A data-centric approach to detecting and mitigating demographic bias in pediatric mental health text
We develop and evaluate a data-centric de-biasing framework to address gender-based disparities in clinical text arising from non-biological differences, such as reporting practices and documentation styles. Our method selectively de-biases data by neutralizing biased language and normalizing information density while preserving clinically relevant content. Further validation across different models is essential before clinical deployment.
Authors
- Sanghyun Choo (ORCID: https://orcid.org/0000-0002-8884-3437)
- Paulina Bondaronek (ORCID: https://orcid.org/0000-0003-0096-1234)
- Greeshma Agasthya (ORCID: https://orcid.org/0000-0002-2301-5964)
- D. Santel
- Jordan Tschida (ORCID: https://orcid.org/0000-0002-1359-4165)
- Julia Ive
- Anuj Kapadia
- John Pestian
- Mayanka Chandrashekar
- Tracy Glauser
- Vishal Yadav
Institutions
- Kumoh National Institute of Technology (KR)
- Cincinnati Children's Hospital Medical Center (US)
- Oak Ridge National Laboratory (US)
- Georgia Institute of Technology (US)
- Queen Mary University of London (GB)
- Health Data Research UK (GB)
- University College London (GB)
- University of Cincinnati (US)
- University of Cincinnati Medical Center (US)
Publication Details
- Journal
- Communications Medicine
- Published
- 2026-03-05
- DOI
- https://doi.org/10.1038/s43856-026-01480-2
- Primary Topic
- Mental Health via Writing
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- U.S. Department of Energy
- Battelle
- Cincinnati Children's Hospital Medical Center
- UT-Battelle