The DOI matches its Crossref record; 138 of the 140 references it cites were found in a registry or an index; 21 of the 100 sources its claims cite were checked for support.
Sex and gender differences and biases in artificial intelligence for biomedicine and healthcare
Open accessAbstract
Precision Medicine implies a deep understanding of inter-individual differences in health and disease that are due to genetic and environmental factors. To acquire such understanding there is a need for the implementation of different types of technologies based on artificial intelligence (AI) that enable the identification of biomedically relevant patterns, facilitating progress towards individually tailored preventative and therapeutic interventions. Despite the significant scientific advances achieved so far, most of the currently used biomedical AI technologies do not account for bias detection. Furthermore, the design of the majority of algorithms ignore the sex and gender dimension and its contribution to health and disease differences among individuals. Failure in accounting for these differences will generate sub-optimal results and produce mistakes as well as discriminatory outcomes. In this review we examine the current sex and gender gaps in a subset of biomedical technologies used in relation to Precision Medicine. In addition, we provide recommendations to optimize their utilization to improve the global health and disease landscape and decrease inequalities.
Overview
CoChat Summary
TL;DR
This narrative review examines how sex and gender differences are neglected in biomedical AI and how resulting biases can harm Precision Medicine. It proposes distinguishing "desirable" from "undesirable" biases and offers recommendations to embed sex and gender considerations across data generation, AI technologies, and fairness frameworks.
Key findings
- The review defines two bias categories: *desirable* biases (deliberately accounting for sex/gender differences to improve diagnosis and treatment) and *undesirable* biases (unintended or unnecessary discrimination), and argues AI can either magnify or mitigate sex and gender inequalities depending on design.
- Sex and gender gaps persist across data sources: an analysis of 234 phenotypic traits from almost 55,000 mice showed findings influenced by sex; women are underrepresented in clinical trials (especially Phases I and II); and 67% of patients enrolled in cardiovascular device trials are male.
- Digital biomarker datasets show representation problems — e.g., a Parkinson's disease digital biomarker study had only 18.6% women — and access gaps exist: in low- and middle-income countries women are 10% less likely to own smartphones and 26% less likely to use the internet, with 1.2 billion women lacking mobile internet access.
- Bias also affects clinical practice and algorithms: only 17% of cardiologists correctly identified women as having greater heart disease risk than men, and a commercial algorithm using healthcare cost as a proxy for health status exhibited significant racial discrimination.
Methods
- Design: narrative review examining sex and gender gaps in a subset of biomedical technologies relevant to Precision Medicine (Big Data Analytics, Natural Language Processing, Robotics, Explainable AI, and algorithmic Fairness).
- Data/sample: synthesis of prior literature, illustrative examples, and referenced studies; no primary data collection or systematic search protocol is described.
- Analysis: conceptual/argumentative synthesis organized around data sources, AI technologies, and fairness/explainability, concluding with four recommendations for incorporating sex and gender into AI development.
Limitations
- The paper does not state explicit limitations of its own review.
- Evident from the text: it is a narrative (non-systematic) review without a stated search strategy, inclusion criteria, or formal quality appraisal, so the selection of examples and technologies may be selective rather than exhaustive.
- Evident from the text: several cited figures (e.g., the 18.6% women in the PD study, the 17% cardiologist figure) are drawn from single prior studies and are used illustratively rather than as generalizable estimates.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Figures



Verification
Claims the paper makes
17 Valid · 5 Not valid · 58 Not checked · 32 Own results and interpretations
Every source the paper cites for this claim was checked and supports it.
In the paper
“for the diagnosis and monitoring of diabetes, haemoglobin A1c (HbA1c) levels are routinely used, despite of differences associated with ethnicities135 and gender136.”
§ Sex and gender differences and biases in artificial intelligence for biomedicine and healthcare > Introduction > Desirable vs. undesirable biases
What the sources say
- Supportedcites [135] Cavagnolli, G., Pimentel, A. L., Freitas, P. A. C., Gross, J. L. & Camargo, J. L. Effect of ethnicity on HbA1c levels in individuals without diabetes: systematic review and meta-analysis. PLoS ONE 12, e0171315 (2017). source
The Introduction says HbA1c has long been used to assess glycaemic control and has also been used as a diagnostic criterion since 2010. The Discussion reports differences in HbA1c levels among ethnic groups and considers their implications for diagnosis and monitoring. The separate citation attached to gender is not part of the claim attributed to this source. (LLM, conf=0.92)
- Supportedcites [136] Bae, J. C. et al. Hemoglobin A1c values are affected by hemoglobin level and gender in non-anemic Koreans. J. Diabetes Investig. 5, 60–65 (2014). source
The Introduction describes HbA1c as a gold-standard measure of glycemic control and says it has been recommended for diagnosing diabetes. The Discussion reports gender differences in HbA1c at similar fasting glucose levels; this source supports the gender portion attached to citation 136. (LLM, conf=0.92)
plainstates an association- 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
Every source the paper cites for this claim was checked and supports it.
In the paper
“Consequently, health research and practices can be entangled with sex and gender inequalities and biases20.”
§ Sex and gender differences and biases in artificial intelligence for biomedicine and healthcare > Introduction
What the sources say
- Supportedcites [20] Hay, K. et al. Disrupting gender norms in health systems: making the case for change. Lancet S0140673619306488. https://doi.org/10.1016/S0140-6736(19)30648-8 (2019). source
The section “How do health systems affect gender inequalities in health?” says health services prioritise care according to traditional gender norms and describes resulting biases in care. “How do gender inequalities manifest in the health-care workforce?” also reports biases in peer review of papers and grants that affect health research. Together, these passages support the claim about both research and practice. (LLM, conf=0.91)
plainstates an association- 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
“constitute relevant sources of variation in a number of clinical and subclinical conditions, affecting risk factors, prevalence, age of onset, symptomatology manifestation, prognosis, biomarkers and treatment effectiveness2”
§ Sex and gender differences and biases in artificial intelligence for biomedicine and healthcare > Introduction
What the sources say
- Partly supportedcites [2] Regitz-Zagrosek, V. Sex and gender differences in health. Science & society series on sex and science. EMBO Rep. 13, 596–603 (2012). sourceExperimental — accuracy not yet measured
The Science & Society passages describe sex- and gender-related differences across many diseases, including risk, prevalence, age patterns, symptoms, outcomes and treatment efficacy. They also distinguish biological sex from socio-cultural gender. Our copy does not establish the claim's specific point about biomarkers, so the full list is not supported. (LLM, conf=0.84)
plainstates an association- 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
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“European male populations are the primary focus in genomics research and its derived clinical findings, neglecting other ethnicities and populations133.”
§ Sex and gender differences and biases in artificial intelligence for biomedicine and healthcare > Introduction > Desirable vs. undesirable biases
What the sources say
- Partly supportedcites [133] Bentley, A. R., Callier, S. & Rotimi, C. N. Diversity and inclusion in genomic research: why the uneven progress? J. Community Genet. 8, 255–266 (2017). sourceExperimental — accuracy not yet measured
The “Preferred Cohort” section says European-ancestry cohorts are preferred in genomic analyses and that their results are often presented as the main findings, while results for other populations receive less attention. “Informing genetic diagnoses” describes clinical consequences of excluding diverse populations. The paper does not establish that the populations receiving this focus are specifically male. (LLM, conf=0.88)
plainstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“There is a current lack of consideration of the demonstrated age-dependent sex differences in the symptomatology related with impairments in social communication and interaction, expressive behaviour, reciprocal conversation, non-verbal gestures for diagnostic purposes123.”
§ Sex and gender differences and biases in artificial intelligence for biomedicine and healthcare > Introduction > Desirable vs. undesirable biases
What the sources say
- Partly supportedcites [123] Werling, D. M. & Geschwind, D. H. Sex differences in autism spectrum disorders. Curr. Opin. Neurol. 26, 146–153 (2013). sourceExperimental — accuracy not yet measured
The Conclusion of Werling & Geschwind (2013) does support the 'lack of consideration for diagnostic purposes' element: it states that 'discussion continues as to whether females present the autistic phenotype differently than males, and further work is needed to determine ... if diagnostic criteria need to be adjusted to effectively identify these girls,' and the 'Presentation of ASD symptoms' section reports sex-differential symptom subdomains (externalizing vs. internalizing, repetitive/restricted behaviours) plus the Dworzynski threshold findings. However, our copy nowhere demonstrates these differences as AGE-DEPENDENT: the cited age-related references report age effects on overall severity (ref 22) and males showing more restricted/repetitive behaviour 'regardless of age' (ref 24). The specific subdomains named in the claim — expressive behaviour, reciprocal conversation, non-verbal gestures — are also absent; the paper's symptom differences concern externalizing/internalizing problems and RRBIs rather than those social-communication/ADI-ADOS items. (LLM, conf=0.75)
boosted- 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
48 more claims · 9 with their cited sources checked
Create a free account to see all 112 claims and to verify the 79 references still open.
- It may state two separate claims as one. 33
- Probably not a substantial claim. 36
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
“Though gender, race, and ethnicity are frequently identified as potential sources of bias, other factors may contribute to bias such as age, socioeconomic differences, and geography [58].”
From Bias in artificial intelligence algorithms and recommendations for mitigation (2023) · cites this paper as [58]
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
“When population data is merged to form a common, model-ready input, biases can emerge through the selection of input features that are maximally available across all subjects, establishing a “one-size-fits-all” approach42.”
From Bias recognition and mitigation strategies in artificial intelligence healthcare applications (2025) · cites this paper as [42]
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
“Data augmentation to address sparsity can generate synthetic data that fails to appropriately reflect true diversity, thus reinforcing biases42.”
From Bias recognition and mitigation strategies in artificial intelligence healthcare applications (2025) · cites this paper as [42]
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“Aggregation may result in bias when a “one-size-fits-all” model is used for groups with different conditional distributions [33].”
From Bias in artificial intelligence algorithms and recommendations for mitigation (2023) · cites this paper as [33]
The cited source says what the claim attributes to it.
In the citing paper
“It is clear that these powerful computational tools, if not diligently designed and audited, have the potential to perpetuate and even amplify existing biases, particularly those related to race, gender, and other societal constructs [40][41][42][43].”
From Fairness and Bias in Artificial Intelligence: A Brief Survey of Sources, Impacts, and Mitigation Strategies (2023) · cites this paper as [42]
3 more claims about this paper
Create a free account to see every claim other papers make about this one.
Citations
Cites
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
(2018) · Cited by 3,510
Dwork, C., Hardt, M., Pitassi, T., Reingold, O., & Zemel, R. (2012) · Proceedings of the 3rd Innovations in Theoretical Computer Science Conference · doi:10.1145/2090236.2090255 · Cited by 3,501 · cited 1×
“A theoretical implementation of such approach, also called fair af fi rmative action, has been proposed as an optimisation problem to obtain, at the same time, both group fairness (a.k.a statistical parity) and individual fairness 113 .”
BackgroundLinguistic Regularities in Continuous Space Word Representations
Mikolov, T., Yih, W.-T., & Zweig, G. (2013) · Cited by 2,912
Fitzpatrick, K. K., Darcy, A., & Vierhile, M. (2017) · JMIR Mental Health · doi:10.2196/mental.7785 · Cited by 2,730
Clinically applicable deep learning for diagnosis and referral in retinal disease
De Fauw, J., Ledsam, J. R., Romera-Paredes, B., Nikolov, S., Tomasev, N., Blackwell, S., Askham, H., Glorot, X., O’Donoghue, B., Visentin, D., et al. (2018) · Nature Medicine · doi:10.1038/s41591-018-0107-6 · Cited by 2,639
A Survey on Explainable Artificial Intelligence (XAI): Toward Medical XAI
Tjoa, E., & Guan, C. (2021) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2020.3027314 · Cited by 2,524
Reitsma, M. B., Fullman, N., Ng, M., Salama, J. S., Abajobir, A., Abate, K. H., Abbafati, C., Abera, S. F., Abraham, B., Abyu, G. Y., et al. (2017) · The Lancet · doi:10.1016/s0140-6736(17)30819-x · Cited by 2,028
The Path to Personalized Medicine
Hamburg, M. A., & Collins, F. S. (2010) · New England Journal of Medicine · doi:10.1056/nejmp1006304 · Cited by 1,970
Prediction of cardiovascular risk factors from retinal fundus photographs via deep learning
Poplin, R., Varadarajan, A. V., Blumer, K., Liu, Y., McConnell, M. V., Corrado, G. S., Peng, L., & Webster, D. R. (2018) · Nature Biomedical Engineering · doi:10.1038/s41551-018-0195-0 · Cited by 1,865
Research works printed but not linked: 5
- McGregor, A. J. Sex bias in drug research: a call for change. Pharmaceutical J. https://www.pharmaceutical-journal.com/opinion/comment/sex-bias-in-drug-research-a-call-for-change/20200727.article (2016).
- Bolukbasi, T., Chang, K.-W., Zou, J., Saligrama, V. & Kalai, A. Man is to computer programmer as woman is to homemaker? Debiasing Word Embeddings. In Proc. 30th International Conference on Neural Information Processing Systems. (eds Lee, D. D., Luxburg, U. V., Garnett, R., Sugiyama, M., Guyon, I. M.). 4356–4364 (NIPS, Barcelona, Spain, 2016).
- Kory-Westlund, J. M. & Breazeal, C. A Persona-Based Neural Conversation Model. In Proc. 18th ACM Interaction Design and Children Conference (IDC). (ed. Fails, J. A.) 38–50, (ACM Press, Boise, Idhao, US, 2019).
- Iacobacci, N. Exponential Ethics. (ATROPOS PRESS, 2018).
- Suresh, H. & Guttag, J. V. A framework for understanding unintended consequences of machine learning. Preprint at https://arxiv.org/abs/1901.10002 (2019). · arXiv
Cited by
Rajpurkar, P., Chen, E., Banerjee, O., & Topol, E. J. (2022) · Nature Medicine · doi:10.1038/s41591-021-01614-0 · Cited by 3,212
Artificial Intelligence and Business Value: a Literature Review
Enholm, I. M., Papagiannidis, E., Mikalef, P., & Krogstie, J. (2022) · Information Systems Frontiers · doi:10.1007/s10796-021-10186-w · Cited by 1,046
Ferrara, E. (2023) · Sci · doi:10.3390/sci6010003 · Cited by 875
Artificial intelligence for multimodal data integration in oncology
Lipkova, J., Chen, R. J., Chen, B., Lu, M. Y., Barbieri, M., Shao, D., Vaidya, A. J., Chen, C., Zhuang, L., Williamson, D. F. K., et al. (2022) · Cancer Cell · doi:10.1016/j.ccell.2022.09.012 · Cited by 777
Bias in artificial intelligence algorithms and recommendations for mitigation
Nazer, L. H., Zatarah, R., Waldrip, S., Ke, J. X. C., Moukheiber, M., Khanna, A. K., Hicklen, R. S., Moukheiber, L., Moukheiber, D., Ma, H., et al. (2023) · PLOS Digital Health · doi:10.1371/journal.pdig.0000278 · Cited by 621
“Though gender, race, and ethnicity are frequently identified as potential sources of bias, other factors may contribute to bias such as age, socioeconomic differences, and geography [58].”
Background
Metadata
- Authors
- Cirillo, D., Catuara-Solarz, S., Morey, C., Guney, E., Subirats, L., Mellino, S., Gigante, A., Valencia, A., Rementeria, M. J., Chadha, A. S., & Mavridis, N.
- Affiliations
- Barcelona Supercomputing Center (ES)
- Women's Brain Project (CH)
- Universitat Oberta de Catalunya (ES)
- Institució Catalana de Recerca i Estudis Avançats (ES)
- Corresponding author
- Cirillo, Davide
- ORCID
- Venue
- npj Digital Medicine
- Published
- 2020-06-01
- Volume
- 3
- Issue
- 1
- Pages
- 81-81
- ISSN
- 2398-6352 (ISSN-L)
- PMC
- PMC7264169
- OpenAlex
- W3029881688
- Semantic Scholar
- aa64d955454464ef5d921cc9df6682ff4921b2e3
- Cited by
- 553
- References
- 140
- Citations by year
- 2026: 71 · 2025: 115 · 2024: 114 · 2023: 103 · 2022: 86 · 2021: 59 · 2020: 5
- FWCI
- 10.57
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- cc-by
- creativecommons.org/licenses/by/4.0 (tdm)
- creativecommons.org/licenses/by/4.0 (vor)
- Locations
- npj Digital Medicine (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- UPCommons institutional repository (Universitat Politècnica de Catalunya) (acceptedVersion, cc-by)
- LA Referencia (Red Federada de Repositorios Institucionales de Publicaciones Científicas) (submittedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Funders
- EC | Horizon 2020 Framework Programme (116030)
- Topics
- Health Sciences › Medicine › Public Health, Environmental and Occupational Health › Sex and Gender in Healthcare
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Physical Sciences › Computer Science › Artificial Intelligence › Ethics and Social Impacts of AI
- Keywords
- sex differences, gender differences, algorithmic bias, bias detection, artificial intelligence in healthcare, precision medicine, health disparities
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 2026-10-04, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-04 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol and 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: Sex and gender differences and biases in artificial intelligence for biomedicine and healthcare. https://app.cochat.ai/papers/10.1038/s41746-020-0288-5
- The paper
- Cirillo, D., Catuara-Solarz, S., Morey, C., Guney, E., Subirats, L., Mellino, S., Gigante, A., Valencia, A., Rementeria, M. J., Chadha, A. S., & Mavridis, N. (2020). Sex and gender differences and biases in artificial intelligence for biomedicine and healthcare. npj Digital Medicine. https://doi.org/10.1038/s41746-020-0288-5
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.