Sex-specific machine learning classification models improve outcome prediction for abdominal aortic aneurysms
ML classification models show promise in improving predictions of patient outcomes for AAA. The higher AAA prevalence rate for males leads to female patients being underrepresented in AAA datasets. In this proof-of-concept study, we demonstrated that sex-specific models outperformed a general model in predicting patient outcomes. Additionally, equalizing sample sizes within the dataset improved predictions for female patients without compromising overall performance of the model. As ML applications in medicine continue to grow, it is important to consider population representation within datasets to reduce model bias.
Authors
- Nathan L. Liang (ORCID: https://orcid.org/0000-0003-2863-4008)
- Pete H. Gueldner (ORCID: https://orcid.org/0000-0002-9177-8162)
- David A. Vorp (ORCID: https://orcid.org/0000-0002-6467-2151)
- Tiziano Tallarita (ORCID: https://orcid.org/0000-0001-6262-0246)
- Katherine E. Kerr (ORCID: https://orcid.org/0000-0003-1238-6564)
- Timothy K. Chung
- Indrani Sen
- Joseph C. Wildenberg (ORCID: https://orcid.org/0000-0001-5489-8467)
Institutions
- University of Pittsburgh (US)
- Magee-Womens Research Institute (US)
- Mayo Clinic Health System (US)
- University of Pittsburgh Medical Center (US)
- McGowan Institute for Regenerative Medicine (US)
- The University of Texas at San Antonio (US)
Publication Details
- Journal
- Biology of Sex Differences
- Published
- 2025-11-11
- DOI
- https://doi.org/10.1186/s13293-025-00765-w
- Primary Topic
- Aortic aneurysm repair treatments
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- National Science Foundation
- American Heart Association
- Nvidia
- University of Pittsburgh
- Jewish Healthcare Foundation
- National Institutes of Health