The DOI matches its Crossref record; 53 of the 53 references it cites were found in a registry or an index.
Unmasking bias in artificial intelligence: a systematic review of bias detection and mitigation strategies in electronic health record-based models
Open accessAbstract
Objectives: Leveraging artificial intelligence (AI) in conjunction with electronic health records (EHRs) holds transformative potential to improve healthcare. However, addressing bias in AI, which risks worsening healthcare disparities, cannot be overlooked. This study reviews methods to handle various biases in AI models developed using EHR data. Materials and Methods: We conducted a systematic review following the Preferred Reporting Items for Systematic Reviews and Meta-analyses guidelines, analyzing articles from PubMed, Web of Science, and IEEE published between January 01, 2010 and December 17, 2023. The review identified key biases, outlined strategies for detecting and mitigating bias throughout the AI model development, and analyzed metrics for bias assessment. Results: Of the 450 articles retrieved, 20 met our criteria, revealing 6 major bias types: algorithmic, confounding, implicit, measurement, selection, and temporal. The AI models were primarily developed for predictive tasks, yet none have been deployed in real-world healthcare settings. Five studies concentrated on the detection of implicit and algorithmic biases employing fairness metrics like statistical parity, equal opportunity, and predictive equity. Fifteen studies proposed strategies for mitigating biases, especially targeting implicit and selection biases. These strategies, evaluated through both performance and fairness metrics, predominantly involved data collection and preprocessing techniques like resampling and reweighting. Discussion: This review highlights evolving strategies to mitigate bias in EHR-based AI models, emphasizing the urgent need for both standardized and detailed reporting of the methodologies and systematic real-world testing and evaluation. Such measures are essential for gauging models’ practical impact and fostering ethical AI that ensures fairness and equity in healthcare.
Overview
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
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
“Thus, its responses can reflect the subjective leanings originated from its data and embedded in its training, poten tially skewing responses in subtle yet significant ways (Chen et al., 2024).”
From If ChatGPT can do it, where is my creativity? generative AI boosts performance but diminishes experience in creative writing (2025) · cites this paper as Chen et al., 2024
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
“However, these reviews often overlook the importance of addressing bias across multiple levels of the development process.”
From Advancing Health Equity through Multi-Level Fairness in Health Informatics (2026) · cites this paper as [6]
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions
Sterne, J. A., Hernán, M. A., Reeves, B. C., Savović, J., Berkman, N. D., Viswanathan, M., Henry, D., Altman, D. G., Ansari, M. T., Boutron, I., et al. (2016) · BMJ · doi:10.1136/bmj.i4919 · Cited by 20,329
High-performance medicine: the convergence of human and artificial intelligence
Topol, E. J. (2019) · Nature Medicine · doi:10.1038/s41591-018-0300-7 · Cited by 10,450
A 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 2×
“Our analyses were further enhanced by consolidating established bias risk assessment tools, such as ROBINS-I, 16 ROBINS-E, 17 and PROBAST, 18 and relevant review articles, such as Mehrabi et al 19 which categorized potential biases in AI applications.”
Rajpurkar, P., Chen, E., Banerjee, O., & Topol, E. J. (2022) · Nature Medicine · doi:10.1038/s41591-021-01614-0 · Cited by 3,212 · cited 1×
“8,9 Biases can emerge at various stages of the model development lifecycle, 10–12 underlining the importance of developing and implementing strategies for effective detection, mitigation, and evaluation of biases.”
PROBAST: A Tool to Assess the Risk of Bias and Applicability of Prediction Model Studies
Wolff, R. F., Moons, K. G. M., Riley, R. D., Whiting, P. F., Westwood, M., Collins, G. S., Reitsma, J. B., Kleijnen, J., Mallett, S., & for the PROBAST Group† (2019) · Annals of Internal Medicine · doi:10.7326/m18-1376 · Cited by 2,984 · cited 1×
“Our analyses were further enhanced by consolidating established bias risk assessment tools, such as ROBINS-I, 16 ROBINS-E, 17 and PROBAST, 18 and relevant review articles, such as Mehrabi et al 19 which categorized potential biases in AI applications.”
Scalable and accurate deep learning with electronic health records
Rajkomar, A., Oren, E., Chen, K., Dai, A. M., Hajaj, N., Hardt, M., Liu, P. J., Liu, X., Marcus, J., Sun, M., et al. (2018) · npj Digital Medicine · doi:10.1038/s41746-018-0029-1 · Cited by 2,648 · cited 1×
“46 On the other hand, AI models may exhibit biases, and if left unaddressed, they have the potential to exacerbate healthcare disparities.”
Implicit bias in healthcare professionals: a systematic review
FitzGerald, C., & Hurst, S. (2017) · BMC Medical Ethics · doi:10.1186/s12910-017-0179-8 · Cited by 2,556 · cited 1×
“22 It often stems from preexisting bias within the data (such as stereotypes and flawed case assumptions) which may occur in the data collection and data preprocessing steps.”
Scalable and accurate deep learning with electronic health records.
Rajkomar, A., Oren, E., Chen, K., Dai, A. M., Hajaj, N., Hardt, M., Liu, P. J., Liu, X., Marcus, J., Sun, M., et al. (2018) · eScholarship (California Digital Library) · Cited by 2,242
Potential Biases in Machine Learning Algorithms Using Electronic Health Record Data
Gianfrancesco, M. A., Tamang, S., Yazdany, J., & Schmajuk, G. (2018) · JAMA Internal Medicine · doi:10.1001/jamainternmed.2018.3763 · Cited by 1,395 · cited 1×
“48,49 In this systematic review, we have comprehensively identified and scrutinized potential biases in AI models developed from EHR data, revealing their potential influence on healthcare disparities.”
Do no harm: a roadmap for responsible machine learning for health care
Wiens, J., Saria, S., Sendak, M., Ghassemi, M., Liu, V. X., Doshi-Velez, F., Jung, K., Heller, K., Kale, D., Saeed, M., et al. (2019) · Nature Medicine · doi:10.1038/s41591-019-0548-6 · Cited by 1,266 · cited 1×
“13 While a few scoping reviews have been conducted to understand bias in broader medical AI, 14,15 a focused review on AI models derived from EHR data is notably absent.”
Research works printed but not linked: 1
- Johnson, 2000, Sampling bias and other methodological threats to the validity of health survey research, Int J Stress Manag, 7, 247, 10.1023/A:1009589812697 · DOI
Cited by
Sorted by citations
Saeidnia, H. R., Fotami, S. G. H., Lund, B. D., & Ghiasi, N. (2024) · Social Sciences · doi:10.3390/socsci13070381 · Cited by 227
“Future research should aim to overcome these limitations to provide a more comprehensive understanding of ethical considerations in AI-based mental health interventions and to inform responsible practice and policy development.”
Bouguettaya, A., Stuart, E. M., & Aboujaoude, E. N. (2025) · npj Digital Medicine · doi:10.1038/s41746-025-01746-4 · Cited by 60
Transparency of medical artificial intelligence systems
Kim, C., Gadgil, S. U., & Lee, S. (2025) · Nature Reviews Bioengineering · doi:10.1038/s44222-025-00363-w · Cited by 60
“Detailed documentation allows developers to mitigate potential biases during model training using computational techniques, even if biases are discovered after data collection 17,66,67 .”
Jamshidi, S., Nikanjam, A., Nafi, K. W., Khomh, F., & Rasta, R. (2025) · Internet of Things · doi:10.1016/j.iot.2025.101531 · Cited by 55
Mei, P., Brewis, D. N., Nwaiwu, F., Sumanathilaka, D., Alva-Manchego, F., & Demaree-Cotton, J. (2025) · Computers in Human Behavior: Artificial Humans · doi:10.1016/j.chbah.2025.100140 · Cited by 50
“Thus, its responses can reflect the subjective leanings originated from its data and embedded in its training, potentially skewing responses in subtle yet significant ways (Chen et al., 2024).”
Collaborative, T. C. (2025) · BMJ · doi:10.1136/bmj-2024-083305 · Cited by 52
“54 58 As with other performance assessments, results should be quantified where possible to facilitate comparative analyses.”
Artificial intelligence tool development: what clinicians need to know?
Chew, B.-H., & Ngiam, K. Y. (2025) · BMC Medicine · doi:10.1186/s12916-025-04076-0 · Cited by 41
“This includes creating synthetic data based on the electronic health records, omics data and bioimages to train diagnostic and predictive models [102, 103].”
Exploring Artificial Intelligence Biases in Predictive Models for Cancer Diagnosis
Smiley, A., Reátegui-Rivera, C. M., Villarreal‐Zegarra, D., Escobar-Agreda, S., & Finkelstein, J. (2025) · Cancers · doi:10.3390/cancers17030407 · Cited by 43
“On the other hand, other types of biases are general across different AI models in healthcare, including implicit bias, selection bias, measurement bias, confounding bias, algorithmic bias, and temporal bias [9].”
Biases in AI: acknowledging and addressing the inevitable ethical issues
Hofmann, B. (2025) · Frontiers in Digital Health · doi:10.3389/fdgth.2025.1614105 · Cited by 52
Mitigating Algorithmic Bias in AI-Driven Cardiovascular Imaging for Fairer Diagnostics
Sufian, M. A., Alsadder, L., Hamzi, W., Zaman, S., Sagar, A. S. M. S., & Hamzi, B. (2024) · Diagnostics · doi:10.3390/diagnostics14232675 · Cited by 38
The ethics of data mining in healthcare: challenges, frameworks, and future directions
Ahmed, M. M., Okesanya, O. J., Oweidat, M. H., Othman, Z. K., Musa, S. S., & Lucero‐Prisno, D. E. (2025) · BioData Mining · doi:10.1186/s13040-025-00461-w · Cited by 41
“A study found that fewer than a quarter of the 20 included studies reported any fairness metric and most failed to analyze performance across demographic subgroups [72].”
Leveraging Large Language Models for Enhancing Safety in Maritime Operations
Miller, T. I., Durlik, I., Kostecka, E., Łobodzińska, A., Łazuga, K., & Kozlovska, P. (2025) · Applied Sciences · doi:10.3390/app15031666 · Cited by 35
Riley, B. K., & Dixon, A. L. (2024) · Current Opinion in Psychology · doi:10.1016/j.copsyc.2024.101833 · Cited by 38
Cofré, S., Sanchez, C., Quezada-Figueroa, G., & López-Cortés, X. A. (2025) · British Journal Of Nutrition · doi:10.1017/s0007114525000522 · Cited by 28
Sousa‐Pinto, B., Marques‐Cruz, M., Neumann, I., Chi, Y., Nowak, A. J., Reinap, M., Awad, M., Nothacker, M., Trucl, M., Brożek, J., et al. (2025) · Annals of Internal Medicine · doi:10.7326/annals-24-02338 · Cited by 27
“For example, these outputs can re fl ect disparities in access to care, differences of attention to speci fi c health conditions, or differential measurement errors across speci fi c population subgroups (20).”
Abogunrin, S., Muir, J. M., Zerbini, C., & Sarri, G. (2025) · Frontiers in Pharmacology · doi:10.3389/fphar.2025.1454245 · Cited by 27
Navigating Uncertainty: A User-Perspective Survey of Trustworthiness of AI in Healthcare
Ojha, J., Presacan, O., Lind, P. G., Monteiro, E., & Yazidi, A. (2025) · ACM Transactions on Computing for Healthcare · doi:10.1145/3716317 · Cited by 31
“By acknowledging and addressing these uncertainties, we can strive for a more reliable and trustworthy integration of AI in healthcare practices [9].”
From Genomics to AI: Revolutionizing Precision Medicine in Oncology
Calvino, G., Farro, J., Zampatti, S., Peconi, C., Megalizzi, D., Trastulli, G., Andreucci, S., Cascella, R., Strafella, C., Caltagirone, C., et al. (2025) · Applied Sciences · doi:10.3390/app15126578 · Cited by 30
Giri, R., Firdhos, S. H., & Vida, T. A. (2025) · Journal of Clinical Medicine · doi:10.3390/jcm14196900 · Cited by 24
“Discussions in medical ethics and law suggest that responsibility in AI-administered anesthesia may extend to multiple stakeholders, including software developers, hospital systems, and the supervising anesthesiologist [55].”
Automatic Classification of Cardiac Arrhythmias Using Deep Learning Techniques: A Systematic Review
Vásquez-Iturralde, F., Flores-Calero, M. J., Grijalva, F., & Rosales, A. (2024) · IEEE Access · doi:10.1109/access.2024.3408282 · Cited by 23
A roadmap to implementing machine learning in healthcare: from concept to practice
Yan, A. P., Guo, L. L., Inoue, J., Arciniegas, S. E., Vettese, E., Wolochacz, A., Crellin‐Parsons, N., Purves, B., Wallace, S., Patel, A., et al. (2025) · Frontiers in Digital Health · doi:10.3389/fdgth.2025.1462751 · Cited by 24
Rochon, M., Tanner, J., Jurkiewicz, J., Beckhelling, J., Aondoakaa, A. S., Wilson, K., Dhoonmoon, L., Underwood, M., Mason, L., Harris, R., et al. (2024) · PLoS ONE · doi:10.1371/journal.pone.0315384 · Cited by 28
Teles, A. S., de Moura, I. R., Silva, F., Roberts, A., & Stahl, D. R. (2025) · Information Fusion · doi:10.1016/j.inffus.2025.102981 · Cited by 23
Biases in Artificial Intelligence Application in Pain Medicine
Jumreornvong, O., Perez, A., Malave, B., Fatimah, M., Kia, A., & Nwaneshiudu, C. A. (2025) · Journal of Pain Research · doi:10.2147/jpr.s495934 · Cited by 22
“62,63 Finally, equitable communication strategies tailored to vulnerable populations can enhance patient care, while machine learning models trained on diverse data can improve the accuracy of pain diagnoses and treatment recommendations, reducing disparities among those most affected by bias.”
Albergante, L., O’Flynn, C., & Meyer, G. D. (2025) · Journal of the American Veterinary Medical Association · doi:10.2460/javma.24.09.0617 · Cited by 16
“75,76 Overall, while progress is being made, the op - portunities provided by AI have not yet moved the needle significantly for early disease detection in the clinic.”
Loaiza‐Bonilla, A., Thaker, N. G., Chung, C. D., Parikh, R. B., Stapleton, S., & Borkowski, P. (2025) · American Society of Clinical Oncology Educational Book · doi:10.1200/edbk-25-100048 · Cited by 16
Machine learning in dentistry: a scoping review
Lakhotia, S., Godrej, H., Kaur, A., Nutakki, C. S., Mun, M., Eber, P., & Celi, L. A. (2025) · PLOS Digital Health · doi:10.1371/journal.pdig.0000940 · Cited by 16
“Our findings are consistent with previous reviews that highlight the growing use of deep learning in medical imaging, but also indicate shortcomings in bias reporting [13] and calibration performance evaluation [14].”
Horsfall, L., Bondaronek, P., Ive, J., & Poduval, S. (2025) · Clinical Epidemiology · doi:10.2147/clep.s527000 · Cited by 15
“83,86 Preprocessing Preprocessing refers to general data cleaning and the conversion of raw data into a format appropriate for analysis.”
InfluentialUse of AI as a Research Recruitment Tool: Ethical Elements, Considerations, and Recommendations
Rentzepis, P. J., Snyder, D. L., Baker, T. J., & Bierer, B. E. (2026) · The American Journal of Bioethics · doi:10.1080/15265161.2026.2649234 · Cited by 14
“Bias is engrained in EHRs, the most common data source for AI recruitment, by who receives care at the healthcare entity, insurance status, location, and choice of language, all of which will impact AI-based natural language processing (Verheij et al. 2018; Harton et al. 2022; Chen et al. 2024).”
Kababji, S. E., Mitsakakis, N., Jonker, E., Beltran-Bless, A.-A., Pond, G. R., Vandermeer, L., Radhakrishnan, D., Mosquera, L., Paterson, A. H. G., Shepherd, L. E., et al. (2025) · Journal of Medical Internet Research · doi:10.2196/66821 · Cited by 15
MLHOps: Machine Learning Health Operations
Khattak, F. K., Subasri, V., Krishnan, A., Pou-Prom, C., Koçak, S. A., Dolatabadi, E., Pandya, D., Seyyed-Kalantari, L., & Rudzicz, F. (2024) · IEEE Access · doi:10.1109/access.2024.3521279 · Cited by 31
“…focus on developing standardized evaluation frameworks that include metrics and benchmark datasets for assessing the bias in the models [11], [51], [94]. b) Regulatory guidelines: Regulatory guidelines should be established to ensure and prioritize bias evaluation [59], [195]. c) Patient…”
Siddiqui, E. F., Haleem, M., Ahmad, S. F., Salhi, A., Zamani, A. T., & Varish, N. (2025) · IEEE Access · doi:10.1109/access.2025.3610526 · Cited by 16
“If you feed LLM modified data when training then LLM can utilize data poisoning and model manipulation to create biased and incorrect outcomes [51].”
Generative artificial intelligence for literature reviews
Wagner, G., Prester, J., Mousavi, R., Lukyanenko, R., & Paré, G. (2026) · Journal of Information Technology · doi:10.1177/02683962261425675 · Cited by 17
“This work can leverage growing research on AI data bias identi fi cation and mitigation (Chen et al., 2024; Nazer et al., 2023; Tejani et al., 2024).”
What Is Fair? Defining Fairness in Machine Learning for Health
Gao, J., Chou, B., McCaw, Z. R., Thurston, H., Varghese, P., Hong, C., & Gronsbell, J. L. (2025) · Statistics in Medicine · doi:10.1002/sim.70234 · Cited by 17
“Similarly, a review of EHR-based prediction models found that most studies investigate the over-all performance of ML models, but do not interrogate potential biases [34].”
Zimmerman, R. M., Hernandez, E. J., Yandell, M. D., Tristani‐Firouzi, M., Silver, R. M., Grobman, W. A., Haas, D. M., Saade, G. A., Steller, J. G., & Blue, N. R. (2025) · BMC Pregnancy and Childbirth · doi:10.1186/s12884-024-07095-6 · Cited by 12
Acosta-Jiménez, S., Mendoza-Mendoza, M. M., Galvan-Tejada, C. E., Galván-Tejada, J. I., Celaya-Padilla, J. M., García-Domínguez, A., Gamboa-Rosales, H., & Solís-Robles, R. (2024) · Network Modeling Analysis in Health Informatics and Bioinformatics · doi:10.1007/s13721-024-00497-8 · Cited by 9
“Not only does this affect the accuracy of artificial intelligence models, but it can also perpetuate disparities in medical care (Chen et al. 2024; Celi et al. 2022).”
Preparing Allergists to Practice in 2050 Using Artificial Intelligence
Khoury, P., Oppenheimer, J., Bunyavanich, S., Ciaccio, C. E., & Portnoy, J. M. (2025) · The Journal of Allergy and Clinical Immunology In Practice · doi:10.1016/j.jaip.2025.09.012 · Cited by 10
AI for all: bridging data gaps in machine learning and health
Wang, M. L., & Bertrand, K. A. (2025) · Translational Behavioral Medicine · doi:10.1093/tbm/ibae075 · Cited by 10
Integrating New Technologies in Lipidology: A Comprehensive Review
Escobar, C., García, J. S., Torremocha, A., Lorenzo, C. C., Lara-García, A., Muñoz, L. C., González, R. M., García, J. V., & Banach, M. (2025) · Journal of Clinical Medicine · doi:10.3390/jcm14144984 · Cited by 10
Sridhar, D., Manikandan, R., Dhanapal, Y., Khute, S., & Subash, P. (2025) · Intelligent Pharmacy · doi:10.1016/j.ipha.2025.08.001 · Cited by 12
Destination (Un)Known: Auditing Bias and Fairness in LLM-Based Travel Recommendations
Andreev, H., Kosmas, P., Livieratos, A. D., Theocharous, A. L., & Zopiatis, A. (2025) · AI · doi:10.3390/ai6090236 · Cited by 11
“It includes sampling bias, where some groups are more likely to be included than others [27,29,30]; labelling bias, where annotator judgements alter ground truth [27] and statistical bias, where historical patterns embed pre-existing inequalities that the model reproduces [27,31].”
Artificial Intelligence and Health Equity
Bright, T. J., & Norris, K. C. (2025) · Annual Review of Medicine · doi:10.1146/annurev-med-043024-125309 · Cited by 8
Samra, R. M. A., & Alí, R. R. (2026) · AgriEngineering · doi:10.3390/agriengineering8070270 · Cited by 8
“AI models may exhibit bias due to imbalanced datasets or regional variability [77].”
Death risk prediction model for patients with non-traumatic intracerebral hemorrhage
Chen, Y., Liu, X., Yan, M., & Wan, Y. (2025) · BMC Medical Informatics and Decision Making · doi:10.1186/s12911-025-02865-4 · Cited by 8
Fair streaming feature selection
Duan, Z., Li, T., Ling, Z., Wu, X., Yang, J., & Jia, Z. (2025) · Neurocomputing · doi:10.1016/j.neucom.2025.129394 · Cited by 7
Managing Transparency Fairness Accountability in AI for Sustainable Human Resource Management
Setiawati, R., Kusmara, A., & Kuusk, T. (2025) · International Conference on Creative Communication and Innovative Technology (ICCIT) · doi:10.1109/iccit65724.2025.11166968 · Cited by 9
“While they enhance efficiency and reduce human error, they also pose risks of algorithmic, implicit, and selection bias if trained on flawed or unrepresentative data [29].”
Hassoon, A., Lin, C. F., Woo, H., Irimia, R. M., Marsteller, J. A., Li, A., Banderia, A., Leo, H., Peng, X., Rastall, D. P. W., et al. (2025) · Annals of Epidemiology · doi:10.1016/j.annepidem.2025.11.004 · Cited by 8
A community-based approach to ethical decision-making in artificial intelligence for health care
Senghor, A. S., Bright, T. J., Kakim, S., Norris, K. C., Antwi, H. A., Cooper, J. K., Mullins, C. D., & Baquet, C. R. (2025) · JAMIA Open · doi:10.1093/jamiaopen/ooaf076 · Cited by 6
“9,14 Additionally, healthcare organizations must critically assess how biases in clinical data influence AI predictions, ensuring that models do not reinforce existing disparities in care delivery.”
The value alignment problem in advisory AI: a systematic literature review
Triantafyllopoulos, L., Paxinou, E., Tzanoulinou, D., Verykios, V. S., & Kalles, D. (2026) · AI and Ethics · doi:10.1007/s43681-026-01015-4 · Cited by 6
“This shift reflects a tran - sition from primarily conceptual discussions of alignment toward technically oriented experimentation with LLMs and alignment methods such as RLHF [39, 97] and DPO [10–12, 28, 89].”
Explainable artificial intelligence for multi-modal cancer analysis: from genomics to immunology
Zhang, M., & Yin, L. (2026) · Critical Reviews in Oncology/Hematology · doi:10.1016/j.critrevonc.2025.105040 · Cited by 5
Dantas, C., Cabrita, M., Midão, L., Carvalho, A. S., & Costa, E. M. (2025) · AI and Ethics · doi:10.1007/s43681-025-00777-7 · Cited by 5
“…achieve a broad standpoint of analy - sis and a holistic knowledge of the policy, technological, research and user perspectives needed to implement real - istic solutions, adjusted to the citizens’ needs and fit to be integrated into national and international policies and work - flows [ 17 , 18].”
Navigating Fairness in AI-based Prediction Models: Theoretical Constructs and Practical ApplicationsPreprint
van der Meijden, S. L., Wang, Y., Arbous, M. S., Geerts, B. F., Steyerberg, E. W., & Hernandez‐Boussard, T. (2025) · medRxiv · doi:10.1101/2025.03.24.25324500 · Cited by 5
“However, group fairness requires high-quality data on protected attributes like race and ethnicity [15, 16].”
The Elusive B Cell Antigen in Multiple Sclerosis: Time to Rethink CNS B Cell Functions
Mailaender, F., Vasilenko, N., Tieck, M. P., Schembecker, S., & Kowarik, M. C. (2025) · International Journal of Molecular Sciences · doi:10.3390/ijms262110771 · Cited by 5
Through a Teacher’s Lens: Combating Bias in AI-Powered Education for a Just Future
Poornesh, M. (2024) · The Clearing House A Journal of Educational Strategies Issues and Ideas · doi:10.1080/00098655.2024.2393153 · Cited by 5
Bridging fair-aware artificial intelligence and co-creation for equitable mental healthcare
Timmons, A. C., Duong, J. B., Walters, S. N., Carta, K. E., Jumonville, G. A., Carrasco, A. S., Romero, D., Ahle, M. W., Comer, J. S., Khurd, I. P., et al. (2025) · Nature Reviews Psychology · doi:10.1038/s44159-025-00491-5 · Cited by 7
“With the rapid adoption of AI-based mental health interventions, frameworks that ensure equitable and effective outcomes are urgently needed 16,21–24 (Fig.”
Information Distortion in Electronic Health Records: A Concept Analysis
Wang, J., Xu, Y., Zhang, R., Yang, Z., Zhang, X., & Pan, H. (2025) · Journal of Advanced Nursing · doi:10.1111/jan.70233 · Cited by 3
Mitigating Ethical Issues for Large Language Models in Oncology: A Systematic Review
Zhou, S., Liu, X., Xu, Z., Zhan, Z., Song, M., Wang, J., Liu, S., Xu, H., & Zhang, R. (2025) · JCO Clinical Cancer Informatics · doi:10.1200/cci-25-00076 · Cited by 5
Raising awareness of potential biases in medical machine learning: Experience from a Datathon
Hochheiser, H., Klug, J., Mathie, T., Pollard, T., Raffa, J. D., Ballard, S. L., Conrad, E. A., Edakalavan, S., Joseph, A. M., Alnomasy, N., et al. (2025) · PLOS Digital Health · doi:10.1371/journal.pdig.0000932 · Cited by 4
Clinical Applications of Artificial Intelligence in Neurology Practice
Hadar, P., & Moura, L. M. V. R. (2025) · CONTINUUM Lifelong Learning in Neurology · doi:10.1212/cont.0000000000001552 · Cited by 4
Automating clinical phenotyping using natural language processing
Schmidt, L., Ibing, S., Borchert, F., Hugo, J., Marshall, A. A., Peraza, J., Cho, J. H., Böttinger, E. P., Renard, B. Y., & Ungaro, R. C. (2026) · Communications Medicine · doi:10.1038/s43856-025-01337-0 · Cited by 4
CLE-SH: Comprehensive Literal Explanation Package for SHapley Values by Statistical Validity
Kim, K., Lee, Y., & Seo, J. (2026) · IEEE Access · doi:10.1109/access.2026.3654890 · Cited by 4
“In these areas, datasets are often vulnerable to data bias, meaning high performance alone is not enough to validate the use of machine learning [1, 2, 3].”
Masison, J., Lehmann, H. P., & Wan, J. (2024) · Journal of Investigative Dermatology · doi:10.1016/j.jid.2024.08.025 · Cited by 3
Ethical Artificial Intelligence and Machine Learning in Self-Service Technologies
Ghosh, R., Çela, E., Chakraborty, S., Gupta, B., Dutta, G., Chakraborty, S., & Roy, S. (2025) · IGI Global eBooks · doi:10.4018/979-8-3373-1295-8.ch007 · Cited by 3
Akinniyi, O., Dixon, J., Aina, J., Weaks, F., Saleh, G. A., Rahman, M. M., Meeker, T. J., Trivedi, H., Gichoya, J. W., & Khalifa, F. (2025) · Computer Methods and Programs in Biomedicine · doi:10.1016/j.cmpb.2025.109036 · Cited by 3
Digital Mental Health Through an Intersectional Lens: A Narrative Review
Yesha, R., Orezzoli, M. C. E., Sims, K., & Landau, A. Y. (2026) · Healthcare · doi:10.3390/healthcare14020211 · Cited by 4
“Several forms of bias can arise throughout the lifecycle of AI systems, including during data collection, clinical documentation, model design, and implementation, which perpetuate existing disparities in mental health care [34–37].”
Hooshmandi, S., Tabatabaii, M., Arani, H. Z., Heidarian, A., & Nazemi, S. (2026) · Discover Artificial Intelligence · doi:10.1007/s44163-026-00843-x · Cited by 3
Addressing Bias and Ensuring Fairness in AI Systems for Healthcare
Zarnegar, A. (2025) · Health Informatics and Knowledge Management Conference (HIKM) · doi:10.1145/3774816.3774817 · Cited by 3
Mitigating Algorithmic Bias in Cancer Site Classification Models
Shivanna, A., Spannaus, A., Tschida, J., Gounley, J. P., Krawczuk, P., & Hanson, H. A. (2026) · JCO Clinical Cancer Informatics · doi:10.1200/cci-25-00250 · Cited by 3
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 5
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 4
“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].”
Kansal, M., Sibal, R., & Ram, P. (2024) · Communications in computer and information science · doi:10.1007/978-3-031-77029-6_8 · Cited by 2
Applying Artificial Neural Networks to Estimate PMI: A New Era of Accuracy in Forensic Science
Ouanes, K. (2025) · Springer eBooks · doi:10.1007/978-981-96-4585-5_11 · Cited by 2
Efthymiou, E., Katsarou, D. V., Sofologi, M., Megari, K., Papadopoulou, S. L., Mantsos, E., Argyriadis, A., & Argyriadi, A. (2025) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3373-0573-8.ch009 · Cited by 2
Nejadshamsi, S., Chu, C. H., McGilton, K. S., Li, X., Ronquillo, C., & Rahimi, S. A. (2025) · JAMIA Open · doi:10.1093/jamiaopen/ooaf171 · Cited by 2
“Chen et al. 26 investigated the impact of preprocessing methods such as reweighting and synthetic oversampling, which were shown to improve fairness in electronic health record (EHR)-based models without significantly compromising predictive accuracy.”
Abouelmagd, K., Saleem, N., Kuhn, K., Oyedele, T. J., Abdelbar, S. M. M., Rath, S., & Alsabri, M. (2025) · Current Emergency and Hospital Medicine Reports · doi:10.1007/s40138-025-00334-w · Cited by 2
“The majority of existing AI models are trained on data from adult populations and high-income countries (HICs), severely limiting their generalizability to the unique anatomical and physiological characteristics of children, particularly neonates and infants [50, 51].”
Bouterse, A., Pruneski, J. A., Oettl, F. C., Zsidai, B., Tischer, T., Longo, U. G., Seil, R., Hirschmann, M. T., & Samuelsson, K. (2026) · Knee Surgery Sports Traumatology Arthroscopy · doi:10.1002/ksa.70339 · Cited by 2
“Additionally, the integrity of AI models is dependent on the data they are trained on, allowing imbalanced or incomplete datasets to introduce algorithmic bias into decision ‐ making models that in fl uence aspects of clinical care [15, 17, 36, 48, 53, 73].”
Lim, Y. X., Ming, L. C., Ng, N. C., Leow, S., & Ingle, R. G. (2026) · Discover Computing · doi:10.1007/s10791-026-10205-x · Cited by 3
“Where imbalances exist, technical fixes can be applied, such as oversampling underrepresented groups, undersampling majority groups, reweighting data points to give more importance to minority samples, or using data augmentation to synthetically create more data for sparse classes [79, 81, 82] [–].”
The BEETS framework for responsible artificial intelligence
Tettegah, S. (2026) · Discover Artificial Intelligence · doi:10.1007/s44163-026-01597-2 · Cited by 3
“The literature on artificial intelligence (AI) often focuses on key themes such as security, ethics, equity, bias, trust, and fairness [11, 19].”
Zhao, Q., Jiang, W., Zhang, C., Ouyang, M., Wu, N., & Guo, Y. (2026) · Frontiers in Public Health · doi:10.3389/fpubh.2026.1863050 · Cited by 3
“This missing pattern reflects structural inequalities such as digital access barriers or specific employment systems, which go beyond the scope of random technical failures (83, 84).”
Bashir, S., Salad, H. A., Abdullahi, Y. B., Abdi, Y. H., Abdi, M. S., Ahmed, N., Musa, S. S., Elehamer, N. M. K., Musa, M. K., Bolarinwa, O. A., et al. (2026) · Frontiers in Digital Health · doi:10.3389/fdgth.2026.1841706 · Cited by 4
Accuracy of administrative data in ascertaining health conditions: a systematic review
Campbell, A. C., Tyler, J., Shuttleworth, R. R., Pearce, L. A., Heerde, J. A., Borschmann, R., Sawyer, S. M., Preen, D. B., Kinner, S. A., & Calais‐Ferreira, L. (2026) · JAMIA Open · doi:10.1093/jamiaopen/ooag109 · Cited by 2
“52 When the underlying codes have variable accuracy, particularly for sensitivity and PPV, these models risk inheriting and amplifying biases in clinical predictions and resource allocation that are difficult to detect and correct once embedded in automated processes.”
AI in the Team Meeting Landscape
Allen, J. A., Grabowski, M., Thiese, M. S., & Eden, E. (2026) · Emerald Group Publishing Limited eBooks · doi:10.1108/s1534-085620260000021010 · Cited by 2
Rudroff, T. (2025) · Infectious Disease Reports · doi:10.3390/idr17020033 · Cited by 2
“, and metabolic indicators, to generate comprehensive risk assessments [59].”
Assessing the risk of discriminatory bias in classification datasets
Dai, K., Kim, J., Džeroski, S., Wicker, J. S., Dobbie, G., & Dost, K. (2025) · Machine Learning · doi:10.1007/s10994-025-06843-9 · Cited by 2
“This approach is particularly problematic if the practitioner is unaware of the bias; even more so in numeric datasets such as in healthcare data, where patients are defined by their age and numeric lab results rather than their gender or race (Chen et al., 2024).”
Narasannagari, L., & Bhardwaj, D. S. (2025) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3693-9521-9.ch003 · Cited by 2
Moving forward on the science of informatics and predictive analytics
Henry, S. B. (2024) · Journal of the American Medical Informatics Association · doi:10.1093/jamia/ocae077 · Cited by 2
AI-Powered clinical assessments: GPT-4o’s role in standardizing CPR skill evaluations
Semeraro, F. F. (2024) · Resuscitation · doi:10.1016/j.resuscitation.2024.110411 · Cited by 2
Frasier, K. M., Hash, M. G., & Pugliese, A. (2025) · Cureus · doi:10.7759/cureus.79109 · Cited by 1
“This is often due to biases in the training data or differences between the model’s training population and real-world application, underscoring the need for further refinement in predictive algorithms [16].”
Demiral, Y. E., & Alsahag, A. M. M. (2025) · Lecture notes in computer science · doi:10.1007/978-981-96-6389-7_32 · Cited by 1
Muñoz-García, V., Consuegra‐Ayala, J. P., & Moreda, P. (2025) · Knowledge-Based Systems · doi:10.1016/j.knosys.2025.113915 · Cited by 1
Emerging Trends in Early Dementia Diagnosis: An Analysis on Advanced Machine Learning Approaches
Gami, B., Agrawal, M., & Katarya, R. (2025) · ACM Computing Surveys · doi:10.1145/3764578 · Cited by 1
“Similarly, Chen et al. [16] suggested systematic bias detection and mitigation techniques.”
Pelka, O., Sigle, S., Werner, P., Schweizer, S. T., Iancu, A., Scherer, L., Kamzol, N. A., Eil, J. H., Apfelbacher, T., Seletkov, D., et al. (2025) · RöFo - Fortschritte auf dem Gebiet der Röntgenstrahlen und der bildgebenden Verfahren · doi:10.1055/a-2651-6653 · Cited by 1
Translating Features to Findings: Deep Learning for Melanoma Subtype Prediction
Guermazi, D., Khemchandani, S., Wahood, S., Nguyen, C. V., & Saliba, E. (2025) · Dermatopathology · doi:10.3390/dermatopathology12040042 · Cited by 1
Lee, J. T., Hsu, S. H., Li, V. C.-S., Anindya, K., Chen, M. M., Wang, C. I., Shen, T. K.-B., Liu, V. T. N., Chen, H.-H., & Atun, R. A. (2025) · PLOS Digital Health · doi:10.1371/journal.pdig.0000951 · Cited by 1
“Addressing these gaps requires fairness-enhancing techniques, such as the Disparate Impact Remover, to ensure consistent model accuracy across diverse groups [19,21].”
InfluentialBalancing Innovation and Responsibility: Ethical and Privacy Challenges in Stroke Digital Health
Muchada, M., Rizzo, F., Brunelli, N., & Molina, C. A. (2025) · Stroke · doi:10.1161/strokeaha.125.050450 · Cited by 1
Golovenko, M. Y., & Larionov, V. (2025) · Artificial Intelligence · doi:10.15407/jai2025.04.010 · Cited by 2
A Meta-Analytic NLP-Driven Mapping of Fairness in Healthcare AI
Tsoni, R., Koutouzis, M., & Verykios, V. S. (2025) · International Conference on Information, Intelligence, Systems & Applications (IISA) · doi:10.1109/iisa66859.2025.11311222 · Cited by 1
Alzahrani, F., & Alnezary, F. S. (2026) · Frontiers in Public Health · doi:10.3389/fpubh.2026.1763446 · Cited by 1
“This ongoing vigilance supports a dynamic, adaptive approach to equity within organizational practice (64, 65).”
Qu, H., Li, J., Gao, Y., Xu, X.-M., Zhang, X.-B., & Yang, S.-D. (2026) · Frontiers in Artificial Intelligence · doi:10.3389/frai.2026.1807400 · Cited by 1
AI for TCM: From target identification to drug design
Wang, M., Wang, W., Shen, M., Fan, X., & Liao, J. (2026) · Journal of Pharmaceutical Analysis · doi:10.1016/j.jpha.2026.101681 · Cited by 1
Metadata
- Authors
- Chen, F., Wang, L., Hong, J., Jiang, J., & Zhou, L.
- Affiliations
- Harvard University (US)
- Brigham and Women's Hospital (US)
- Wellesley High School , Wellesley, MA 02481, United States
- Corresponding author
- Chen, Feng
- Venue
- Journal of the American Medical Informatics Association
- Publisher
- Oxford University Press (OUP)
- Published
- 2024-03-23
- Volume
- 31
- Issue
- 5
- Pages
- 1172-1183
- ISSN
- 1067-5027 (ISSN-L)
- 1527-974X
- PMC
- PMC11031231
- OpenAlex
- W4393119757
- Cited by
- 178
- References
- 48
- Citations by year
- 2026: 72 · 2025: 91 · 2024: 15
- FWCI
- 13.55
- Type
- journal article
- Language
- en
- Open access
- green
- www.ncbi.nlm.nih.gov
- Locations
- Journal of the American Medical Informatics Association (publishedVersion)
- PubMed Central (publishedVersion)
- PubMed (publishedVersion)
- arXiv (Cornell University) (submittedVersion, cc-by)
- arXiv (Cornell University) (cc-by)
- Funders
- National Library of Medicine (1R01LM014239)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Ethics and Social Impacts of AI
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning in Healthcare
- Keywords
- algorithmic bias, bias detection, bias mitigation, electronic health records, artificial intelligence in healthcare, health care disparities, fairness metrics, implicit bias, selection bias, confounding, measurement bias, temporal bias, resampling, reweighting, ethical AI
- MeSH terms
- Algorithms, Artificial Intelligence, Female, Humans, Models, Theoretical, Pregnancy, Bias, Selection Bias, Benchmarking, Electronic Health Records
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.
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
