Generative pre-trained transformer reinforces historical gender bias in diagnosing women’s cardiovascular symptoms
GPT-4o underrepresented women in simulated cardiovascular scenarios and less accurately diagnosed female patients with critical conditions. These biases risk reinforcing historical disparities in cardiovascular care. Future efforts should focus on bias detection and mitigation.
Authors
- Mohamed Rahouma (ORCID: https://orcid.org/0000-0002-8678-2060)
- Giovanni Soletti (ORCID: https://orcid.org/0000-0002-8691-5951)
- Katherine Krieger (ORCID: https://orcid.org/0009-0002-8706-9977)
- Irbaz Hameed (ORCID: https://orcid.org/0000-0003-4954-4685)
- Charles A. Mack (ORCID: https://orcid.org/0000-0001-8142-9818)
- Miodrag Savic (ORCID: https://orcid.org/0000-0002-0211-8868)
- Aina Hirofuji
- Polina Mantaj
- Alexander Gregg
- Giorgio Quer (ORCID: https://orcid.org/0000-0003-2208-7912)
- Camilla S Rossi
- Mario Gaudino
Institutions
- Scripps Research Institute (US)
- Medical University of Graz (AT)
- Cornell University (US)
- Yale University (US)
- Kepler Universitätsklinikum (AT)
- Weill Cornell Medicine (US)
Publication Details
- Journal
- European Heart Journal - Digital Health
- Published
- 2025-11-06
- DOI
- https://doi.org/10.1093/ehjdh/ztaf131
- Citations
- 1
- Primary Topic
- Acute Myocardial Infarction Research
- Type
- article
- Field-Weighted Citation Impact
- 1.39