Integration of fairness-awareness into clinical language processing models
This study demonstrates that fairness can be integrated into clinical language models, though effects vary by model type. Architectures aligned with clinical text structure inherently promote fairness, yet mixed fairness constraint outcomes highlight the need for tailored interventions. Persistent demographic disparities show that algorithmic bias often reflects upstream documentation inequities. This framework offers a scalable path toward equitable NLP for clinical artificial intelligence.
Authors
- Soha Ahmadi (ORCID: https://orcid.org/0000-0002-7182-9049)
- Rawan Abulibdeh (ORCID: https://orcid.org/0009-0008-1179-4201)
- Ervin null Sejdić
- Yihang Lin
- Karen Tu
- Qiuyi Zhao
- Leo Anthony Celi
Institutions
- Beth Israel Deaconess Medical Center (US)
- University Health Network (CA)
- Harvard University (US)
- University of Toronto (CA)
- Toronto Western Hospital (CA)
- North York General Hospital (CA)
- Hospital for Sick Children (CA)
- Massachusetts Institute of Technology (US)
Publication Details
- Journal
- Communications Medicine
- Published
- 2026-02-24
- DOI
- https://doi.org/10.1038/s43856-026-01433-9
- Primary Topic
- Artificial Intelligence in Healthcare and Education
- Type
- article
- Field-Weighted Citation Impact
- 0.00