Bias in medical AI: Implications for clinical decision-making
Open accessThe DOI matches its Crossref record; 124 of the 124 references it cites were found in a registry or an index; 30 of the 140 sources its claims cite were checked for support.
Abstract
Biases in medical artificial intelligence (AI) arise and compound throughout the AI lifecycle. These biases can have significant clinical consequences, especially in applications that involve clinical decision-making. Left unaddressed, biased medical AI can lead to substandard clinical decisions and the perpetuation and exacerbation of longstanding healthcare disparities. We discuss potential biases that can arise at different stages in the AI development pipeline and how they can affect AI algorithms and clinical decision-making. Bias can occur in data features and labels, model development and evaluation, deployment, and publication. Insufficient sample sizes for certain patient groups can result in suboptimal performance, algorithm underestimation, and clinically unmeaningful predictions. Missing patient findings can also produce biased model behavior, including capturable but nonrandomly missing data, such as diagnosis codes, and data that is not usually or not easily captured, such as social determinants of health. Expertly annotated labels used to train supervised learning models may reflect implicit cognitive biases or substandard care practices. Overreliance on performance metrics during model development may obscure bias and diminish a model’s clinical utility. When applied to data outside the training cohort, model performance can deteriorate from previous validation and can do so differentially across subgroups. How end users interact with deployed solutions can introduce bias. Finally, where models are developed and published, and by whom, impacts the trajectories and priorities of future medical AI development. Solutions to mitigate bias must be implemented with care, which include the collection of large and diverse data sets, statistical debiasing methods, thorough model evaluation, emphasis on model interpretability, and standardized bias reporting and transparency requirements. Prior to real-world implementation in clinical settings, rigorous validation through clinical trials is critical to demonstrate unbiased application. Addressing biases across model development stages is crucial for ensuring all patients benefit equitably from the future of medical AI.
Overview
CoChat Summary
TL;DR
This narrative review argues that bias in medical AI arises and compounds across the entire lifecycle—training data, model development and evaluation, deployment, and publication—and can worsen clinical decisions and health disparities. It catalogs mitigation strategies including diverse data collection, statistical debiasing, subgroup evaluation, interpretability, bias reporting, and clinical-trial validation.
Key findings
- Bias can enter at multiple stages: imbalanced sample sizes, nonrandomly missing data, uncaptured social determinants of health, biased labels, race/ethnicity in clinical algorithms, overreliance on whole-cohort performance metrics, publication biases, and implementation/end-user biases.
- The review cites specific examples and numbers: over half of published clinical AI models in 2019 used data from the US or China; radiology accounted for over 40% of AI publications in 2019; authors were 75% male; a MIMIC-III mortality model yielded recall rates as low as 25% under imbalanced racial representation; and the Epic Sepsis Model missed two-thirds of sepsis cases after deployment.
- Mitigation strategies discussed include large diverse data sets, oversampling methods such as SMOTE and ADASYN, imputation, SDoH capture, expert consensus labeling, subgroup analysis, bias-centered metrics like equalized odds and predictive parity, interpretability methods such as SHAP and LIME, adversarial debiasing, ongoing monitoring, and clinical trials for AI validation.
- Medical large language models pose unique bias risks because they are autoregressive, can reproduce biased or inaccurate training content, cannot assess their own outputs, and may generate inconsistent recommendations; statistical debiasing methods for clinical LLMs remain limited.
Methods
- Narrative review, not a systematic review, synthesizing literature and real-world or hypothetical examples across diverse medical domains and stages of the AI development pipeline.
- Includes Table 1 summarizing bias types by development stage, illustrative examples, potential clinical consequences, and mitigation strategies.
- No primary data collection, systematic search protocol, or quantitative meta-analysis is reported; the work draws on cited studies and examples.
Limitations
- Stated in the paper: more research is needed; there is a lack of standardized bias reporting guidelines; there are no explicit FDA regulations examining fairness or bias in medical AI outputs; and statistical debiasing methods for real-world clinical LLMs remain limited.
- Evident from the text: as a narrative review, it lacks systematic search and selection criteria and does not quantitatively synthesize evidence; examples are illustrative or hypothetical and not exhaustive; and the proposed mitigations are not empirically validated in this paper.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
17 Valid · 2 Not valid · 65 Not checked · 42 Own results and interpretations
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Fewer studies have concentrated specifically on the implications and consequences of biased AI in healthcare [19–22].”
§ Bias in medical AI: Implications for clinical decision-making > Related work
What the sources say
- Not checked yetcites [19] RB Parikh, 2019, Addressing bias in artificial intelligence in health care, JAMA source
- Supportedcites [20] MD Abràmoff, 2023, Considerations for addressing bias in artificial intelligence for health equity, NPJ Digit Med source
The cited paper is an example of a study focused on biased AI in healthcare and its consequences. Its Introduction describes AI exacerbating health disparities and contributing to worse outcomes for Black patients; the “Bias in the healthcare process” section examines unequal outcomes across groups. The citation supports identifying this paper as one of the studies, rather than establishing a count of how many such studies exist. (LLM, conf=0.86)
- Partly supportedcites [21] R Daneshjou, 2021, Lack of transparency and potential bias in artificial intelligence data sets and algorithms: a scoping review, JAMA Dermatol sourceExperimental — accuracy not yet measured
The Discussion says that few dermatology AI studies report patient race, ethnicity, or skin tone, limiting assessment of potential bias. That supports a narrower gap in studying bias, but the paper does not establish that fewer studies across healthcare have examined the implications and consequences of biased AI. (LLM, conf=0.82)
- Not checked yetcites [22] AS Albahri, 2023, A systematic review of trustworthy and explainable artificial intelligence in healthcare: Assessment of quality, bias risk, and data fusion., Inf Fusion. source
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.
The cited sources support only part of this claim — read each source’s verdict below.
In the 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].”
§ Bias in medical AI: Implications for clinical decision-making > Biases in training data > Imbalanced sample sizes
What the sources say
- Not checked yetcites [54] RF Wolff, 2019, PROBAST: a tool to assess the risk of bias and applicability of prediction model studies, Ann Intern Med source
- Partly supportedcites [55] T Hernandez-Boussard, 2020, MINIMAR (MINimum Information for Medical AI Reporting): Developing reporting standards for artificial intelligence in health care., J Am Med Inform Assoc sourceExperimental — accuracy not yet measured
The Introduction and General Principles of the MINIMAR Design say that reporting training-data characteristics can reveal potential biases and help assess whether an AI model generalizes to a clinical population. But the paper proposes MINIMAR; the passages found do not identify PROBAST or say that PROBAST was developed to assess biases from imbalanced data. No page numbers are available in our copy. (LLM, conf=0.83)
- Could not verifycites [56] Szczekocka E, Tarnec C, Pieczerak J, editors. Standardization on bias in Artificial Intelligence as industry support. 2022 IEEE International Conference on Big Data (Big Data); 2022: IEEE. sourceExperimental — accuracy not yet measured
The Standardization Specifics section says standards could help identify data-related biases, but it does not discuss PROBAST or a reporting checklist. Searches for PROBAST, imbalanced data, and clinical settings found no passage establishing the claimed purpose or use of that tool. (LLM, conf=0.76)
plainstates a cause- 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.
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“The researchers found that imbalanced representation of racial groups had a significant effect on model performance (yielding recall rates as low as 25%), raising critical concerns for the model’s potential application in real clinical settings [44].”
§ Bias in medical AI: Implications for clinical decision-making > Biases in training data > Imbalanced sample sizes
What the sources say
- Partly supportedcites [44] E Röösli, 2022, Peeking into a black box, the fairness and generalizability of a MIMIC-III benchmarking model., Sci Data. sourceExperimental — accuracy not yet measured
The Discussion reports significantly worse prediction performance for Black patients and says their low representation might partly explain the bias; it does not establish that imbalanced racial representation caused the performance gap. The Discussion also says the model identified only every fourth to fifth high-risk patient, but that low recall concerns the high-risk class overall, not a racial group. (LLM, conf=0.90)
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.
- Partly supported
The sources with a verdict support this claim; the rest weren’t checked yet, have no full text we could read, or their verdicts aren’t rated yet.
In the paper
“Furthermore, the degree of practitioner misclassification can vary along sociodemographic factors.”
§ Bias in medical AI: Implications for clinical decision-making > Biases in training data > Data labels and misclassification
What the sources say
- Supportedcites [84] GH Rauscher, 2013, Potentially missed detection with screening mammography: does the quality of radiologist’s interpretation vary by patient socioeconomic advantage/disadvantage?, Ann Epidemiol source
The INTRODUCTION states the goal was 'to examine whether variation in the quality of mammography interpretation as performed by the original reading radiologist varied by patient sociodemographic characteristics,' i.e. radiologist (practitioner) error by sociodemographic group. The RESULTS (p. 3, Table 3 discussion) report that potentially missed detection differed by minority ethnicity (+17 pp), income, education and lack of private insurance (+30 pp), and the DISCUSSION concludes the observed racial/ethnic and socioeconomic disparities 'shed light on differences in the quality of interpretation for different racial/ethnic and sociodemographic groups.' The Abstract's CONCLUSIONS affirm the association, so the hedged claim that practitioner misclassification 'can vary' along sociodemographic factors is supported; only the education association was attenuated in the selection-bias sensitivity analysis. (LLM, conf=0.90)
- Not checked yetcites [85] AM Kilbourne, 2005, Care without coverage: too little, too late., J Natl Med Assoc
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
“There are several well-documented instances of insufficient evidence for race and ethnicity to be used as predictive factors in models.”
§ Bias in medical AI: Implications for clinical decision-making > Biases in training data > Race and ethnicity in clinical algorithms
What the sources say
- Partly supportedcites [92] MA Marzinke, 2022, Limited evidence for use of a black race modifier in eGFR calculations: a systematic review, Clin Chem sourceExperimental — accuracy not yet measured
The Discussion (page unavailable) concludes that there is little evidence for including a Black race modifier in eGFR calculations. This documents an instance of insufficient evidence for using race in a model, but the paper does not establish the broader claim of several instances involving race and ethnicity across models. (LLM, conf=0.85)
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
65 more claims · 8 with their cited sources checked
Create a free account to see all 126 claims and to verify the 110 references still open.
- It may state two separate claims as one. 41
- Probably not a substantial claim. 28
- The quote only partly states this claim. 1
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
“Generating educational resources without clinical context review risks reinforcing data biases [172]”
From Applications, Challenges, and Prospects of Generative Artificial Intelligence Empowering Medical Education: Scoping Review (2025) · cites this paper as [172]
Experimental — accuracy not yet measuredThe check wasn’t confident either way, so it gives no verdict — never disputed.
In the citing paper
“In contrast, ensemble classification, combining multiple models, demonstrated improved accuracy and robustness [21,22].”
From Machine Learning in Maternity: Early Diagnosis of Placenta Accreta Spectrum (2026) · cites this paper as [22]
Experimental — accuracy not yet measured
Citations
Cites
LeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126
SMOTE synthetic minority over-sampling technique
Chawla, N. V., Bowyer, K. W., Hall, L. O., & Kegelmeyer, W. P. (2002) · Journal of Artificial Intelligence Research · doi:10.1613/jair.953 · Cited by 32,865 · cited 1×
“…which generates synthetic examples for the minority class by interpolating between existing minority class instances, and Adaptive Synthetic Sampling (ADASYN), which uses a similar method to SMOTE but focuses more on examples that are harder to classify (near the decision boundary) [49,50].”
MethodA Unified Approach to Interpreting Model Predictions
Lundberg, S., & Lee, S.-I. (2017) · arXiv · doi:10.48550/arxiv.1705.07874 · Cited by 23,707
Gauriat, C.-M., Pencolé, Y., Ribot, P., & Brouillet, G. (2024) · HAL (Le Centre pour la Communication Scientifique Directe) · doi:10.4230/oasics.dx.2024.27 · Cited by 12,586
Crenshaw, K. (2018) · Living with Contradictions · doi:10.4324/9780429499142-5 · Cited by 7,232
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 1×
“In reality, however, AI models are frequently found to be biased toward certain patient groups, leading to disparities not only in performance but also in potential or actual clinical benefits [7–9].”
BackgroundA Survey on Bias and Fairness in Machine Learning
Mehrabi, N., Morstatter, F., Saxena, N., Lerman, K., & Galstyan, A. (2022) · ACM Computing Surveys · doi:10.1145/3457607 · Cited by 5,812 · cited 1×
“The topic of bias in AI has garnered significant attention in recent years, with multiple review papers addressing different aspects of this critical issue [14–18].”
BackgroundCrenshaw, K. (1998) · Feminism And Politics · doi:10.1093/oso/9780198782063.003.0016 · Cited by 5,672
Social determinants of health inequalities
Marmot, M. (2005) · The Lancet · doi:10.1016/s0140-6736(05)71146-6 · Cited by 5,086
ADASYN: Adaptive synthetic sampling approach for imbalanced learning
He, H., Bai, Y., Garcia, E. A., & Li, S. (2008) · 2008 IEEE International Joint Conference on Neural Networks (IEEE World Congress on Computational Intelligence) · doi:10.1109/ijcnn.2008.4633969 · Cited by 4,521
Research works printed but not linked: 12
- Food, Administration D., 2019, Proposed regulatory framework for modifications to Artificial Intelligence/Machine Learning (AI/ML)-based Software as a Medical Device.
- D Roselli, 2019, editors. Managing bias in AI. Companion proceedings of the 2019, world wide web conference
- KPUL McFarling, 2024, Doctors use problematic race-based algorithms to guide care every day. Why are they so hard to change?, STAT
- Ribeiro MT, Singh S, Guestrin C. Model-agnostic interpretability of machine learning. arXiv preprint arXiv:160605386. 2016.
- Raji ID, Bender EM, Paullada A, Denton E, Hanna A. AI and the everything in the whole wide world benchmark. arXiv preprint arXiv:211115366. 2021.
- Chauhan VK, Clifton L, Salaün A, Lu HY, Branson K, Schwab P, et al. Sample Selection Bias in Machine Learning for Healthcare. arXiv preprint arXiv:240507841. 2024.
- Yang S, Wang Y, Chu X. A survey of deep learning techniques for neural machine translation. arXiv preprint arXiv:200207526. 2020.
- Topal MO, Bas A, van Heerden I. Exploring transformers in natural language generation: Gpt, bert, and xlnet. arXiv preprint arXiv:210208036. 2021.
- Schmidgall S, Harris C, Essien I, Olshvang D, Rahman T, Kim JW, et al. Addressing cognitive bias in medical language models. arXiv preprint arXiv:240208113. 2024.
- Ebrahimi S, Chen K, Asudeh A, Das G, Koudas N. AXOLOTL: Fairness through Assisted Self-Debiasing of Large Language Model Outputs. arXiv preprint arXiv:240300198. 2024.
- Li J, Tang Z, Liu X, Spirtes P, Zhang K, Leqi L, et al. Steering LLMs Towards Unbiased Responses: A Causality-Guided Debiasing Framework. arXiv preprint arXiv:240308743. 2024.
- Zhang Y-F, Yu W, Wen Q, Wang X, Zhang Z, Wang L, et al. Debiasing large visual language models. arXiv preprint arXiv:240305262. 2024.
Cited by
Items 101–200 of 716
Olia, J. B. H., Raman, A., Hsu, C., Alkhayyat, A., & Nourazarian, A. (2025) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2025.109984 · Cited by 10
Razavi, J. K., Dhafar, H. O., & BaHammam, A. S. (2025) · Saudi Journal for Health Sciences · doi:10.4103/sjhs.sjhs_26_25 · Cited by 15
Arora, T., Muhammad-Kamal, H., & Beam, K. (2025) · Seminars in Perinatology · doi:10.1016/j.semperi.2025.152100 · Cited by 12
Alami, H., Pozelli Sabio, R., Pérez, E. J., Gagnon, M.-P., Langlois, L., Denis, J.-L., Malas, K., Rivard, L., Savoldelli, M., Ag Ahmed, M. A., et al. (2026) · Journal of Medical Internet Research · doi:10.2196/87448 · Cited by 9
“Data security oversha-dowing data quality in current governance frameworks is a key insight of this study, as dataset quality can have a significant impact on mitigating bias and ensuring equity in the delivery of care using AI technologies [92-94].”
Hamamoto, R., Koyama, T., Takahashi, S., Yasuda, T., Kobayashi, K., Akagi, Y., Kouno, N., Sudo, K., Hirata, M., Sunami, K., et al. (2026) · Journal of Hematology & Oncology · doi:10.1186/s13045-026-01781-y · Cited by 8
“Benchmark and dataset biases may further obscure clinical validity if evaluation does not reflect real-world case mix and institutional heterogeneity [162].”
Representation of Demographics in Otolaryngology by Artificial Intelligence Text‐to‐Image Platforms
Shaari, A. L., Saad, A. M., Patel, A. M., & Filimonov, A. (2025) · Laryngoscope Investigative Otolaryngology · doi:10.1002/lio2.70152 · Cited by 8
“…mechanisms, increased transparency in the types of training data and processes, larger datasets, bias reporting, increased vigilance in determining what AI learns from, and increasing transparency of code between engineers and designers to avoid a lack of diversity in datasets [27, 28, 34–36].”
Impact of analytical bias on machine learning models for sepsis prediction using laboratory data
Yeşil, M. R., Talli, I., Pelloso, M., Cosma, C., Pangrazzi, E., Plebani, M., Üstündağ, Y., & Padoan, A. (2025) · Clinical Chemistry and Laboratory Medicine (CCLM) · doi:10.1515/cclm-2025-0491 · Cited by 8
Wislocki, K. E., Sami, S., Liberzon, G., & Zalta, A. K. (2025) · JMIR Mental Health · doi:10.2196/80801 · Cited by 9
“Our findings align with past research showing that GAI models demonstrate evidence of a variety of biases that may impact health care [19,21,24,44,45].”
Alrashed, F. A., Ahmad, T., Alsabih, A. O., Mahmoud, S., Almurdi, M. M., & Abdulghani, H. M. (2025) · Healthcare · doi:10.3390/healthcare13182377 · Cited by 10
“Medical doctors, who are trained to make complex, often subjective decisions based on a combination of medical knowledge and patient context, may be hesitant to fully trust AI systems in clinical practice [12].”
Lee, Y., Jeon, S., Kim, D., Auh, Q., Lee, J. H., & Noh, Y. (2025) · Communications Medicine · doi:10.1038/s43856-025-01081-5 · Cited by 12
“Logisticregression,inparticular,is highlyinterpretable,easilyimplemented,andwell-alignedwithclinical work fl ows 44 .”
Ethical Considerations in the Deployment of Artificial Intelligence in Surgery
Barrios, P. A., Meredyth, N. A., Morris, R., Haines, K., Kim, G. J., & Peterson, C. Y. (2025) · Journal of Surgical Research · doi:10.1016/j.jss.2025.08.019 · Cited by 9
He, Z., Zhao, L., Li, G., Wang, J., Cai, S., Tu, P., Chen, J., Wu, J., Zhang, J., Chen, R., et al. (2025) · Frontiers in Digital Health · doi:10.3389/fdgth.2025.1670510 · Cited by 8
“These issues align with the findings of Cross et al., highlighting the variability in the quality of AI-generated responses to medical inquiries (26).”
Oettl, F. C., Pruneski, J. A., Zsidai, B., Yu, Y., Cong, T., Tischer, T., Hirschmann, M. T., & Samuelsson, K. (2025) · Knee Surgery Sports Traumatology Arthroscopy · doi:10.1002/ksa.70145 · Cited by 11
“(cid:129) Ethical considerations and bias : One risk associated with medical AI is the potential to perpetuate and even amplify existing biases in healthcare [9, 12, 30, 63].”
AI-based neoadjuvant immunotherapy response prediction across pan-cancer: a comprehensive review
Deng, Y., Li, T., Wang, Y., Chen, S., Tang, F., Zhu, T., Ran, J., Yang, B., Zhang, X., Xu, R., et al. (2025) · Cancer Cell International · doi:10.1186/s12935-025-04063-8 · Cited by 11
“Moreover, biases in data collection, such as disparities in imaging quality between well-funded and resource-limited healthcare settings, can further exacerbate these issues [271, 272].”
Guo, X., Zhang, H., Yao, L., & Ma, W. (2025) · Critical Reviews in Oncology/Hematology · doi:10.1016/j.critrevonc.2025.105076 · Cited by 10
Olszewski, R., Brzeziński, J., Watros, K., & Rysz, J. (2026) · Applied Sciences · doi:10.3390/app16031423 · Cited by 9
Cai, X., Li, W., Shi, W., Cai, Y., & Zhou, J. (2026) · Journal of Medical Internet Research · doi:10.2196/82170 · Cited by 9
“If these biases are not addressed, they can result in inaccurate medical judgments and exacerbate existing health care disparities [18].”
Role of artificial intelligence in congenital heart disease
Niyogi, S. G., Nag, D. S., Shah, M. M., Swain, A., Naskar, C., Srivastava, P., & Kant, R. (2025) · World Journal of Clinical Pediatrics · doi:10.5409/wjcp.v14.i3.105926 · Cited by 12
AlGoraini, Y., Alsayyali, M., Alotaibi, O., Almeshawi, I., Alaifan, F., & Alrashed, R. (2025) · Frontiers in Public Health · doi:10.3389/fpubh.2025.1634638 · Cited by 9
From lab to life: technological innovations in transforming cancer metastasis detection and therapy
Basu, S., Sasikumar, S., Sur, S., Vaidya, V., Kheur, S. M., Gupta, S., Ranjan, A., Paul, M. K., Nawani, N. N., Bhat, A., et al. (2025) · Discover Oncology · doi:10.1007/s12672-025-02910-8 · Cited by 7
Xue, M., Yang, J., Wang, H., Yan, Z., Chen, X., Gao, W., Luo, R., Lv, X., & Ye, Z. (2025) · Journal of Anesthesia and Translational Medicine · doi:10.1016/j.jatmed.2025.06.003 · Cited by 10
“90 Recent analyses have revealed substantial performance degradation when AI models developed in one institution are deployed in different clinical environments, with accuracy drops of 10–25 % documented in cross-institutional validations.”
Nwokedi, V. U., Odedele, M. K., Yusuff, T. A., Anderson, I. A., Akanbi, O. O., Agu, C. P., & Omoike, A. O. (2025) · Epidemiology and Health Data Insights · doi:10.63946/ehdi/17369 · Cited by 8
“Because EHR data reflect historical care patterns — often marred by existing health inequities — models trained on them may reproduce or even mag-nify bias [39, 40].”
The balance between artificial and human intelligence in clinical practice
Marrella, D., Anttila, T., Ryhänen, J., Miller, R., Liu, B., & Liverneaux, P. A. (2026) · Journal of Hand Surgery (European Volume) · doi:10.1177/17531934251401382 · Cited by 8
“(4) Data-related – diversifying AI training datasets is essential (Cross et al., 2024).”
Artificial intelligence in emergency medicine: a narrative review
Rego, A., Arango-Ibañez, J. P., Taylor, R. A., Smith, M. E., Jones, D. D., Pelletier, J., Colletti, J. E., Gottlieb, M. A., & Long, B. (2026) · The American Journal of Emergency Medicine · doi:10.1016/j.ajem.2026.01.028 · Cited by 8
The role of artificial intelligence in sarcopenia: Advances, applications, and future directions
Yousaf, M. W., Nadeem, A. H., Nadeem, M. F., & Qaisar, R. (2026) · Computational Biology and Chemistry · doi:10.1016/j.compbiolchem.2026.108930 · Cited by 7
(2025) · Academic Medicine · doi:10.1097/acm.0000000000006099 · Cited by 21
Shostakovych-Koretska, L. R. (2025) · Інфекційні хвороби · doi:10.11603/1681-2727.2025.2.15302 · Cited by 7
Garcia, M. B., de Almeida, R. S., Acut, D. P., Almeida, R., Garcia, P. S., & Stefani, E. (2025) · IGI Global eBooks · doi:10.4018/979-8-3373-1519-5.ch005 · Cited by 9
Evaluation of large language model-generated medical information on idiopathic pulmonary fibrosis
Chérrez-Ojeda, I., Frye, B. C., Hoheisel, A., Cortés-Telles, A., Robles‐Velasco, K., Mateos‐Toledo, H., Figueiredo, R. G., Ryerson, C. J., Rodas‐Valero, G., & Calderón, J. C. (2025) · Frontiers in Artificial Intelligence · doi:10.3389/frai.2025.1618378 · Cited by 6
Heinz, M., Hakam, H. T., Salzmann, M., Prill, R., & Ramadanov, N. (2025) · Medicina · doi:10.3390/medicina61122080 · Cited by 9
Lim, E., & Lim, C. E. D. (2025) · AI and Ethics · doi:10.1007/s43681-025-00892-5 · Cited by 6
“A growing body of literature highlights that AI models in healthcare often lack critical sociodemographic information, which contributes to the skewing of diagnostic or treatment recommendations, particularly for underrepresented communities [17, 20 ].”
Zikopoulos, A., Moustakli, E., Potiris, A., Louis, K., Arkoulis, I., Vogiatzoglou, A. L., Tzeli, M., Kathopoulis, N., Christopoulos, P., Thomakos, N., et al. (2026) · Journal of Clinical Medicine · doi:10.3390/jcm15020686 · Cited by 7
“In addition to imbalanced datasets and biased clinical or demographic samples, predictive accuracy can also be impacted by variations in reference standards, preprocessing procedures, or outcome definitions between studies [80].”
Muda, N. B., & Sulaiman, M. A. (2026) · Scientific Reports · doi:10.1038/s41598-026-36550-x · Cited by 7
Shalmani, M. A., Khani, M., Taleban, A., Yi, Z., Fink, J. T., Weber, C. E., Lu, Q., & Luo, J. (2026) · Neural Computing and Applications · doi:10.1007/s00521-025-11695-4 · Cited by 6
Ngabo-Woods, H., Dunai, L., Verdú, I. S., & Tîrșu, V. (2026) · Information · doi:10.3390/info17020168 · Cited by 6
Cary, M. P., Russell, R. G., Silcox, C., Lytle, K. S., Lehmann, L. S., Hightower, M., Economou-Zavlanos, N., & Guilamo‐Ramos, V. (2025) · Journal of the American Medical Informatics Association · doi:10.1093/jamia/ocaf224 · Cited by 6
Artificial Intelligence in Cardiovascular Medicine: Focus on Hypertension
Varzideh, F., Mone, P., Kansakar, U., Pande, S., Jankauskas, S. S., & Santulli, G. (2026) · Hypertension · doi:10.1161/hypertensionaha.126.26094 · Cited by 10
Zhang, Y., Kirchler, T., & Wang, A. (2026) · Journal of Medical Systems · doi:10.1007/s10916-026-02376-3 · Cited by 7
Mishra, S. K., Hamadou, M., Gurung, K., Subba, A., Mukhia, P., Chhetri, T., Roy, A., Rai, P., & Georrge, J. J. (2026) · Archives of Computational Methods in Engineering · doi:10.1007/s11831-026-10617-4 · Cited by 6
Artificial Intelligence in Medicine: Basic Knowledge, Applications, Ethics and Perspectives
Cascella, M., Avella, M. T., Pumpo, M. D., Sebastiani, E., Bussa, M., Tagliaferri, A., Brunetti, B., Meloni, P., Bianchini, E., Graziani, P., et al. (2026) · Journal of Evaluation in Clinical Practice · doi:10.1111/jep.70484 · Cited by 7
Bangla, S., Kapoor, A., & Jeer, G. (2026) · International Journal for Equity in Health · doi:10.1186/s12939-026-02903-1 · Cited by 7
“Proprietary algorithms are often treated as “black boxes,” with developers unwilling to disclose training data, model architecture, or validation results [18].”
InfluentialWu, D., Yu, Q., Wu, Q., Wang, X., Wei, H., Li, F., Li, G., Wang, X., & Wang, X. (2026) · Frontiers in Immunology · doi:10.3389/fimmu.2026.1854718 · Cited by 6
“For liquid-biopsy and multi-omics AI systems, pre-analytical variability, batch effects and platform-speci fi c noise may further shift the input distribution and reduce model transportability (202).”
ChatGPT's Influence on Dental Education: Methodological Challenges and Ethical Considerations
Ardila, C. M., & Yadalam, P. K. (2024) · International Dental Journal · doi:10.1016/j.identj.2024.11.014 · Cited by 6
“5,6 While the authors discuss the implications of these biases for clinical and educational settings, their recommendations lack specificity.”
Han, Y., Lam, J. C. K., Li, V. O. K., & Cheung, L. Y. L. (2025) · Data & Policy · doi:10.1017/dap.2025.8 · Cited by 9
Comparative evaluation of LLMs in orthopedic surgery
Chundi, G., Dawar, A., Sarwar, S., Prasad, S., Vosbikian, M. M., & Ahmed, I. H. (2025) · Journal of Orthopaedic Reports · doi:10.1016/j.jorep.2025.100728 · Cited by 6
Jia, Q., Wen, Y., Liu, Y., Zhao, H., Yu, Q., Long, Y., Sun, D., & Yu, Y. (2026) · PLoS ONE · doi:10.1371/journal.pone.0348819 · Cited by 9
Zhang, M., Wu, K., Long, S., Jin, X., & Liu, B. (2025) · Frontiers in Pharmacology · doi:10.3389/fphar.2025.1578117 · Cited by 6
AI-generated data contamination erodes pathological variability and diagnostic reliabilityPreprint
He, H., Xiang, S., Zhang, Y., Zhu, Y., Zhang, J., Lu, Y., Deng, H., Alsentzer, E., Liu, Y., Chen, Q., et al. (2026) · medRxiv · doi:10.64898/2026.01.19.26344383 · Cited by 8
ВИКОРИСТАННЯ ШТУЧНОГО ІНТЕЛЕКТУ В КЛІНІЧНІЙ МЕДИЦИНІ Й НАУКОВИХ ДОСЛІДЖЕННЯХ
Shostakovych-Koretska, L. R., & Kopcha, V. S. (2025) · Медична освіта · doi:10.11603/m.2414-5998.2025.1.15382 · Cited by 5
Nielsen, J. P. S., Mikkelsen, A. K., Künzel, J., Sebelik, M., Madani, G., Yang, T., & Todsen, T. (2025) · Diagnostics · doi:10.3390/diagnostics15151848 · Cited by 5
“These especially include potential biases [34].”
Khasawneh, Y. J. A., Khasawneh, N. A. S., & Khasawneh, M. A. S. (2025) · Journal of Religion and Health · doi:10.1007/s10943-025-02390-8 · Cited by 6
“AI technologies can help reduce cognitive errors and decision-making biases by processing complex data and uncovering hidden patterns, thereby providing a data-driven foundation for informed decision-making (Cross et al., 2024).”
Yang, Y. K., Chen, Z., Shen, X., Chen, R. C., & Gao, H. (2025) · BioData Mining · doi:10.1186/s13040-025-00464-7 · Cited by 6
“(iv) Patient-Specific Issues: Heterogeneous health trajectories [9, 17] and the “cold-start” problem for new patients.”
Khalifa, A., & Hussein, R. (2026) · Journal of Medical Systems · doi:10.1007/s10916-026-02391-4 · Cited by 5
Using Artificial Intelligence and Machine Learning to Promote Child Health Equity
Cheung, R., Butler, M., Cooper, C., & Dhanoa, H. (2025) · PEDIATRICS · doi:10.1542/peds.2025-070739l · Cited by 5
León-Domínguez, U. (2025) · Ergonomics · doi:10.1080/00140139.2025.2560581 · Cited by 6
What Is Implementation Science
Fayaz-Bakhsh, A., Tania, J., Lutfi, S. L., Jha, A. K., & Rahmim, A. (2025) · PET Clinics · doi:10.1016/j.cpet.2025.09.002 · Cited by 5
Yang, Y., He, H., Yu, C., & Nia, P. S. (2026) · European Journal of Cardio-Thoracic Surgery · doi:10.1093/ejcts/ezag072 · Cited by 5
Li, F., Zhao, H., Huang, Q., Zhang, C., & Chen, G. (2026) · BMC Medical Education · doi:10.1186/s12909-026-08779-7 · Cited by 5
Ulizio, C., Dua, D., Mukul, N. M., Areti, S., Kostick, K. M., & Rahimzadeh, V. N. (2026) · JMIR AI · doi:10.2196/85163 · Cited by 5
“Additionally, many concerns have been raised regarding potential problems with health AI tools, including bias [47], transparency [32], and issues in clinical implementation [48], among others.”
Patients’ Perspectives on the Implementation of AI in Radiological Diagnostics: Focus Group Study
Karger, C. R. (2026) · Journal of Medical Internet Research · doi:10.2196/89178 · Cited by 2
“Such distortions can result in suboptimal performance for certain patient groups and incorrect diagnoses [12-14].”
Drug Discovery and Computational Chemistry: A State‐of‐the‐Art Review
Prajapati, A., Kachhadiya, R. N., & Patel, H. D. (2026) · ChemistrySelect · doi:10.1002/slct.202507418 · Cited by 6
Mehta, V., Mathur, A. S., Bhadania, M., Galletti, C., & Fraile, J. F. (2026) · Clinical and Experimental Dental Research · doi:10.1002/cre2.70375 · Cited by 7
“This can be a risky attribute in clinical decision‐making, leading to a clinician's dilemma of accepting or rejecting an outcome generated by AI models (Chinta et al. 2025; Cross et al. 2024).”
Generative AI in healthcare: foundations, applications, challenges and future directions
Zheng, J., Li, B., Li, H., & Lu, Y. (2026) · Journal of Management Analytics · doi:10.1080/23270012.2026.2647940 · Cited by 6
Bak, M., Schut, M. C., & Ploem, M. C. (2025) · Journal of Responsible Technology · doi:10.1016/j.jrt.2025.100133 · Cited by 10
Khakpoor, F. L., van der Vliet, W., Deng, J. D., & Pornpattananangkul, N. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.11.12.688133 · Cited by 4
Mitigated deployment strategy for ethical AI in clinical settings
Abdulrahman, S., & Trengove, M. (2025) · BMJ Health & Care Informatics · doi:10.1136/bmjhci-2024-101363 · Cited by 5
Torales, J. C., O’Higgins, M. G., Ortíz, B. D. M., Kunihiro, D., Carbajal, L., Pereira, F., Franco, E. D., Torres-Romero, A. D., Ventriglio, A., Castaldelli-Maia, J. M., et al. (2025) · International Journal of Social Psychiatry · doi:10.1177/00207640251391255 · Cited by 4
“Our findings reinforce this initial body of evidence by providing data from a different sociocultural context and in a language underrepresented in the literature on AI in healthcare (Cross et al., 2024).”
Veneziano, F. A., Mistrulli, R., Gioia, F. A., & Luca, L. D. (2025) · Mini-invasive Surgery · doi:10.20517/2574-1225.2025.119 · Cited by 4
Bias-Mitigated AI as a Foundation for Resilient and Effective Health Systems
Júnior, J. B. D. S., Birminghamm, M., Rivière-Cinnamond, A., Boisson, E., Valdez, M. L., García-Saisó, S., Souza, J. P., Martí, M., Richards, L.-M. L.-M., Guzman, J., et al. (2025) · JMIR Public Health and Surveillance · doi:10.2196/88457 · Cited by 4
“These actions align with global regulatory trends, including the EU AI Act, which mandates oversight for high-risk AI systems [20].”
Fantus, S., Li, J., Wang, T., & Tang, L. (2026) · Journal of Medical Internet Research · doi:10.2196/79613 · Cited by 5
“Yet, clinicians often assume that AI tools used in clinical settings have been ethically scrutinized prior to deployment [27].”
Mondillo, G., Perrotta, A., Masino, M., Colosimo, S., Frattolillo, V., & Abbate, F. G. (2026) · Therapeutics · doi:10.3390/therapeutics3010006 · Cited by 4
“Healthcare AI systems are particularly susceptible to bias because they reflect the training data, and if models are trained predominantly on data from certain demographic groups, they may underperform when applied to underrepresented populations [78].”
Montero, A. M. B., Huet‐Dastarac, M., Herranz-Hernández, S. M., Tengler, B., Buhl, E. S., Galapon, A., Cárdenas, C. E. S., Fusella, M., Herbin, G., de Hond, Y. J. M., et al. (2026) · Physics and Imaging in Radiation Oncology · doi:10.1016/j.phro.2026.100940 · Cited by 7
“Thus, inconsistent and insufficient documentation remains a widespread issue, undermining trust, reproducibility, and safe clinical integration, while also limiting AI literacy and adoption among clinical teams [14 – 17].”
Pepito, J. A., Babate, F. J., Ismael, J. D., Al‐Jumayile, S. S., & Siddique, M. K. B. (2026) · SAGE Open Nursing · doi:10.1177/23779608261433312 · Cited by 4
“Nurses must contextualize AI-generated recommendations, assess their provenance and limitations, and integrate them with patient values and situational factors (Cross et al., 2024).”
Balsano, C., Alisi, A., Burra, P., Calvaruso, V., Cammà, C., Campagner, A., Donatelli, P., Germani, G., Gerussi, A., Giuffrè, M., et al. (2026) · Digestive and Liver Disease · doi:10.1016/j.dld.2026.03.005 · Cited by 4
Quach, J., Theou, O., Maxwell, S. P., Abidi, S. S. R., Song, X., Rockwood, K. J., & Kehler, D. S. (2026) · Ageing Research Reviews · doi:10.1016/j.arr.2026.103122 · Cited by 4
Implementing Generative AI in Pharmacy Experiential Education: A Practical Guide for Preceptors
Alexander, K. M., Noel, Z. R., Nei, S. D., & Krauter, A. K. (2026) · JACCP JOURNAL OF THE AMERICAN COLLEGE OF CLINICAL PHARMACY · doi:10.1002/jac5.70216 · Cited by 6
“Another major ethical concern is the potential for AI systems to perpetuate or amplify existing biases in health care data, which may disproportionately impact marginalized populations and worsen health inequities [41–43].”
Wang, H. J., Shi, G., Liu, X., Zhang, J., Chen, S., Shi, J., Li, Y., Yue, G., & Wu, Y. (2026) · Expert Systems · doi:10.1111/exsy.70259 · Cited by 8
Artificial Intelligence Across the Drug Development Lifecycle
Demyashkin, G. A., Алексеевич, П. М., Zyryanov, S., Yavorskiy, A., Shegai, P., & Kaprin, A. (2026) · Medical Sciences · doi:10.3390/medsci14020248 · Cited by 6
“The risk of AI models amplifying errors or pre-existing biases in their training data is a significant concern, raising questions about the generalizability of insights across diverse patient populations and the ethical implications of biased outcomes [249,250].”
Epistemic Compression and Symbolic Absence: Rethinking the Human and the Political
Lutfi, R. (2026) · Dia-noesis. · doi:10.12681/dia.45799 · Cited by 4
Malkoç, M. (2026) · JMIR Infodemiology · doi:10.2196/96664 · Cited by 4
Virtual cells for predictive immunotherapy
Bergman, D. R., & Fertig, E. J. (2025) · Nature Biotechnology · doi:10.1038/s41587-025-02583-2 · Cited by 4
Pawan, S. J., Muellner, M., Lei, X., Desai, M. M., Varghese, B. A., Duddalwar, V. A., & Cen, S. (2025) · Multimodal Technologies and Interaction · doi:10.3390/mti9050049 · Cited by 4
“Most cases in cluster A corresponded to iris species 2, though some instances from cluster C were also mixed into this category.”
Shafiullah (2025) · World Journal of Urology · doi:10.1007/s00345-025-05832-w · Cited by 4
Sallam, M., Beaini, C., Mijwil, M. M., & Sallam, M. (2025) · Journal of Studies in Medical Sciences · doi:10.20849/jsms.v1i1.1515 · Cited by 4
“As AI becomes a diagnostic gatekeeper, it risks encoding historical biases into future care unless equity is proactively engineered into its development, deployment, and regulation (Chen et al., 2024; Cross et al., 2024).”
Young, A., Paloka, R., Islam, A., Prasanna, P., Hill, V. B., & Payne, D. L. (2025) · Academic Radiology · doi:10.1016/j.acra.2025.09.008 · Cited by 3
AI-driven strategies for advancing corneal cell therapy: a promising frontier
Tafti, M. F., Khorrami‐Nejad, M., Arabfard, M., Ghiasi, M., Afkhamizadeh, F., Jadidi, K., & Aghamollaei, H. (2025) · Frontiers in Medicine · doi:10.3389/fmed.2025.1563891 · Cited by 3
“If unaddressed, biased AI can drive non-standard clinical decisions and exacerbate healthcare disparities (112, 113).”
Olariu, M.-E., Burlacu, A., Brinza, C., & Iftene, A. (2025) · Future Cardiology · doi:10.1080/14796678.2025.2573566 · Cited by 3
Miao, Y.-R., Wang, Y., Shi, L., Lv, J.-T., & Yang, X. (2025) · World Journal of Gastroenterology · doi:10.3748/wjg.v31.i40.111389 · Cited by 3
“Such bias may influence clinical decision-making at multiple stages of the AI lifecycle, potentially exacerbating healthcare disparities[89].”
Predicability of PD-L1 expression in cancer cells based solely on H&E-stained sections
Faa, G., Fraschini, M., Ziranu, P., Pretta, A., Porcu, G. S., Saba, L., Scartozzi, M., Shokun, N., & Rugge, M. (2025) · Journal of Pathology Informatics · doi:10.1016/j.jpi.2025.100524 · Cited by 3
Kostick, K. M., Hurley, M. E., Ayaz, S., Herrington, J. D., Zampella, C., Parish‐Morris, J., Tunç, B., Lázaro‐Muñoz, G., Blumenthal‐Barby, J., & Storch, E. A. (2025) · JMIR Mental Health · doi:10.2196/79182 · Cited by 6
“Critics have also warned against overreliance on algorithmic inferences about patients’ health status [32,33], especially in “black box” systems that resist clinical scrutiny and accountability and compromise informed clinical decision-making [34].”
Lam, K., Latorre, J. C., Yiu, A. J., Navin, G., Tan, A., & Qiu, J. (2025) · The Lancet Digital Health · doi:10.1016/j.landig.2025.100922 · Cited by 7
Joamets, K. (2025) · TalTech Journal of European Studies · doi:10.2478/bjes-2025-0037 · Cited by 4
“Bias in ADM processes refers to systematic assumptions or flaws embedded within AI systems, their underlying algorithms or the datasets used for training, which produce unfair, distorted or discriminatory outcomes (Cross et al ., 2024; Bhattacharya et al ., 2024; Hofmann, 2025).”
Wang, J., Xun, J.-J., & Zhao, F.-F. (2026) · Frontiers in Public Health · doi:10.3389/fpubh.2025.1724546 · Cited by 3
Can AI train the next generation of counsellors? Ethical challenges and opportunities
Athanassopoulos, L. (2026) · Journal of Psychology and AI · doi:10.1080/29974100.2025.2606673 · Cited by 3
Ibrahim, Z. A., & Al-Sambawee, M. A. J. (2026) · Discover Artificial Intelligence · doi:10.1007/s44163-026-00956-3 · Cited by 3
Yu, C. (2026) · Frontiers in Public Health · doi:10.3389/fpubh.2026.1756755 · Cited by 5
“(3) Exacerbation of health inequalities: AI bias may exacerbate existing healthcare gaps, leading to underserved groups (such as ethnic minorities, low-income populations, and people in remote areas) receiving poor treatment outcomes and poor performance of AI models in these groups (133).”
Nurses confront artificial intelligence-generated health information at the bedside: A case study
Cueto, S., Gagne, J. C. D., Merrill, T., & Cary, M. P. (2026) · Nursing Outlook · doi:10.1016/j.outlook.2026.102738 · Cited by 5
Ali, U., Shafi, I., Ahmad, J., Caceres, A. Z., Montero, T. C. C., Rehman, H. M. R. U., & Ashraf, I. (2026) · Computational and Structural Biotechnology Journal · doi:10.34133/csbj.0023 · Cited by 4
“To make progress on this challenge, larger and more diverse datasets must be created to train simpler models or learn from smaller datasets or unannotated datasets (synthetic data generation and few-shot learning) [ 115 ].”
AI ‐Driven Epidemiology: The Next Frontier in Precision Public Health
Prema, S. S., Ganapathy, D., & Shanmugamprema, D. (2026) · Wiley Interdisciplinary Reviews Data Mining and Knowledge Discovery · doi:10.1002/widm.70077 · Cited by 6
“If these biases are not identified and mitigated, AI systems risk reinforcing discriminatory patterns, leading to unequal disease surveillance, misallocation of healthcare resources, and biased risk assessments (Cross et al. 2024).”
Biomarkers of Treatment Response in Paediatric Medulloblastoma
Bayanova, M., Saliev, T. M., Zhakupov, A., Abdikadirova, A., Sapargaliyeva, M., Ibraimov, B., & Bolatov, A. (2026) · Diagnostics · doi:10.3390/diagnostics16071089 · Cited by 3
“Imaging biomarkers and artificial intelligence–based models are particularly susceptible to protocol-dependent variation, overfitting, and limited generalisability, raising concerns about transparency, interpretability, and unintended bias [93].”
Metadata
- Authors
- Cross, J. L., Choma, M. A., & Onofrey, J. A.
- Affiliations
- Yale University (US)
- Corresponding author
- Cross, James L.
- Venue
- PLOS Digital Health
- Publisher
- Public Library of Science (PLoS)
- Published
- 2024-11-07
- Volume
- 3
- Issue
- 11
- Pages
- e0000651
- ISSN
- 2767-3170 (ISSN-L)
- PMC
- PMC11542778
- OpenAlex
- W4404134492
- Cited by
- 607
- References
- 120
- Citations by year
- 2026: 382 · 2025: 222 · 2024: 3
- FWCI
- 13.83
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- PLOS Digital Health (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- RePEc: Research Papers in Economics (submittedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- 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
- algorithmic bias, artificial intelligence in healthcare, clinical decision-making, health care disparities, biased training data, missing data, label bias, model evaluation, bias mitigation, model interpretability, social determinants of health, clinical validation
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 (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
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.