Top 10 Tips for Writing Materials and Methods in Radiology: A Brief Guide for Authors

HomeRadiologyVol. 310, No. 3 PreviousNext Reviews and CommentaryFree AccessEditorialTop 10 Tips for Writing Materials and Methods in Radiology: A Brief Guide for AuthorsSarah L. Atzen Sarah L. Atzen Author AffiliationsFrom the Radiological Society of North America, 820 Jorie Blvd, Oak Brook, IL 60523.Address correspondence to the author (email: [email protected]).Sarah L. Atzen Published Online:Mar 19 2024https://doi.org/10.1148/radiol.240306MoreSectionsPDF ToolsAdd to favoritesCiteTrack CitationsPermissionsReprints ShareShare onFacebookXLinked In IntroductionThe Materials and Methods section is important for any research manuscript. However, it is especially important for a journal such as Radiology, where complex imaging protocols must be reproducible. Yet, with a current word limit of only 800 words, authors may struggle with what information to include. Although now published online only, adhering to word count in Radiology remains important. There are two reasons for this: readability and expediting the production of your manuscript.Therefore, this editorial provides key terms and tips to guide authors in writing their Materials and Methods in Radiology. These should be relevant for most human studies. For systematic reviews and meta-analyses of diagnostic test accuracy studies, please review and complete the Preferred Reporting Items for a Systematic Review and Meta-Analysis of Diagnostic Test Accuracy, or PRISMA-DTA, checklist (1) before submission.Specific tips for manuscripts related to artificial intelligence (AI) will be covered in an upcoming article. However, a good place to start is the following publication from the Radiology Editorial Board: "Assessing Radiology Research on Artificial Intelligence: A Brief Guide for Authors, Reviewers, and Readers" (2) and the Checklist for Artificial Intelligence in Medical Imaging (CLAIM) (https://pubs.rsna.org/doi/suppl/10.1148/ryai.2020200029/suppl_file/ryai200029_CLAIMchecklist.docx) (3). For more writing tips, please see "Getting Started with your Paper" found at https://pubs.rsna.org/page/radiology/author-instructions#getStarted (4–6).Key terms often included in Radiology's Materials and Methods sections are described in Table 1 (7–16). Of note, authors often mislabel their radiologic research studies (10). Therefore, a good understanding of the difference between prospective and retrospective studies, as well as other terms, is essential before publishing in Radiology. To better understand the difference between study types, along with their pros and cons, please see the article by Alvin et al (9). Once you understand these terms, please see the following top 10 tips for writing your Materials and Methods, as well as Table 2. The checklists (STARD [Standards for Reporting of Diagnostic Accuracy Studies], STROBE [Strengthening the Reporting of Observational Studies in Epidemiology], CONSORT [Consolidated Standards of Reporting Trials]) mentioned in Table 2 can be found at https://pubs.rsna.org/page/radiology/author-instructions#checklist.Table 1: Key Terms for the Materials and Methods SectionTable 2: Top Tips from Radiology for Writing Your Materials and Methods1. Get your regulatory approvals out of the way first.At the beginning of Materials and Methods, name your type of study (randomized trial, prospective, retrospective, secondary analysis). Include the following items as applicable:Any financial support from industry;Institutional review board or ethics committee approval;Informed consent obtained (specify written, verbal) or waiver granted;Health Insurance Portability and Accountability Act approval (United States only);Animal Care Committee approval (animal studies only);Registry name and number (eg, ClinicalTrials.gov: NCT01234567) for all randomized trials and prospective studies, and secondary analyses as applicable.2. Indicate the use of any AI or AI-assisted technologies in preparing your manuscript.At submission, we ask authors to indicate if they have used any AI or AI-assisted technologies in their manuscript preparation. These tools are numerous and growing. They include large language models (LLMs), chatbots, (eg, OpenAI's ChatGPT, Google Bard [now Gemini]), and image creators. Radiology expects authors to be transparent about their use of these technologies. Authors must report how they used such tools and provide information about their use and details such as the name and version of the AI tool, date of access, and name of the manufacturer or creator. We ask authors to provide this information in the Materials and Methods section or in the Acknowledgments or a relevant section of the manuscript (eg, figure legends for AI-generated figures) (17). Please review the journal's LLM policy for more details at https://pubs.rsna.org/page/policies#llm.3. Use appropriate subheadings and ensure the relevant information is included under each.A common template may be used as follows:Study design and patients/participants (who, where, when, and what);Definitions of terms;Imaging protocols (enough detail that imaging can be repeated);Image analysis (independent readers or consensus, what was analyzed, reader blinding);Statistical analysis (statistical tests and analyses performed).4. Specify who, where, when, and what.Who you studied. Mention your sampling technique: These include nonprobability sampling (consecutive, convenience) or random sampling (7,18). Flinders University provides helpful definitions (19). Indicate whether matched or paired samples: A common error is stating that groups are matched or paired but treating them as independent in the statistical analysis. For matched studies, provide matching criteria and the number of exposed and unexposed (cohort studies) or number of controls per case (case-control studies) (16). Eligibility: Describe how patients (for retrospective studies) or participants (for prospective studies) were selected. A key part of your study is determining eligibility criteria. Therefore, clearly state both inclusion and exclusion criteria with a simple format (inclusion criteria consisted of a, b, c, etc; exclusion criteria consisted of a, b, c, etc). Although how you determined study eligibility (via inclusions and exclusions) is part of the Materials and Methods, the number excluded for each criterion and the final study sample should be reported in the Results. Overlap: For retrospective studies, note any overlap in patients with other studies, even if in a different type of study with different outcome assessments. Reference all studies published or in press. Also, explain if there was a different analysis of data. If too many to reference, include a supplemental table.Where your study took place: Mention if your study was a single-center or multicenter study. Please specify the type of institution where these patients were identified (eg, tertiary care center, academic medical center, community hospital, university teaching hospital). If multicenter, mention how many institutions were involved. Blind institution(s) name at first submission.When your study took place. Provide the start and end dates (month and year) of data collection or participant accrual. Examples:Patients underwent CT between July 2022 and June 2023.Patients were prospectively enrolled between July 2022 and June 2023.What you analyzed. Describe prespecified primary and secondary outcome measure(s), other variables analyzed, and interventions compared (treatment vs control group). When performing a subgroup analysis for a randomized controlled trial, mention if the subgroup analysis was prespecified (planned before looking at data) or post hoc (planned after looking at the data) (13,20). If post hoc analysis, address in statistical analysis and include as a limitation in discussion.5. Do not forget definitions of key terms.Consider including a separate section titled Definitions of Terms. This will also make your study more understandable to general radiologists. Move any long equations to the supplement for ease of reading.6. Provide imaging protocols.Be brief but include enough detail that the imaging can be repeated. For MRI, include the key sequences, such as repetition time (TR) and echo time (TE). If more detail is needed to understand the results, the full pulse sequence is likely better placed in a separate table or supplement. For CT, indicate contrast agent or noncontrast, plane, section thickness, and section interval. If necessary, provide more details in a separate table or supplement.7. Avoid promoting a particular brand or manufacturer.For any devices or contrast agents, use the proprietary (trade) name when first mentioned. Also provide manufacturer name. Thereafter, use nonproprietary (generic) names. If multiple brands of systems (eg, MRI, CT) are used in the article, a table is suggested.8. For diagnostic accuracy studies, clearly state your reference standard and index test.In a diagnostic accuracy study, you compare the index test (whatever method you are evaluating) against the reference standard (the best method currently available) (7). For example, comparing CT (index test) with MRI or pathology (reference standards). Of note, there is no perfect reference standard. For example, pathology-based reference standards are the accepted standard for the diagnosis of malignancy (21). However, they are also more prone to selection bias than clinical reference standards that exclude pathologic examination (22). Also, different diseases or conditions have different imaging reference standards. For example, MRI is considered the reference standard for early diagnosis of occult fractures rather than radiography or CT and bone scanning (23,24). However, for bone age assessment, left hand and wrist radiographs are commonly used (25). Yet, in parts of the world without access to radiography machines, portable US may offer a better alternative (25). Ensure you are using the appropriate reference standard to address your study question, and the reference standard is explicitly defined.9. Give relevant specifics for each reading and/or evaluation.Provide the number of readers and/or investigators, along with their expertise, and years of experience (for residents, indicate in-training).Provide initials for any authors. If those contributors are not authors, mention them in acknowledgments but obtain written permission to do so. Nonauthor contributor initials should not appear in the text of the article.State if using automated or semiautomated software to analyze or complete postprocessing of images; provide software name, version, and manufacturer.Mention if studies were read independently, the use of any consensus agreements, and how disagreements, if any, were resolved.Describe the blinding of data or images. Clearly state what was blinded and what was not blinded.For several imaging methods taken at one time for the same patient or participant, describe methods used to blind the reader and prevent recall.10. Wrap it up with your statistical analysis.Summarize the statistical tests and analyses performed: The statistical analysis section should be your last section before the Results. In this section, include the statistical tests performed, on what data, and for what determination.Mention descriptive statistics when summarizing data from your sample alone (non-inferential statistics). Please use standard measures (eg, percentages, means ± SDs; medians and IQRs [quartile 1-quartile 3]) (14,26).For inferential statistics (generalizing your findings to the larger population), include 95% CIs to indicate the range of likely values for key measures. The 95% CIs are required for area under the curve values, sensitivity, and specificity. For comparative studies, provide 95% CIs for the differences (effects).Mention univariable and/or multivariable analysis. For multivariable analysis, distinguish between confounding variables you adjusted for (eg, age, sex) and independent variables being investigated. Mention summary measures such as regression coefficients, correlations, odds ratios, and hazard ratios. If calculating hazard or odds ratios, mention reference categories.For randomized controlled trials, mention any post hoc subgroup analysis.Provide appropriate analysis for matched data.Reader evaluation: When using readers, indicate how intrareader and/or interreader variability was measured (eg, κ statistic, intraclass correlation coefficient) (27) and how coefficients were interpreted (eg, cutoff values for weak, moderate, strong).Intrareader variability measures differences in one reader.Interreader variability measures differences between two or more independent readers (often with varying expertise) (28).Power: Justification of sample size and/or power calculation is strongly recommended, especially in studies where the primary conclusion is of nonsignificance. In your statistical analysis section, briefly mention power, significance, statistical test, and magnitude of effect. For example, to detect an effect size of xxx with a t test with 80% power at a two-sided significance of 5% (P <. 05), yyy patients/participants are needed. If more detail is necessary to understand your calculation, please include it in a supplement. There are free programs available for sample size estimation, such as G*Power (https://www.psychologie.hhu.de/arbeitsgruppen/allgemeine-psychologie-und-arbeitspsychologie/gpower) (29). However, these should be checked by a biostatistician due to their complexity (14).Do not forget the P value used for significance, statistical software, and biostatistician. Provide the P value used for significance (for example, P < .05) and for multiple comparisons (Bonferroni) correction, if applicable. Please indicate the name of any statistical software (version, manufacturer) used for your analysis. If your biostatistician is an author, then provide their initials. Adding a biostatistician as a coauthor is a common request by the Editor at the revision stage. For more statistical tips, please review "Statistics 101 for Radiologists" by Anvari et al (14) and "Submissions to Radiology: Our Top 10 List of Statistical Errors" by Levine et al (30).Disclosures of conflicts of interest: S.L.A. RSNA employee.AcknowledgmentsThis editorial was adapted from the author's April 2023 blog post published in Radiology: Behind the Scenes. I would like to thank David Bluemke, MD, PhD; Elkan Halpern, PhD; Ariane R. Panzer, PhD; and Jenna Saleh, MS, for their insightful comments and edits.References1. Frank RA, Bossuyt PM, McInnes MDF. Systematic Reviews and Meta-Analyses of Diagnostic Test Accuracy: The PRISMA-DTA Statement. Radiology 2018;289(2):313–314. Link, Google Scholar2. Bluemke DA, Moy L, Bredella MA, et al. Assessing Radiology Research on Artificial Intelligence: A Brief Guide for Authors, Reviewers, and Readers-From the Radiology Editorial Board. Radiology 2020;294(3):487–489. Link, Google Scholar3. Mongan J, Moy L, Kahn CE Jr. Checklist for Artificial Intelligence in Medical Imaging (CLAIM): A Guide for Authors and Reviewers. Radiol Artif Intell 2020;2(2):e200029. Link, Google Scholar4. Scientific Style Guide: Writing a Manuscript for Radiology. Radiological Society of North America Web site. https://pubs.rsna.org/page/radiology/author-instructions/scientificediting. Accessed December 21, 2023. Google Scholar5. Atzen SL, Bluemke DA. Top 10 Tips for Writing Your Scientific Paper: The Radiology Scientific Style Guide. Radiology 2022;304(1):1–2. Link, Google Scholar6. Atzen SL, Bluemke DA. How to Write the Perfect Abstract for Radiology. Radiology 2022;305(3):498–501. Link, Google Scholar7. Bossuyt PM, Reitsma JB, Bruns DE, et al. STARD 2015: an updated list of essential items for reporting diagnostic accuracy studies. Radiology 2015;277(3):826–832. Link, Google Scholar8. Walsh T. Fuzzy gold standards: Approaches to handling an imperfect reference standard. J Dentistry 2018;74(Suppl 1):S47–S49. Crossref, Medline, Google Scholar9. Alvin MD, Moshiri M, Bluemke DA. Prospective Clinical Trials: 10-year Trends in Radiology. Radiology 2020;295(1):1–2. Link, Google Scholar10. Halligan S, Kenis SF, Abeyakoon O, Plumb AAO, Mallett S. How to avoid describing your radiological research study incorrectly. Eur Radiol 2020;30(8):4648–4655. Crossref, Medline, Google Scholar11. Smith AK, Ayanian JZ, Covinsky KE, et al. Conducting high-value secondary dataset analysis: an introductory guide and resources. J Gen Intern Med 2011;26(8):920–929. Crossref, Medline, Google Scholar12. JMIR Publications Knowledge Base and Help Center. Does my trial (RCT) have to be registered? JMIR Publications. https://support.jmir.org/hc/en-us/articles/115001389307-Does-my-trial-RCT-have-to-be-registered-#:~:text=In%20accordance%20with%20ICMJE%20recommendations,registry%20such%20as%20clinicaltrials.gov. Accessed December 21, 2023. Google Scholar13. Schulz KF, Altman DG, Moher D; CONSORT Group. CONSORT 2010 statement: updated guidelines for reporting parallel group randomised trials. BMJ 2010;340:c332. Crossref, Medline, Google Scholar14. Anvari A, Halpern EF, Samir AE. Statistics 101 for Radiologists. RadioGraphics 2015;35(6):1789–1801. Link, Google Scholar15. Wang X, Kattan MW. Cohort Studies: Design, Analysis, and Reporting. Chest 2020;158(1S):S72–S78. Crossref, Medline, Google Scholar16. von Elm E, Altman DG, Egger M, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. J Clin Epidemiol 2008;61(4):344–349. Crossref, Medline, Google Scholar17. Editorial Policies: Guidelines for Use of Large Language Models. Radiological Society of North America Web site. https://pubs.rsna.org/page/policies#llm. Accessed December 21, 2023. Google Scholar18. Setia MS. Methodology Series Module 5: Sampling Strategies. Indian J Dermatol 2016;61(5):505–509. Crossref, Medline, Google Scholar19. Student Learning Centre. Sampling. Flinders University. https://students.flinders.edu.au/content/dam/student/slc/sampling.pdf. Accessed December 21, 2023. Google Scholar20. Schühlen H. Pre-specified vs. post-hoc subgroup analyses: are we wiser before or after a trial has been performed? Eur Heart J 2014;35(31):2055–2057. Crossref, Medline, Google Scholar21. Ronot M. Performance of LI-RADS for the Noninvasive Diagnosis of HCC: Pathology Should Not Be the Only Acceptable Reference. Radiology 2022;303(3):546–547. Link, Google Scholar22. van der Pol CB, McInnes MDF, Salameh JP, et al. Impact of Reference Standard on CT, MRI, and Contrast-enhanced US LI-RADS Diagnosis of Hepatocellular Carcinoma: A Meta-Analysis. Radiology 2022;303(3):544–545. Link, Google Scholar23. Sadineni RT, Pasumarthy A, Bellapa NC, Velicheti S. Imaging Patterns in MRI in Recent Bone Injuries Following Negative or Inconclusive Plain Radiographs. J Clin Diagn Res 2015;9(10):TC10–TC13. Medline, Google Scholar24. Haj-Mirzaian A, Eng J, Khorasani R, et al. Use of Advanced Imaging for Radiographically Occult Hip Fracture in Elderly Patients: A Systematic Review and Meta-Analysis. Radiology 2020;296(3):521–531. Link, Google Scholar25. Wan J, Zhao Y, Feng Q, Lv P, Hong K, Zhang C. Statistical Confirmation of a Method of US Determination of Bone Age. Radiology 2021;300(1):176–183. Link, Google Scholar26. Psoter KJ, Roudsari BS, Dighe MK, Richardson ML, Katz DS, Bhargava P. Biostatistics primer for the radiologist. AJR Am J Roentgenol 2014;202(4):W365–W375. Crossref, Medline, Google Scholar27. Benchoufi M, Matzner-Lober E, Molinari N, Jannot AS, Soyer P. Interobserver agreement issues in radiology. Diagn Interv Imaging 2020;101(10):639–641. Crossref, Medline, Google Scholar28. Zhao B, Tan Y, Bell DJ, et al. Exploring intra- and inter-reader variability in uni-dimensional, bi-dimensional, and volumetric measurements of solid tumors on CT scans reconstructed at different slice intervals. Eur J Radiol 2013;82(6):959–968. Crossref, Medline, Google Scholar29. Andrade C. Sample Size and its Importance in Research. Indian J Psychol Med 2020;42(1):102–103. Crossref, Medline, Google Scholar30. Levine D, Bankier AA, Halpern EF. Submissions to Radiology: Our Top 10 List of Statistical Errors. Radiology 2009;253(2):288–290. Link, Google ScholarArticle HistoryReceived: Jan 30 2024Revision requested: Jan 31 2024Revision received: Jan 31 2024Accepted: Jan 31 2024Published online: Mar 19 2024 FiguresReferencesRelatedDetailsRecommended Articles Statistical Confirmation of a Method of US Determination of Bone AgeRadiology2021Volume: 300Issue: 1pp. 176-183Point-of-Care Bone Age Evaluation: The Increasing Role of US in Resource-limited PopulationsRadiology2020Volume: 296Issue: 1pp. 170-171Generalizability and Bias in a Deep Learning Pediatric Bone Age Prediction Model Using Hand RadiographsRadiology2022Volume: 306Issue: 2How AI May Transform Musculoskeletal ImagingRadiology2024Volume: 310Issue: 1Performance of a Deep-Learning Neural Network Model in Assessing Skeletal Maturity on Pediatric Hand RadiographsRadiology2017Volume: 287Issue: 1pp. 313-322See More RSNA Education Exhibits Working "Smarter" : A Review of Artificial Intelligence in Musculoskeletal RadiologyDigital Posters2022Checklist For The Wrist MRIDigital Posters2021Evidence Based Radiology in the age of Artificial Intelligence: The PICO/DATO ModelDigital Posters2020 RSNA Case Collection Right dorsal triquetral fracture with hook of hamate fractureRSNA Case Collection2020Pisiform DislocationRSNA Case Collection2021Kienbock's DiseaseRSNA Case Collection2021 Vol. 310, No. 3 Metrics Altmetric Score PDF download