The DOI matches its Crossref record; 26 of the 26 references it cites were found in a registry or an index; 3 of the 17 sources its claims cite were checked for support.
MINIMAR (MINimum Information for Medical AI Reporting): Developing reporting standards for artificial intelligence in health care.
Open accessAbstract
The rise of digital data and computing power have contributed to significant advancements in artificial intelligence (AI), leading to the use of classification and prediction models in health care to enhance clinical decision-making for diagnosis, treatment and prognosis. However, such advances are limited by the lack of reporting standards for the data used to develop those models, the model architecture, and the model evaluation and validation processes. Here, we present MINIMAR (MINimum Information for Medical AI Reporting), a proposal describing the minimum information necessary to understand intended predictions, target populations, and hidden biases, and the ability to generalize these emerging technologies. We call for a standard to accurately and responsibly report on AI in health care. This will facilitate the design and implementation of these models and promote the development and use of associated clinical decision support tools, as well as manage concerns regarding accuracy and bias.
Overview
CoChat Summary
TL;DR
MINIMAR is a proposal for minimum reporting standards for medical AI, covering study population and setting, patient demographics, model architecture, and model evaluation. It argues that transparent reporting of training data, model development, and validation is needed to assess bias, reproducibility, and generalizability in health care.
Key findings
- Proposes MINIMAR around 4 essential components: (1) study population and setting, (2) patient demographic characteristics, (3) model architecture, and (4) model evaluation.
- Argues that lack of reporting standards limits understanding of intended predictions, target populations, hidden biases, and the ability to generalize AI models in health care.
- Cites an empirical evaluation of 81 studies comparing AI models against clinicians that showed major problems with lack of transparency, bias, and unjustified claims, likely because key details were often missing.
- Calls for a standard to accurately and responsibly report AI in health care, including transparency on training data, model design/development, evaluation, validation, and code/data sharing.
Methods
- Design: a proposal/commentary, not an empirical study; MINIMAR is presented as a starting point for broader community discussion.
- Data or sample: no primary dataset; the paper draws on literature and uses two cited example studies in Table 1 to illustrate proposed reporting fields.
- Main analysis: narrative rationale and a proposed checklist/table of 4 reporting components with features, descriptions, and examples.
Limitations
- The paper states MINIMAR is a starting point for broader community discussion and hopes it will stimulate discussion and improvement; no formal validation or consensus process is reported.
- Evident from the text: MINIMAR is not empirically tested for usability, reliability, or impact, and the illustrative examples are limited to two cited studies.
- Evident from the text: no formal Delphi process, stakeholder validation, or adoption study is described.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
3 Valid · 0 Not valid · 10 Not checked · 4 Own results and interpretations
Every source the paper cites for this claim was checked and supports it.
In the paper
“Therefore, the data used for training and the definition of the task—be it classification or prediction—become more important than the specifics of the machine learning algorithm.7”
§ MINIMAR (MINimum Information for Medical AI Reporting): Developing reporting standards for artificial intelligence in health care > INTRODUCTION
What the sources say
- Supportedcites [7] Obermeyer, 2019, Dissecting racial bias in an algorithm used to manage the health of populations, Science source
The Discussion (p. 8) states the comparison the citing sentence attributes directly: 'Changing the procedures by which we fit algorithms (for instance, by using a new statistical technique for decorrelating predictors with race...) is not required. Rather, we must change the data we feed the algorithm - specifically, the labels we give it,' and 'labels are the key determinant of both predictive quality and predictive bias.' The Problem formulation section (p. 6) calls label choice 'perhaps the single most important decision made in the development of a prediction algorithm,' and Experiments on label choice (p. 6) shows algorithms held constant in training method whose labels differ produce nearly twofold variation in the Black share of the highest-risk group. Our copy frames 'definition of the task' as label choice / 'problem formulation' (turning an amorphous concept into a concrete predictable variable) rather than a classification-versus-prediction distinction, but the substance of the attributed claim - data and task definition outweighing algorithmic specifics - is stated explicitly. (LLM, conf=0.82)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“It is clear that the performance of any AI model broadly depends on its reliability and its ability to generalize to the setting and population in which it is applied, rather than its performance represented by the training and test data alone.10”
§ MINIMAR (MINimum Information for Medical AI Reporting): Developing reporting standards for artificial intelligence in health care > INTRODUCTION
What the sources say
- Partly supportedcites [10] Riley, 2016, External validation of clinical prediction models using big datasets from e-health records or IPD meta-analysis: opportunities and challenges, BMJ sourceExperimental — accuracy not yet measured
The source states the substance of the claim for clinical prediction models: the Summary points section defines external validation as using 'new participant level data, external to those used for model development, to examine whether the model's predictions are reliable... in individuals from potential population(s) for clinical use'; 'Current shortcomings of external validation studies' says predictive performance 'tends to vary across settings, populations and periods' so that multiple external validations are needed 'to fully appreciate the generalisability of a prediction model'; and the Conclusions state that reporting only overall (averaged) performance 'is not sufficient' and that users need to know whether a model is 'reliable or transportable to all the settings, populations, and groups'. What the source does not support is the claim's scope: our copy frames this entirely around statistical multivariable regression risk models (Framingham, QRISK2, Nottingham prognostic index, Wells score) and never mentions artificial intelligence or machine-learning models, and its contrast is with development-data/overall averaged performance rather than with a modern AI 'training and test split'. So the generalisability-and-reliability principle is confirmed, but the extension to 'any AI model' is not this paper's. (LLM, conf=0.72)
boostedstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“A lack of transparency regarding the training data used for model development directly affects the reproducibility, generalizability, and interpretability of a proposed model.”
§ MINIMAR (MINimum Information for Medical AI Reporting): Developing reporting standards for artificial intelligence in health care > INTRODUCTION
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
In the paper
“However, the characteristics of the data necessary to assess how these predictive models perform are not being adequately reported in the literature,11 leaving uncertainty and doubt about the application in the broader healthcare setting.”
§ MINIMAR (MINimum Information for Medical AI Reporting): Developing reporting standards for artificial intelligence in health care > INTRODUCTION
What the sources say
- Not checked yetcites [11] Bozkurt, 2020, Reporting of demographic data, representativeness and transparency in machine learning models using electronic health records, JAMA Netw Open
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Detailed knowledge of the data used to train the model (ie, the training data) and the population those data represent—or often, does not represent—is essential to understanding the validity and generalizability of the “AI solution.””
§ MINIMAR (MINimum Information for Medical AI Reporting): Developing reporting standards for artificial intelligence in health care > INTRODUCTION
plainstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: unclear — It is unclear whether it could be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
5 more claims
Create a free account to see all 17 claims and to verify the 14 references still open.
- It may state two separate claims as one. 7
- Probably not a substantial claim. 3
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Standardized reporting checklists such as the Prediction Model Risk of Bias Assessment (PROBAST) tool have been developed to assess the risk of imbalanced data biases, which allows both developers and downstream users (e.g., physicians, hospitals) to better understand the suitability and limitations of AI-based systems in specific clinical settings [54–56].”
From Bias in medical AI: Implications for clinical decision-making (2024) · cites this paper as [55]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“One significant gap is the absence of guidance on how to report post-deployment model evaluation, particularly in terms of their impact on equitable health outcomes.”
From Advancing Health Equity through Multi-Level Fairness in Health Informatics (2026) · cites this paper as [12]
Experimental — accuracy not yet measuredIn the citing paper
“This includes MINIMAR (MINimum Information for Medical AI Reporting), which defines a minimal set of reporting elements covering study population and setting, patient demographics, model architecture, and model evaluation (198)”
From Artificial Intelligence in Oncology: Current Landscape, Challenges, and Future Directions (2024) · cites this paper as [198]
Citations
Cites
Preferred Reporting Items for Systematic Reviews and Meta-Analyses: the PRISMA statement
Moher, D., Liberati, A., Tetzlaff, J., Altman, D. G., & for the PRISMA Group (2009) · BMJ · doi:10.1136/bmj.b2535 · Cited by 83,823
von Elm, E., Altman, D. G., Egger, M., Pocock, S. J., Gøtzsche, P. C., Vandenbroucke, J. P., & Initiative, F. T. S. (2007) · PLoS Medicine · doi:10.1371/journal.pmed.0040296 · Cited by 23,020
von Elm, E., Altman, D. G., Egger, M., Pocock, S. J., Gøtzsche, P. C., & Vandenbroucke, J. P. (2007) · The Lancet · doi:10.1016/s0140-6736(07)61602-x · Cited by 21,320
Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement
Moher, D., Liberati, A., Tetzlaff, J., & Altman, D. G. (2010) · International Journal of Surgery · doi:10.1016/j.ijsu.2010.02.007 · Cited by 12,156
SPIRIT 2013 Statement: Defining Standard Protocol Items for Clinical Trials
Chan, A.-W., Tetzlaff, J. M., Altman, D. G., Laupacis, A., Gøtzsche, P. C., Krleža-Jerić, K., Hróbjartsson, A., Mann, H., Dickersin, K., Berlin, J. A., et al. (2013) · Annals of Internal Medicine · doi:10.7326/0003-4819-158-3-201302050-00583 · Cited by 7,363
Dissecting racial bias in an algorithm used to manage the health of populations
Obermeyer, Z., Powers, B., Vogeli, C., & Mullainathan, S. (2019) · Science · doi:10.1126/science.aax2342 · Cited by 7,137 · cited 2×
“Therefore, the data used for training and the definition of the task—be it classification or prediction—become more important than the specifics of the machine learning algorithm.(7) Detailed knowledge of the data used to train the model (ie, the training data) and the population those data represent—or often, does no…”
BackgroundMethodSPIRIT 2013 Statement: defining standard protocol items for clinical trials.
Chan, A., Tetzlaff, J., Altman, D. G., Laupacis, A., Gøtzsche, P. C., A-Jerić, K. K., Hrobjartsson, A., Mann, H., Dickersin, K., Berlin, J. A., et al. (2015) · PubMed · Cited by 4,897
Minimum information about a microarray experiment (MIAME)-toward standards for microarray data
Brazma, A., Hingamp, P., Quackenbush, J., Sherlock, G., Spellman, P., Stoeckert, C., Aach, J., Ansorge, W., Ball, C. A., Causton, H. C., et al. (2001) · Nature Genetics · doi:10.1038/ng1201-365 · Cited by 4,109
Hannun, A. Y., Rajpurkar, P., Haghpanahi, M., Tison, G. H., Bourn, C., Turakhia, M. P., & Ng, A. Y. (2019) · Nature Medicine · doi:10.1038/s41591-018-0268-3 · Cited by 3,148
Improving the Quality of Reporting of Randomized Controlled Trials: The CONSORT Statement
Begg, C. (1996) · JAMA · doi:10.1001/jama.1996.03540080059030 · Cited by 3,127
Research works printed but not linked: 1
- Villarosa, 2018, Why America’s black mothers and babies are in a life-or-death crisis, The New York Times Magazine, 11
Cited by
Collins, G. S., Moons, K. G. M., Dhiman, P., Riley, R. D., Beam, A. L., Van Calster, B., Ghassemi, M., Liu, X., Reitsma, J. B., van Smeden, M., et al. (2024) · BMJ · doi:10.1136/bmj-2023-078378 · Cited by 4,718
Díaz-Rodríguez, N., Del Ser, J., Coeckelbergh, M., López de Prado, M., Herrera-Viedma, E., & Herrera, F. (2023) · Information Fusion · doi:10.1016/j.inffus.2023.101896 · Cited by 800
Bias in medical AI: Implications for clinical decision-making
Cross, J. L., Choma, M. A., & Onofrey, J. A. (2024) · PLOS Digital Health · doi:10.1371/journal.pdig.0000651 · Cited by 607
Ethical and Bias Considerations in Artificial Intelligence/Machine Learning
Hanna, M. G., Pantanowitz, L., Jackson, B., Palmer, O., Visweswaran, S., Pantanowitz, J., Deebajah, M., & Rashidi, H. H. (2025) · Modern Pathology · doi:10.1016/j.modpat.2024.100686 · Cited by 535
Sharma, A., Lin, I. W., Miner, A. S., Atkins, D. C., & Althoff, T. (2023) · Nature Machine Intelligence · doi:10.1038/s42256-022-00593-2 · Cited by 500
Metadata
- Authors
- Hernandez-Boussard, T., Bozkurt, S., Ioannidis, J. P. A., & Shah, N. H.
- Affiliations
- Department of Medicine, Stanford University, Stanford, California, USA
- Stanford Medicine (US)
- Venue
- Journal of the American Medical Informatics Association
- Publisher
- Oxford University Press (OUP)
- Published
- 2020-06-28
- Volume
- 27
- Issue
- 12
- Pages
- 2011-2015
- ISSN
- 1067-5027 (ISSN-L)
- 1527-974X
- PMC
- PMC7727333
- OpenAlex
- W3037405958
- Cited by
- 410
- References
- 26
- Citations by year
- 2026: 94 · 2025: 77 · 2024: 80 · 2023: 65 · 2022: 51 · 2021: 38 · 2020: 5
- FWCI
- 11.77
- Type
- journal article
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- Locations
- Journal of the American Medical Informatics Association (publishedVersion, cc-by-nc-nd)
- PubMed Central (publishedVersion, cc-by-nc-nd)
- PubMed (publishedVersion)
- Funders
- National Cancer Institute (R01CA183962)
- Topics
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Physical Sciences › Computer Science › Artificial Intelligence › Ethics and Social Impacts of AI
- Physical Sciences › Computer Science › Artificial Intelligence › Explainable Artificial Intelligence (XAI)
- Keywords
- reporting guidelines, artificial intelligence in healthcare, clinical decision support, model transparency, algorithmic bias, generalizability, clinical prediction model
- MeSH terms
- Artificial Intelligence, Delivery of Health Care, Humans, Electronic Health Records, Research Report
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-05, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-05 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (qwen3.8-omni-flash).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: MINIMAR (MINimum Information for Medical AI Reporting): Developing reporting standards for artificial intelligence in health care. https://app.cochat.ai/papers/10.1093/jamia/ocaa088
- The paper
- Hernandez-Boussard, T., Bozkurt, S., Ioannidis, J. P. A., & Shah, N. H. (2020). MINIMAR (MINimum Information for Medical AI Reporting): Developing reporting standards for artificial intelligence in health care. Journal of the American Medical Informatics Association. https://doi.org/10.1093/jamia/ocaa088
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.