Diagnostics Submission Guide: MDPI Process (2026)

Source: Diagnostics Instructions for Authors and MDPI research and publication ethics policy (accessed June 2026)

Diagnostics is not a stronger version of a subscription radiology or laboratory-medicine journal, and it is not a weaker one. It is a different model. MDPI built it around speed and soundness-based review: the editorial question is whether the work is methodologically sound, in scope, and correctly reported, not whether it ranks among the most selective diagnostic findings of the year. That model shapes everything about how you should prepare the package.

It helps to anchor the soundness-versus-selectivity distinction against journals authors already know. At a flagship clinical title like The Lancet, JAMA, or NEJM, the editorial question is whether the result should change practice, and most submissions are desk-rejected for novelty before reporting is ever assessed; the bar at Nature-tier journals is similar but for biological reach. Diagnostics inverts the order.

A correct, complete, in-scope diagnostic study clears the bar even when the finding is incremental, so the leverage is not in pitching importance the way you would for The Lancet or JAMA. It is in proving the diagnostic claim is reported and validated to the standard the field expects.

Three consequences matter most.

• Diagnostics is strongly section-based, so scope fit is assessed against a named diagnostic discipline rather than a vague "is this interesting" bar.
• The pre-check is fast and partly template-driven, so completeness is rewarded and incompleteness is punished early.
• Reporting guidelines are load-bearing: a paper that claims diagnostic accuracy but does not report it the STARD way reads as not ready, even when the underlying data are good.

A technically interesting model with a missing STARD checklist, or a single-center retrospective study claiming clinical deployability with no external validation, can be returned or sent straight to major revision before it gains any momentum. A competent, complete, in-scope study with metrics reported correctly moves quickly.

The core fit for most submissions is the original research article. It works best when the diagnostic or prognostic question is central, the reference standard is explicit, the methods are reproducible from the text, and the declarations and reporting package are complete on first upload.

Ask these questions before you submit:

• is a diagnostic or prognostic question the actual subject of the paper, or is diagnosis a downstream application of a basic-science finding?
• can a reader reproduce the analysis from the manuscript and supplementary files alone, including the reference standard?
• are the accuracy metrics reported with confidence intervals, or as bare point estimates?
• does the reporting follow the checklist that matches the study design?

If the answers are uncertain, the pre-check problem is usually more important than the science problem.

The pre-check editor is answering a short list of questions fast.

On scope, the editor asks whether the manuscript belongs in a diagnostics journal and in which section. If the diagnostic relevance is thin or bolted on, the paper is redirected or returned. On soundness, the question is whether the design supports the diagnostic claim: is there a reference standard, is the spectrum of patients representative, are the metrics appropriate for the question. Diagnostics does not require the finding to be field-defining, but it does require the work to be done and reported correctly.

On integrity, the editor checks whether ethics approvals, consent, image-integrity expectations, and data availability are in order. MDPI runs integrity and plagiarism checks at pre-check, and gaps here trigger fast returns. On completeness, the editor looks for the declarations block and the design-appropriate reporting checklist. A diagnostic-accuracy study with no STARD checklist, or a prediction model with no TRIPOD statement, reads as not ready, even when the science is fine.

Manuscript structure: Diagnostics expects a defined section set: Abstract, Keywords, Introduction, Materials and Methods, Results, Discussion, Conclusions, plus the declarations block. Original research and systematic reviews need a structured abstract of around 200 to 250 words. The abstract is the first thing the pre-check editor reads, so the diagnostic question, the reference standard, and the headline accuracy result (with its confidence interval) all need to be visible there.

Reporting and methods readiness: Provide full methodological detail so results can be reproduced, and follow the design-appropriate guideline. STARD governs diagnostic-accuracy studies; TRIPOD (and TRIPOD-AI for machine-learning models) governs prediction-model development and validation; PRISMA governs systematic reviews and meta-analyses (with a registered protocol); CARE governs case reports. A diagnostic or prediction-model paper that does not map cleanly onto its reporting checklist is the most common reviewer-stage friction point.

Declarations and ethics: Draft the Institutional Review Board statement, Informed Consent statement, Author Contributions (by initials), Funding, Data Availability, and Conflicts of Interest sections before you upload. These are not post-acceptance paperwork at MDPI; they are pre-check gates. For retrospective imaging or laboratory studies, the consent or consent-waiver language has to be explicit.

Figures, supplementary, and abstract assets: A graphical abstract is optional but commonly used; if supplied, it should be a high-resolution PNG, JPEG, or TIFF. Supplementary materials should carry the completed reporting checklist, the full confusion matrix or calibration data, and detail that would slow the main narrative. ORCID is expected for the submitting author, and the system will ask for suggested reviewers in the relevant diagnostic subspecialty.

In our pre-submission review work with Diagnostics manuscripts, four failure patterns generate the most consistent pre-check returns and reviewer friction, and each one is testable against your own manuscript before you upload. Across the diagnostics and imaging manuscripts we pre-screen, the single most common reason a technically sound paper draws a major-revision verdict is a model or accuracy claim that the validation evidence does not support, not a flaw in the underlying science.

Across our diagnostics and clinical pre-submission reviews, the pattern that surprises authors most is that the Diagnostics pre-check is not a selectivity filter in the Clinical Chemistry sense; it is a fit-and-completeness filter where the reporting guideline does most of the work. The manuscripts that get returned fastest are rarely bad science. They are competent studies whose section angle, declarations block, accuracy reporting, or validation strategy is not ready for a fast, template-driven screen. Manuscripts coming through pre-submission review for Diagnostics split along these four lines.

Scope-thin diagnostic framing that the section editor cannot place

The single most common pattern we see is a manuscript whose diagnostic relevance is downstream rather than central. The study is really a cell-biology, bioinformatics, or methods paper, and a diagnostic angle (a biomarker panel, a classifier, a TCGA-style dataset) has been added so the work can target a diagnostics journal.

Diagnostics is strongly section-based, so the pre-check editor has to place the manuscript in a specific section. When the diagnostic question is not the actual subject, the section assignment fails and the paper is returned or redirected fast.

The testable version: read your own abstract and introduction, and ask whether a section editor could name the Diagnostics section from the first paragraph alone. If the diagnostic application only appears in the discussion, rebuild the introduction and abstract around the diagnostic question and the reference standard rather than around the underlying biology or method.

Check whether your Diagnostics scope angle reads as diagnostic from the abstract →

A diagnostic-AI model with no external validation, framed as clinically deployable

The second pattern is specific to this journal because so much of its volume is machine learning and imaging. We repeatedly see a model trained and tested on a single-center retrospective dataset, evaluated only with internal cross-validation, and then framed in the discussion as ready for clinical deployment.

Reviewers in this space now expect external or temporal-geographic validation. Its absence is the common reason a methodologically clean model gets major revision instead of acceptance.

The testable version: if your paper claims clinical utility, confirm the model was validated on data it never saw during training and tuning, ideally from a different site, scanner, or assay. If the only evidence is internal cross-validation on one cohort, frame the manuscript as a development study and map it against STARD-AI or TRIPOD-AI before submission.

Check whether your Diagnostics model claims match its validation evidence →

Accuracy metrics reported as bare point estimates without a reference standard

The third pattern is diagnostic-accuracy reporting that does not meet the bar STARD sets. We see sensitivity and specificity reported as single numbers with no confidence intervals, AUC values quoted without a comparison or a defined positive class, and no clear definition of the reference standard against which accuracy is measured.

In diagnostics, the reference standard is the whole argument: an AUC of 0.95 means nothing if the ground truth is itself uncertain or circular.

The testable version: for every accuracy number in your abstract and results, confirm there is a confidence interval, the reference standard is named and justified, and the patient spectrum is described well enough for a reader to judge generalizability.

Check whether your Diagnostics accuracy metrics meet the STARD bar →

Reporting that does not map onto its guideline checklist

The fourth pattern shows up at the reviewer stage: reporting that does not follow the checklist matching the study design. Examples include a diagnostic-accuracy study with no STARD flow diagram, a prediction model with no TRIPOD statement, a systematic review with no PRISMA diagram and no registered protocol, or a case report that ignores CARE.

Each one forces reviewers to spend their attention on missing structure rather than on the science. In diagnostics, clinical relevance depends on how the reference standard, sample size, and statistical analyses are reported.

The testable version: identify the guideline that matches your design, walk your manuscript against every checklist item, and supply the completed checklist and relevant diagram in the supplementary files. If half your checklist items point to "see Methods" without the Methods actually covering them, the reporting is not ready.

Check whether your Diagnostics reporting matches its guideline checklist →

Each of these is something you can check against your own draft before you commit the submission. This guide tells you what Diagnostics editors look for; the review tells you whether YOUR paper passes the pre-check before you upload.

Of 60+ manuscripts our team reviewed for diagnostics, laboratory-medicine, and imaging journals, including Diagnostics and its open-access peers, the recurring pre-check weakness was a diagnostic claim whose section fit, validation evidence, or reporting checklist did not match the abstract. Paid Manusights reviews include a 60-day money-back guarantee, and we do not train models on submitted manuscripts. Run a [Diagnostics submission package check](/ai-review?

target_journal=Diagnostics&source_blog=diagnostics-submission-guide&primary_concern=submission_readiness) to see whether your scope framing, validation strategy, accuracy reporting, and declarations block will clear the MDPI pre-check.

Diagnostics reports a median first decision near 22 days and median acceptance-to-publication near 2.7 days. Treat these as planning ranges, not promises: clinical imaging and large diagnostic-accuracy manuscripts often run longer because finding reviewers with the right subspecialty takes time.

• Day 0: Submission via SuSy. The portal accepts the package and routes it to the section editor for pre-check.

• Days 1 to 3: Editorial pre-check. The editor screens section-scope fit, ethics completeness, integrity and plagiarism checks, reporting-checklist presence, and basic soundness.

The fastest returns happen here, before any reviewer is invited.

• Days 3 to 7: Reviewer invitation. Manuscripts that pass pre-check enter single-blind reviewer search, typically targeting two or more reviewers in the relevant diagnostic subspecialty.

• Days 7 to 22: Peer review and first decision. Reviewer reports return and the editor issues the first decision, with a median near 22 days from submission.

Major revision is the most common outcome for papers that clear pre-check.

• Days 22 to 40: Revision and acceptance. Revisions are usually requested on a short clock; resubmission and a second review cycle commonly land acceptance inside a few weeks for in-scope, complete packages.

• Days 40 to 43: Production and publication. Acceptance to publication runs near 2.7 days at median, so the slow part of the calendar is reviewer search and revision, not production.

This guide was researched and built from primary sources: the sources we checked include the Diagnostics Instructions for Authors, the journal's aims-and-scope, sections, and editorial-process pages, MDPI's research and publication ethics policy, the STARD and TRIPOD reporting-guideline statements, and Manusights pre-submission review patterns from diagnostics and imaging manuscripts deciding between Diagnostics and peer open-access and selective diagnostics journals. We reviewed and compared current MDPI author guidance with recent Manusights work reviews from authors weighing Diagnostics, Journal of Clinical Medicine, Clinical Chemistry, and European Radiology. Last reviewed by the Manusights clinical editorial team on 2026-06-07.

Source limitations: MDPI can update the APC, article-format details, abstract caps, and editorial-process numbers after this review date, so verify final administrative details against the official Diagnostics author pages before upload. Median timelines are reported by the journal and vary by subspecialty. Use this guide for the decision the official instructions cannot answer: whether your scope framing, validation strategy, accuracy reporting, and declarations block are ready for the MDPI pre-check.

• Pre-submission review for radiology and imaging research
• Diagnostics journal profile
• IEEE Transactions on Medical Imaging submission guide
• Rejected from IEEE Transactions on Medical Imaging, where next?
• Best medical education journals

Before you upload, run your manuscript through a Diagnostics submission readiness check to catch the scope, validation, accuracy-reporting, and ethics gaps the MDPI pre-check filters for. The check is free to run (/ai-review) and takes a single upload.