm diameter indenter). As expected, the SSS sensor measurement for both Anterior-Posterior (AP) and Medial-Lateral (ML) shear loading decreased as the loading distance moved away from the sensor center. This is due to a decrease in deformation of the shear stress sensor as the loading is applied further away from the sensor center. However, it is important to note that there was still a measurable signal at these distances as they are not yet relatively far away from the sensor. This means that measured shear stress from an embedded sensor will not just be from the coincident anatomical location but also have a contribution from adjacent and other relatively close anatomies (e.g. first metatarsal head located sensor may be measuring shear stress contribution from the second metatarsal head). This is due to material coupling, which is that stress applied in one area of the material, in this case the silicone insole, will stress surrounding areas of the material. The implication is that the shear stress sensor will provide more accurate measurements if the loading application location is coincident with the centre of the sensor. This emphasizes the importance of the placement of these discrete sensors, which is why a participant specific sensing insole was manufactured, placing sensors at the exact anatomical location of the boney landmarks, where peak loading is expected.

Although this paper presents the shear stress sensor sensitivities to calibration loading area and calibration loading locations for this sensor it is likely that these observations are true for other embedded in-shoe shear stress sensors. Other researchers measured in-shoe peak shear stresses from gait varied from 9 kPa to 140 kPa and calibration loading area varied from 20 mm diameter area (314 mm 2 ) –10,000 mm 2 (up to half the insole, approximated from the experimental Fig 3 in the paper as there was insufficient detail to give conclusive information on the loading area used) [ 5 , 15 ]. It is likely that these variations in measurements are not due to inherent sensor inaccuracy or participant gait differences but likely to stem from calibration method differences. To the authors’ knowledge, calibration loading area has not been investigated in other published studies, but it is suggested that calibration should be considered for all future in-shoe shear stress measurements.

Human plantar loading consideration for sensor calibration

Fig 7 shows that calibration loading indenter diameters should be 20 mm and 40 mm for the hallux and both the first metatarsal head and the calcaneus respectively. However, due to limitations on sensor sensitivity beyond 30 mm from the center of the sensor a 30 mm indenter diameter was chosen for the first metatarsal head and calcaneus. These choices of calibration indenter diameters were determined from the comparison of the bench top testing normal stress profiles of different indenter diameters, with the participants’ measured normal stress profile during walking. The bench top test showed that all the indenters resulted in normally distributed normal stress profile curves ( Fig 7A ), increasing in curve width with increasing indenter diameters, reflecting a larger contact area of the applied force. An increasing curve width is also expected for the normal stress profiles of anatomical bones with increasing diameters (first metatarsal head ~15 mm, hallux ~20 mm, and calcaneus ~ 40 mm [ 31 – 36 , 38 ]). The participants’ measured normal stress for the hallux and the calcaneus regions of the foot had normal pressure distribution profiles that reflected their anatomical sizes, however, the presence of the second close metatarsal bone influenced the normal stress profile in the first metatarsal head area and widened the normal stress profile, more than what is expected from its anatomical diameter of ~15 mm ( Fig 7B ). The R 2 results of the multiple linear regression reflected this ( Fig 7C ), as the first metatarsal head correlates to the indenter size of 40 mm diameter. The R 2 value of the metatarsal head, however, is small at 0.41, indicating that there may be variability in the pressure distributions in that area, likely from gait variability within a participant’s walk or between participants. The hallux and calcaneus regions of the foot have a normal pressure distribution profile that reflects the loading of the anatomical bones clearly (R 2 ≥ 0.95) and can be matched with an indenter of a similar size to give a representative loading for calibration of 20 mm and 40 mm respectively. However, loading area results from Fig 6A show that sensor sensitivity converges for indenter areas greater than 25–30 mm diameter. Therefore, calibration indenter diameters were reduced to 30 mm for the first metatarsal head and calcaneus.

[A] Experimental normal pressure profiles: (i) Indenter experimental setup, (ii) Normal pressure profile curves width increases with increasing indenter diameter, (iii) F-scan pressure result that shows the cross section used to obtain these values used in ii. [B] Participant pressure profiles: (i) In-shoe gait lab experimental setup (ii) An example of participant’s pressure profile over 20 gait cycles, showing the three normal pressure profiles of the foot at the calcaneus, first met head and hallux, (iii) F-scan pressure result that shows the cross section used to obtain the values. The image also shows the peaks for these three regions (Calcaneus peak CP, Hallux Peak HP and the metatarsal peaks MHP1 and MHP2). [C] Graphical representation of the regression analysis’ coefficient of determination (or R-squared) results. Larger circles indicate a higher R-squared value, and red circles indicate the maximum R-squared in the sensor group. R-squared values are shown above the circles, and maximum is indicated as red font.

https://doi.org/10.1371/journal.pone.0309514.g007

The implications of this for the SSS sensor are that calibration indenter sizes should be between 10–30 mm dependent on expected shear stress application areas. This finding is likely to be true for other embedded in-shoe shear stress sensors in the literature. The limitation from this finding is that to obtain accurate shear stress measurements the user must know something about the shear stress loading profile which may be unknown. A possible way to mitigate for this may be to calibrate the sensor for a range of loading areas and to use a normal stress sensor to determine which indenter calibration area to use in post-processing.

Shear sensor calibration and bench top mechanical test validation

The SSS sensor was highly accurate and repeatable when compared against the bench top mechanical test as seen in Fig 8 . Results from Table 3 show that calibration error was insignificant with the mean absolute error (MAE) over the entire cycle in calibration < 0.00007 kPa for all magnitudes of loading, and errors at peak loading were < 5.8%.

[A] Sensor calibration for both anterior-posterior (AP) and medial-lateral (ML) directions at a ‘medium’ level of posterior and medial shear loading of 1 Hz cyclic loading of up to 70 kPa shear stress, at a constant normal stress of 140 kPa. [B] Sensor validation test result at medium level of shear cyclic loading (up to 70 kPa), at a different loading frequency (~0.85 Hz) and different constant normal stress (125kPa). All results for the different configurations of loading are shown in Table 3 .

https://doi.org/10.1371/journal.pone.0309514.g008

https://doi.org/10.1371/journal.pone.0309514.t003

The errors in the validation of the sensors at loading conditions different from the calibration were higher, but still showed a high accuracy for the sensors. The sensor was most accurate for low–medium shear stress magnitudes with up to <1.8 kPa for MAE, and < 8.7% for error at peak loading (see example of medium magnitude measurements in Fig 8 ). Followed by the measurements at a resultant loading angle of 45° clockwise from the anterior direction (MAE <1.4 kPa; <11.5% peak error). These small errors could be attributed to errors in the validation setup, as an error of ± 5° would correspond to a peak shear stress error of up to 4.6%. The SSS sensor also showed good repeatability for all loading conditions (>97% repeatability in calibration and >96% repeatability in validation).

The highest errors in validation were at high shear stress magnitudes, over the expected plantar shear stress from gait, these were MAE <17.3 kPa and peak error <22.4%. This was likely due to the mechanical coupling of the high normal stress, pushing the total material deformation higher up the hyperelastic stress-strain curve of the sensor material ( Fig 2D ). At this region of the stress-strain curve, very small strains relate to high changes in stress making the SSS sensor more prone to measurement errors. However, the maximum errors translate to an error of ± 31.3 kPa, which is within the standard deviation of most plantar stress measurements from the literature of ± 50 kPa for shear stress [ 1 – 5 , 15 ].

Treadmill walking validation

For treadmill walking the SSS sensors measured the magnitude of shear stresses between 66.5 kPa—152.6 kPa in the AP direction, and 28.4 kPa– 128 kPa in the ML direction, full results are shown in Table 4 . As expected, the ML shear range was lower than the AP shear range, as loading was expected to be predominantly in the AP direction. Loads were cyclic going from zero to peak value with the same frequency as gait which were at speeds of 0.92 and 0.95ms -1 for the healthy participant and participant with diabetes respectively. The only notable differences were in the direction of some of the peak plantar shear stresses.

https://doi.org/10.1371/journal.pone.0309514.t004

No significant differences between both participants peak plantar stress values were observed (t-test of mean peak plantar stresses PPS, p>0.36, p>0.58 and p>0.57). This was expected, as both participants had a similar walking speed (0.92–0.95 ms -1 , and weight per insole area 32.4–35 kPa). However, this study aimed to demonstrate the feasibility, accuracy, and repeatability of the SSS system so no conclusions should be drawn on plantar stress for general people with diabetes and healthy populations for this study.

The shear measurements of the SSS sensor was highly repeatable when comparing data recorded for both within the 15-minute treadmill walk (intra-walk), and between the two 15-minute walks (inter-walk). The mean and standard deviation of the percentage difference of peak plantar stresses were ≤ 8% ± 6% for both investigations. Intra-walk differences were lower than inter-walk–with the highest percentage difference of 21% measured by the SSS sensor for the ML Shear (Hallux, Left foot, participant with diabetes). Other measurements from the shear stress sensors were < 15% difference. For inter-walk, the highest PPS percentage difference was measured by the commercial Flexiforce sensor of 47% difference in normal stress (Hallux, left foot, participant with diabetes), followed by 37% for the AP shear of the SSS sensor (Hallux, right, healthy) and 33% for the ML shear of the SSS sensor (Calcaneus, right, participant with diabetes).

Calibration and material coupling for shear stress sensors

To the author’s knowledge, this study is the first to address in-shoe shear sensing material coupling and unexplored complexities in calibration for shear sensing. The results illustrate that due to sensor and material coupling with adjacent structures the area which contributes to the measured shear can be larger than the area of the sensor. This has important implications for shear sensor calibration, firstly in terms of the location of the sensor and the anatomical region that is to be measured, and secondly in terms of the indenter area used for calibration. These results have significance for all researchers developing systems to measure in-shoe plantar shear stress as these factors will affect the magnitude of shear sensed. Furthermore, these results may partially explain the variation in magnitudes of shear measured at the same anatomical locations by different researchers. A suggested approach for shear sensor calibration is shown below (for detail see methods ‘Human plantar loading specific sensor calibration’ section):

• Determine the sensing area : Material coupling between the shear sensor and adjacent regions can result in the area sensed being greater than then actual area of the sensor.
• Determine the distribution of plantar loading : Normal stress distribution will be indicative of shear stress distribution, whilst foot anatomy, for example the hallux, will determine the loading area.
• Decision for calibration indenter area : Informed by both the sensing area and the distribution and magnitude of plantar loading.

Developed shear stress system sensor

Sensor performance..

A novel Shear Stress System (SSS) sensor composed of a strain gauge rosette, normal pressure sensor and stiffener to concentrate loading at the desired sensor location and mitigate against material coupling was developed and evaluated. Sensor locations were anatomically matched and measured the plantar loading profiles to inform calibration of each sensor at a specific location. This study conducted a thorough experimental validation of the shear sensor through mechanical bench top testing and with human participant treadmill walking. Shear sensing results demonstrated high repeatability (>97%) and high accuracy in the expected measurement range for plantar shear stress (mean absolute errors < ±2 kPa) with error increasing for very high shear stresses (mean absolute errors < ±17 kPa) compared to bench top mechanical tests and repeatability for treadmill walking of 15-minutes duration with less than 21% variability within walking, and less than 37% variability between walks (which was lower than the commercial normal pressure sensors of 47% used in this study).

Limitations.

A rosette strain gauge was chosen for determining unknown principal directions, however it restricted complete strain separation in the AP and ML directions. For exclusive separation, a 0°–90° strain gauge in the ML and AP axes could be adopted. The manual assembly of the sensors and alignment of the sensor in relation to the AP and ML directions affect shear measurement. This has been controlled through careful manufacture, but some small errors will remain. The chosen alignment of the strain gauge rosette in the ML direction was to reduce the fatigue on the soldered joints, this resulted in a decreased sensitivity in the AP direction due to the 45° off-alignment of the gauges with this axis.

Relative stiffness of the silicone and the strain gauge rosette will affect strain transfer between the two materials. Material properties of the silicone is highly important for measurement accuracy, sensitivity, and range, and warrants further investigation.

Future work.

A three-part linear fitting procedure was adopted to calibrate the SSS sensor accommodating the hyperelastic material properties, in the future consideration of alternative fits to capture viscoelastic effects could be made. Despite observing minimal shear sensor temperature response, variability between 20–30°C, literature indicates foot temperatures may be as high as 35° in people with diabetes [ 39 , 40 ], this should be considered in the future. In this proof-of-concept study, the size of calibration area was based on average pressure profiles, a suitable assumption with little participant variation. However, future larger studies may require participant-specific calibration to address varying loading profiles, particularly due to gait variability.

• View Article
• PubMed/NCBI
• Google Scholar
• 22. Amemiya A, Noguchi H, Oe M, Sanada H, Mori T. Establishment of a measurement method for in-shoe pressure and shear stress in specific regions for diabetic ulcer prevention. 2016 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC). IEEE; 2016.