10 • To provide the basis for conducting research and publishing results in scientific journals outlining the steps for successful accomplishment of tasks related to formulation of research question(s), search of up-to-date literature related to the topic of investigation, conduct of research, drafting of manuscripts, review process by authors, submission of manuscripts for consideration of publication, and reply to journal’s editors and reviewers’ comments. • To discuss frequently asked questions concerning authorship. The evidence found in peer-reviewed publications constitutes the foundation for advances in clinical practice. Conducting and disseminating clinically relevant research upon high ethical and editorial standards is imperative for improvement of patient care. Further, publication may enable advancement of individual career goals. Although the portfolio of the academic radiologist encompasses clinical, research, and teaching activities, the dogmatic paradigm “Publish or Perish” prevails for a successful academic career considering the fact that publication is a metric parameter of scholarly productivity in association with other parameters such as successful grants and awards.1 The assessment of an individual’s publication track record is an important part of the academic promotion and tenure process. In this chapter we discuss the steps for preparation of a manuscript as the final output of a research project from organizing tasks in a timely fashion to accruing information obtained throughout the research conduct, finalizing the manuscript draft, and submitting it for consideration of publication in a scientific journal. Previous studies have shown that the chances of professional success in a given activity substantially increase with higher levels of “preparation.”2 In addition, authors have been attributed tasks that are not urgent, but important, such as planning and preparation (writing proposals, manuscripts, and grants), studying, prevention of conflicts, and relationship building, as the key components that efficient workers apply on a day-by-day to achieve professional success.3 Before the trainee contacts potential supervisors with respect to conducting a research project he/she should be able to juggle his/her available time between clinical and educational duties during his/her internship, residency, or fellowship. This includes devotion of time where he/she can perform at his/her best avoiding pre- and post-call periods that can be stressful and exhausting. Booking academic days in the beginning of the week (Monday or Tuesday) enables trainees to catch up and finalize pending tasks such as communicating with supervisors and collaborators, submitting documents, and completing short activities during the week.4 If more than one trainee will participate in a research project, specific roles should be assigned to the trainees prior to the commencement of the study. The authorship positions of the trainees (which should be concurrent with the level of commitment and efforts of the trainees) should be openly discussed with the supervisor prior to starting the project. If the trainee or junior staff does not have a special interest for a specific topic in radiology he/she should evaluate local strengths prior to making a decision on which direction (research topic) to go. For this purpose, the trainee should obtain information about local (institutional, departmental) resources in terms of infrastructure and financial support, and collaborators with a strong track record of publications and research grants, if available. The trainee should aim to find a supervisor who has a proven track record of coming up with project ideas and successfully mentoring other trainees through the process from start to final publication. Selecting a supervisor who practices in an institution geographically close to the locale of the trainee’s rotations may facilitate (but is not exclusive) trainee–supervisor communication along the project. If the trainee does not have his/her own ideas for a research project, then he/she can inquire for ideas from his/her supervisor. Having a supervisor for the research project does not exclude the possibility that the trainee looks for further guidance from a co-supervisor who typically has expertise in an area of relevance for the research project (e.g., epidemiology, rheumatology, orthopaedic surgery) different from the area of expertise of the supervisor (e.g., musculoskeletal radiology). The trainee should investigate the feasibility of the project before committing to it. Is there a sufficient number of patients or imaging studies meeting criteria for the study? Can they be adequately accessed? Does the study have adequate power? Is there a good reference standard? Is the allotted time to complete the study appropriate? Many studies initiated by trainees and young faculty fail because they weren’t feasible to start with. A good supervisor should aid the trainee to identify what questions to ask about feasibility and to make these determinations. There is not a predetermined maximum number of investigators for a research project; however, each investigator should have a well-defined role in the project that should be determined prior to the commencement of the study. Depending on the level of difficulty of the research project, more investigators may be required such as in the case of multi-centric clinical trials where investigators from different institutions are committed to participating in different steps of the conduct of the research project. The research question should be based on the PICO (P, Patient/Problem; I, Intervention; C, Comparator/Control; O, Outcome) concept which was developed for formulation of questions and searching for evidence5 as discussed in Chapter 1. Nevertheless, prior to finalizing this step the investigators should ensure that the overarching research question(s) can add relevant information to the available bulk of knowledge in the literature. The candidate study can be complementary to prior research in a given field, but ideally should not propose research questions similar to those already proposed in previously published studies, unless there is a reason for this, such as the existence of controversy on the results of previous studies on a given topic. In this case the current study should hold potential for providing better evidence on the topic of interest (e.g., sample size with sufficient power, more rigorous methodology, multicentric characteristics, etc.). To accomplish a broad review of the existent literature on a given topic, a power literature search is recommended. This search has two objectives: (1) to determine the legitimacy of the proposed research question within the context under investigation, that is, to determine if the project is sufficiently novel; and (2) to provide a summary of the available publications in the literature that have common search term(s). It will help the investigators to frame the current project by providing a knowledge base of what is known and what requires further investigation, thus helping the authors to design a better study. The steps for a power literature search are: (1) to create a searchable question using a concept map or PICO, (2) to select an appropriate resource such as a database, (3) to run a search using database subject headings, or text words, (4) to combine results for each search term using Boolean operators, (5) to review search results, (6) to repeat steps 3 and 4 with additional terms if search results are not relevant, (7) to save search results, (8) to save search strategy (search steps), and if desired, (9) to create auto alerts. To find articles, books, or other materials on a subject, the investigator will need to develop a search query from the research question. This involves breaking down the research question into main ideas. There are different methods that can be used to help with the process of identifying the main ideas or key concepts. One method is to create a concept map. For this purpose the investigator should: (1) write the question in sentence form—it can be the “perfect article title”—something reflecting exactly what is being sought; and (2) underline the key concepts. For example: Is MRI more effective than radiography (X-rays) in diagnosing rheumatoid arthritis? Each underlined concept can be placed in its own column in a table. Within each column of the table, the investigator should add synonyms for the underlined concept, alternative spellings, and alternative word endings. In a search for literature, the terms within a column are combined using the Boolean operator “or”; the terms in separate columns are combined using the Boolean operator “and.” The PICO method is another way of breaking down questions into smaller concepts.5,6 A PICO for the search question “Is MRI more effective than radiography (X-rays) in diagnosing rheumatoid arthritis?” would be:
Conducting and Publishing Research
Learning Objectives
Introduction
Reasons for Publishing
Steps of Research Conduct
Preparation for the Conduct of a Research Project
Trainee as the Primary Author
Search for Research Project(s)/Supervisor(s)
Co-Investigators
Definition of the Research Question(s)
Conduct of a Power Literature Search
Creation of a Searchable Question Using a Concept Map or PICO
Patient/Problem: | Rheumatoid arthritis |
Intervention: | Radiography |
Comparison: | MRI |
Outcome: | Effectiveness |
Selection of a Database
Once the search concepts have been identified, the next step is to select appropriate resources within which to conduct searches. Databases are one example of a resource that can be used to identify relevant literature. They are collections of information that are controlled, organized, and accessible by computer. MEDLINE and EMBASE are two examples of databases that index journals in medicine and related areas (e.g., nursing, dentistry, basic sciences).
Use of Subject Headings, or Text Words
Many databases like MEDLINE and EMBASE have their own unique search vocabulary. This controlled vocabulary, also called subject headings, consists of standardized terms used to tag, or describe, the contents of an article. Subject headings are assigned by indexers to reflect what is discussed in the article. It is important to become familiar with, and use, the appropriate controlled vocabulary or database subject headings when conducting searches. The system of medical subject headings used in MEDLINE is called MeSH for short. The underlined terms in a concept map, or the PICO terms, can be used to help identify pertinent database subject headings.
In the cases where the search topic is a rare syndrome, new technology, new drug, new catchword, or brand name, it is very likely that there will not be a relevant database subject heading. The researcher will need to conduct a text word search. In this type of search, terms entered into the search interface are matched against word(s) that appear in specific fields such as the title and/or abstract of an article. To ensure that articles are not missed when conducting a text word search, it is important to include synonymous terms, singular and plural forms of words, as well as variations in British and American spelling. The Help section of the search interface for a database provides information about features such as truncation, wildcards, proximity or adjacency commands, and field searching that can be used to create efficient, complete text word searches.
Literature searching involves balancing specificity and sensitivity. Specific searches retrieve a higher ratio of relevant to irrelevant references. For example, searches conducted using database subject headings tend to retrieve more references that are relevant. Sensitive searches are designed to find all possible references and, as a result, retrieve many that are irrelevant. Text word searches tend to retrieve more irrelevant references; search term(s) may be present in the title or abstract, but not used in the sense that is required. An example of a specific search would be a literature review search—just key articles are needed. Systematic review searches, on the other hand, need to be sensitive. They should incorporate both database subject headings and text words to reduce the probability of missing any potential relevant records.
Once the type of search needed—specific or sensitive—has been determined, the next step is to begin searching. Rather than typing all the search terms into the database interface at once, it is preferable to look for each concept separately. For example, search for all the references about rheumatoid arthritis, then look for all the references about radiography, and finally all the references about MRI. In most database search interfaces, the results retrieved with each search concept are assigned a unique set number. These set numbers can be used to combine the different results together using the appropriate Boolean operators. When reviewing the final set of results, be alert for additional terms that may be useful. These additional terms can be searched and combined with the sets of previous terms.
Combination of Results of Search Terms, Review and Saving of Results, Repetition of Steps, Creation of Auto Alerts
All the sets created during the search process form what is referred to as the search strategy. It is worthwhile taking a few moments at the end of a search to do two things: save the search strategy and create automated updates. Saving the search strategy means saving all the steps in the search. The saved search can simply be run again at a later time rather than having to re-create the strategy from scratch. In addition, it may be necessary to provide a brief description of the search in the Methods section of the manuscript and to include the search history as an appendix. Having a document of the search strategy makes this process easier. As it is likely that there will be a long period of time between the initial literature search and the preparation of a final manuscript reporting the results of a project, it will be necessary to track new publications on the topic subsequent to the initial search. The saved search strategy can become the basis of an alert about these new publications. Results of the alerts may be emailed or saved in a user-created account. Depending on the search interface, an alert can be set to run daily, weekly, monthly, or quarterly. The name for an alert varies by interface: Auto Alerts (Ovid interface), Email Alerts (PubMed interface), or Alerts (Web of Knowledge interface).
Investigators are encouraged to use a reference management software program (e.g., Zotero and Mendeley, are freely available on the web) to store, in a library or database, information about relevant articles on the topic of study. These can be the results of the searches imported directly from the database interface into the reference management program. Many of the reference management products can link with popular word processing programs to let authors insert citations from the database or library into a document as they are writing. The software will format the citations and create a reference list or bibliography in accordance with the specific instructions of a selected journal or publisher. During the process of document revision, the citations and reference list or bibliography will be updated as changes are made.
Design, Methodology, and Sample Size Discussions with a Statistician
Once Sir Ronald Fisher (February 17, 1890 to July 29, 1962), an evolutionary biologist, geneticist, and statistician, said: “To call in the statistician after the experiment is done may be no more than asking him to perform a post-mortem examination: he may be able to say what the experiment died of.”7
Statisticians can assist researchers in designing the research question(s), selecting the study design, defining appropriate outcomes, and offering suggestions on design implementation. As part of the feasibility aspect of a research project, it is extremely important for the investigators to estimate the required sample size (Chapter 15). It is also recommended (and solicited by most institutional Research Ethics Board committees) for the investigators to prepare a database spreadsheet for the project prior to its commencement, which should be discussed with the statistician ahead of time. Examples of databases for radiology projects are available in Chapter 11.
Authorship and Contributorship Discussions among Investigators
It is recommended that discussions about authorship should be initiated by the principal investigator of the project (which can be the trainee’s supervisor) early during the conduct of the project. Several guidelines are available on the web to determine minimum standards for authorship of manuscripts.8–11 A priori agreement among the project’s investigators on the adherence to these guidelines may reduce the possibility of future disagreements on authorship among investigators at the phase of finalization of the project.
An author is considered someone who has made relevant contributions to the bulk of knowledge of the publication, and who continues to hold important academic and financial commitments with the published work.12 As per the International Committee of Medical Journal Editors (ICMJE) an author must take responsibility for at least one component of the work, should be able to identify who is responsible for each other component, and should ideally be confident in their co-authors’ ability and integrity.13 In the past, readers were rarely provided with information about contributions to studies from persons listed as authors and in acknowledgments.14 Nevertheless, some journals now request and publish information about the contributions of each author of the study, at least for original research.
Although contributorship and guarantor-ship policies remove much of the ambiguity surrounding contributions, they are unable to solve the question of the quantity and quality of contribution that qualify for authorship. Because of this, the ICJME has determined three minimum requirements for authorship credit on manuscripts submitted to medical journals since 199713: (1) substantial contributions to conception and design, acquisition of data, or analysis and interpretation of data; (2) drafting the article or revising it critically for important intellectual content; and (3) final approval of the version to be published. Authors should meet conditions 1, 2, and 3. These criteria are still appropriate for journals that distinguish authors from other contributors.
“This is a (insert language name) language translation/reprint of the ICMJE Recommendations for the Conduct, Reporting, Editing and Publication of Scholarly Work in Medical Journals. (insert name of organization) prepared this reprint/translation with support from (insert name of funding source, if any). The ICMJE has not endorsed nor approved the contents of this reprint/translation. The official version of the Recommendations for the Conduct, Reporting, Editing and Publication of Scholarly Work in Medical Journals is located at www.ICMJE.org. Users should cite this official version when citing the document.”
In multicenter studies, the group of investigators should identify the individuals who accept direct responsibility for taking different roles in the study, including drafting, reviewing, and submitting the final manuscript.15 The individuals involved in the production of the manuscript(s) should fully meet the criteria for authorship/contributorship a priori determined by the group. This information should be conveyed to the journal by the time of submission of the manuscript for completion of journal-specific author and conflict-of-interest disclosure forms. Acquisition of funding, collection of data, or general supervision of the research group alone does not constitute acknowledged criteria for authorship. Ideally discussions and decisions on authorship should be documented by the principal investigator (PI) during investigators’ meetings.
Honorary and ghost authorships represent unethical practice.16 Honorary authorship relates to the fact that an author is named but does not meet authorship criteria or has not provided any significant assistance to the study. Ghost authorship relates to the fact that an individual meets the criteria for authorship, but is not listed as an author. This typically occurs when the ghost author voluntarily excludes him/herself from authorship on request; otherwise it is considered misconduct or poor ethical conduct. A known example of misconduct ghost authorship relates to the case of Rosalind Franklin, who was a Cambridge-educated chemist who took the first X-ray diffraction picture of DNA (Photograph 51).17 She showed the Photograph 51 picture to her colleague and co-author Gosling, who without Franklin’s knowledge or permission, showed it to a scientist named Wilkins, a competitor of Franklin’s. Wilkins showed it to Watkins, who looked at it and immediately realized that DNA was helical, which led directly to his theory. He then drafted and published a paper that did not acknowledge the photograph directly. In 1953, Nature published three papers: (1) one by Watson and Crick; (2) one by Wilkins, the collaborator who had shown Photograph 51 to Watson; and (3) one by Franklin and Gosling, which described Photograph 51. Watson did not cite Franklin’s paper in his article, only acknowledged it in an endnote at the bottom.
In 1958 Franklin died of ovarian cancer at the age of 37, and was not eligible to receive the Nobel Prize posthumously. In 1962, Watson, Crick, and Wilkins received the Nobel Prize in Physiology or Medicine for their proposal on the DNA double helix structure. The fact that Franklin took the picture but could not recognize what it meant at that moment does not take the credit from her work. This case illustrates the need for proper scientific credit and recognition of collaborative work.
Documentation of the Project
The documentation of a research project has three major parts: (1) description (Materials/Patients and Methods); testing (Results); and analysis of results (Discussion) (Table 10.1). It can be obtained during and/or after the conduct of the project. Although there is no limit for the maximum number of authors who contribute for the drafting of a manuscript, it is advisable that a single senior author carefully reviews the entire document to ensure that the document has a logical sequence and coherence, and that different parts of the document connect with each other.
Prior to starting the write up of the manuscript, the authors should define the manuscript type (original research, technical development, review, editorial, letter to the editor, survey, case report, pictorial essay, etc.) as per guidelines of the scientific journal to which the manuscript will be submitted. The authors should also review the Information for Authors document of the journal (including maximum number of words format for the text, tables, and figures). In this chapter, we focus on the process for preparation of original research manuscripts. Investigators should read, understand, and adhere to the journal’s Information for Authors (Author Instructions). Lack of adherence to the journal’s instructions raises concern from the editors and reviewers for how careful the investigators have been with their project conduct.
The sections of the manuscript should be organized as shown in Table 10.2.
In the following paragraphs, we describe a suggested chronologic order for preparation of manuscript draft sections.
Materials/Patients and Methods Section
The Materials/Patients and Methods section of a manuscript allows readers to verify the methodology used in the study, which has implications, along with other factors, on the quality of the study. If the project has been well thought out in advance and has a well-written proposal, part of this section should already be available in the proposal, which is a good starting point for the first manuscript draft. Ideally the Materials/Patients and Methods section is written before the project is conducted, not after. It is advised that the first sentence of this section states whether institutional Research Ethics Board approval has been received for conduct of the study and whether informed consent has been obtained from patients or patients’ parents, if applicable, for clinical studies. Likewise, for animal studies, authors should state whether the study protocol was approved by the institutional animal care committee and whether it complied with the local Council (or equivalent organization) on Animal Care Guidelines.
The information available in the Materials/Patients and Methods section of a manuscript should be described in sufficient detail (references to methods used, including statistical methods, and rationale for using new or modified methods) to allow other investigators to reproduce the results of the study. If the reader cannot reproduce the study after reading the Materials/Patients and Methods section of a manuscript, then it is insufficient. It is recommended that authors follow the checklist of the Standards for Reporting of Diagnostic Accuracy (STARD) statement18 for studies dealing with diagnostic accuracy, the Consolidated Standards of Reporting Trials (CONSORT) statement19 for randomized controlled trials, and the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) tool20 for meta-analyses of randomized controlled trials. The authors should avoid nonstandard terms, novel or ambiguous abbreviations, and self-aggrandizement their own work (e.g., “excellent,” “innovative,” “unique”). The use of abbreviations should be minimized. If standard abbreviations are used in the text, the authors should define/spell them out in full when it is first used in the abstract, in the text, and in each figure legend. Abbreviations should be defined in each table, usually in a footer.
Table 10.2 Outline of common sections to be included in manuscript submissions
Order of files | Content |
1. Cover letter | Cover letter to the journal’s editor |
2. Title page | Abbreviated title page (includes complete title of manuscript, “short running title” if requested, all authors’ names including middle initial in the appropriate order as well as corresponding author information, affiliation, mailing and email addresses, telephone and fax numbers, and source of support) |
3. Manuscript body | Blank page (includes title and type of manuscript: original research, review, pictorial essay, etc.) Abstract (number of words), Key words Introduction Materials/Patients and Methods Results Discussion/Conclusion/Number of words (from Introduction to Conclusion) Acknowledgment References Tables |
4. Figures | Figures (no composite images; each image should be a separate figure part and represent a separate image file; for each figure, a separate tiff file with a corresponding legend should be provided; insert legends by the time the figures are uploaded onto the journal’s website). One figure may contain multiple images, which should be labeled as a,b,c,d, etc. |
5. Supplemental material (Appendix) | Online-only publication of supplemental material allows publication of material that cannot be accommodated in print. This material can include (a) multimedia (e.g., animation, dynamic image sets [movies], audio), (b) large numbers of relevant images whose number would exceed the limits of print publication, (c) relevant data in the form of tables or text that could not be accommodated in the print version, and (d) interactive materials such as Java applets and other programs for expanding browser capabilities and interactivity in areas such as image display and computer-assisted instruction |
6. *Financial disclosure | Financial disclosure may be required for each author |
7. *Conflicts of interest statement | Conflicts of interest statement may be required for each author |
8. *Transfer of copyright agreement | Transfer of copyright agreement is required for each author |
9. *Authors’ contributions | Some journals require specifications of individual contributions of authors according to the ICMJE submitted by the guarantor of the integrity of the work who can be the senior author or all authors |
Abbreviations: ICMJE, International Committee of Medical Journal Editors.
For example, for a journal that considers 3,000 as the maximum limit of words for an Original Research manuscript, e.g., Radiology, the Materials/Patients and Methods section is expected to not exceed 800 words.21
Results Section
Once the statistical analysis of the data has been performed and the preliminary or final results become available, the investigators should discuss them among themselves and with the study statistician in order to determine the face validity of the results that relates to their ability to be credible22 (whether the results make sense either clinically or experimentally). After ensuring that the results are accurate, credible, and were obtained using appropriate statistical methods, the investigators should hypothesize rationales to justify the findings. The PI can keep a log file of the investigators’ discussions during meetings. The assigned investigator of the study can then start drafting the Results section of the manuscript. The Results section should mirror the Materials/Patients and Methods section. Corresponding results should be presented in the same order as in the methods in the Materials/Patients and Methods section. There should be nothing in the Results section that was not introduced in the Materials/Patients and Methods section.
The Results section should be short, but contain at least three paragraphs comprising: (1) demographics of the population (this information could be presented in a table format instead); (2) results of the research questions; and (3) adverse events, if applicable. The results should be presented in logical sequence in the text, along with tables and illustrations. The authors should not repeat information that has already been covered in tables and/or figures. Only important information should be summarized in the text, tables, or figures. Results should be given for all items evaluated as mentioned in Materials/Patients and Methods. Numerators and denominators must be provided either in the text or in the tables for all percentages given. P values and confidence intervals should be provided wherever appropriate.
The Results section is expected to not exceed 1,000 words for a journal that considers 3,000 as the maximum limit of words for an Original Research manuscript, e.g., Radiology.21
Discussion Section
After the results have been summarized in the text of the manuscript, and before writing the discussion, the authors should update the literature search for new papers in the field of study. If the authors have created alerts for new citations at the time of the initial search then they should select and retrieve papers that may support or contradict their study results. The authors should highlight the relevant concordant and discordant information and add this information in the Discussion section of their manuscript, as appropriate, following a sequential order. Each paragraph of the Discussion section should correspond to one paragraph of the Results section. If new citations provide information similar to that generated by the author’s study the authors should try to explain (if possible) how their study results can complement the published information. The process of drafting the Discussion section can be the most challenging and time-consuming aspect of writing a manuscript. It requires a broad knowledge of the information that is available in the literature and the ability to relate the results of the study to those reported in other relevant studies raising points of similarity and disagreement. The authors should also raise hypothesizes to justify unexpected results of the study and report limitations of the study design, techniques used, and overall conduct. Setting aside sequential academic days for drafting the Discussion section can facilitate the production of a cohesive document.
The Discussion section should emphasize the advances in knowledge made by the study, which can be made in the first paragraph of the section. The authors should not repeat results in detail, but should rather include the implications of the findings and their limitations. On the final paragraphs of this section the authors should address study limitations. A good discussion of limitations shows that the authors have the ability to critically assess their own work. If the limitations are written well, they will adequately anticipate and address concerns that reviewers identify with the study. Limitations should be written within a paragraph (or paragraphs) at the end of the discussion, just prior to the conclusion.
Table 10.3 Suggestions for drafting the discussion section of a manuscript
Discussion subsections | Focus |
Initial paragraph | State the main findings of the study, which were raised as objectives in the Introduction section. |
Body paragraphs | Establish the novelty or uniqueness of the results of the paper. Support your study results by comparison with results published by other groups of investigators. Cite other studies’ results that support or are discordant in relation to your study results explaining how they fit or conflict. Defend your results generating hypotheses for why the answers of your study are different from the answers from previous published studies. Discuss unexpected results that do not support the hypotheses of your study. Discuss the strengths, weaknesses, and limitations of your study, indicating how design limitations of your study may have led to over- or underestimation of results’ values. |
Final paragraph | Restate the answer to the primary research question and indicate the significance of the study results with regard to clinical applications, implications for the health care system, and recommendations. Suggest future needs for investigation and unanswered questions. End with a strong, clear, evidence-supported conclusion (“Summary Statement”). |
The Discussion section should conclude by linking the conclusions with the goals of the study, avoiding unqualified statements and conclusions not supported by the data. For articles on experimental studies, the authors should describe the importance of the conclusions as they relate to potential future practical applications. Details of the Discussion section are available in Table 10.3.23
For example, for a journal that considers 3,000 as the maximum limit of words for an Original Research manuscript, e.g., Radiology,21 the Discussion section is expected to not exceed 800 words.
Upon completion of the first draft of Materials/Patients and Methods, Results, and Discussion, the authors should re-read all the document, focus on refining thoughts, correct typos, and start drafting the Introduction and Abstract sections.
Introduction Section
Only strictly pertinent background information that informs the reader as to why the study was performed should be included in the Introduction section. Brevity and focus are key. No detailed review of the topic is expected in this section. What is said in the Introduction section does not need to be repeated in the Discussion section. An effective Introduction has three segments: a section that provides a scoping statement (why the authors decided to investigate the proposed research question[s]), a section that focuses somewhat on the research area (what is currently available in terms of technology, processes, and procedures to address the question[s]), and a section that focuses on the specific area of research for the current study (how the proposed method or strategy can potentially be superior to the methods or strategies currently available).24 In the final paragraph of this section the authors should state the objectives/hypotheses of the study. For example, for an Original Research manuscript that allows a maximum of 3,000 words, the information in this section is expected to not exceed 400 words.21 Details of the Introduction section are available in Table 10.4.
Table 10.4 Suggestions for drafting the Introduction section of a manuscript
Introduction subsections | Focus |
Initial paragraph | Define the disease or problem the authors decided to evaluate and its implications on the morbidity and mortality for patients. |
Body paragraphs | Address the problem or gap in the literature and state why the authors have decided to conduct this research study and why it is so important to fill in these gaps. Comment on the diagnostic approaches that are currently available to fill in the literature gaps, how did the authors plan to do so in the study, and what imaging technology they have used in their research study. If the technology is novel, they should briefly explain what the technique is about, stating advantages over conventional techniques. Link the problem with the technique that is available to diagnose the problem (disease). In this paragraph, the authors can briefly report results of other studies in the field that provided incomplete information on the technique–problem linkage and how the current study will be complementary to the questions raised. Authors should also state whether no previous study has addressed the proposed research questions of their study, which is indeed needed for this study to be relevant. |
Final paragraph | Clearly state the hypothesis and purpose of the study in a fashion similar to the Purpose statement in the abstract. |
Abstract Section
Once the four sections of the manuscript (Materials/Patients and Methods, Results, Discussion, and Introduction) are completed, the author should re-read them and draft the Abstract section. Therefore, it is written last, not first. For Reviews, Original Articles, or other similar submissions, either structured or unstructured abstracts (typically 100–300 words depending on the journal) should summarize the content of the submission, and specific headings (Objectives, Methods, Results, and Conclusion) should or should not be included as per guidelines of the scientific journal (Table 10.5). Authors should include key words (typically an average of six; no less than four and no more than eight is a good guideline) for the study on a subsection of the abstract. Key words should be concise and authors should avoid abbreviations. For radiology studies, at a minimum, investigators should include keywords identifying the patient age group, the imaging modality studied, the organ system, and the disease process.
References
Reference formatting is not the same for all journals. In addition to reading instructions, authors may look at articles from the same journal as examples for proper formatting. If the authors have used a reference management software program during the writing of the manuscript, then the software will number the references as they are inserted in the manuscript and format the references according to the style requirements of a particular journal or publisher. Otherwise, authors should manually number references consecutively in the order in which they are first mentioned in the manuscript for most journals, although a minority of journals may use alphabetical order. Articles that appear in online journals should follow the same citation format as print articles, with the addition of the URL and the date the article was accessed. In the case of books, the authors of a chapter, title of the chapter, editor(s), title of the book, edition, city and state, publisher, year, and specific pages must be provided. For web content, author(s) (if any); title of the page or content; name or owner of the website; URL; and publication, update, and access dates should be listed. The authors should search for types of references in the online Publication Information for Authors of the proposed journal. Most journals state that it is the responsibility of the authors to verify the accuracy of their study references to ensure proper linking of referenced articles in the online journal.
Tables
Tables and figures can be prepared during drafting of other sections or at the end of the writing process. This process requires accuracy in that information of tables matches with that reported in the text of Results. Tables should be numbered in Arabic numerals and should have a title. Each table should have sufficient information that allows it to stand by itself. All abbreviations used in the table should be explained in a footnote. Abbreviations already explained in the manuscript text should be repeated in the labels or footnotes of tables and legends of figures. The authors should provide the unit of measurement for each variable, report counts along with percentages, specify the type of measure of central tendency (mean, median, etc.), and measure of dispersion (e.g., standard deviation, standard error, interquartile range). For results of comparisons, authors should specify the summary measure (mean difference, relative risk, odds ratio, etc.) and provide a measure of precision around the estimate (e.g., 95% confidence intervals). Some journals accept supplemental material, which can be uploaded as an Appendix on the journal’s website.
Illustrations
Figures provide visual aids in the form of drawings, surgical photographs, and imaging captions to facilitate the reader understanding on the research or teaching point. They should appear sequentially in the text supporting information reported in Results, and less likely, in Discussion or Materials/Patients and Methods. A legend must be supplied for each illustration, including drawings and graphs, and should not duplicate text or table material. It is essential that the legend describes all labels and annotations placed on an illustration. Judicious use of annotations is recommended, enough to assure reader understanding, but not too many as to be distracting. Use of simple annotations (black and/or white, not colored) is suggested. Legends describing radiologic images should include the type of image, its imaging projection or plane, use of contrast material, pulse sequence information (repetition, echo, and inversion times; flip angle) for magnetic resonance images, as appropriate, and the features to be observed by the reader highlighted by arrows and other indicators (Fig. 10.1). Preferably, patient age/gender is indicated in the legend. Regarding graphs, the x-axis and y-axis are labelled including units, and all components of the graph should be adequately explained in the legend. If flow diagrams of research designs or protocols are used, they should indicate the events in a chronologic fashion, with the earlier events being displayed on the left-hand side of the figure and the later events on the right-hand side (Fig. 10.2). Concerning photo-micrographs and pathology slides, the authors should include the stain and original magnification. Figures should be provided in high resolution according to the journal’s instructions. Most journals require digital images of at least 300 dpi (dots per inch; 1200 dpi for line art). Specifications regarding image preparation can be found in the online Publication Information for Authors of journals. Color figures may require the payment of a fee from the authors for paper copy publication.
