Skip to Main content Skip to Navigation
Conference papers

Creating Fair Models of Atherosclerotic Cardiovascular Disease Risk

Abstract : Guidelines for the management of atherosclerotic cardiovascular disease (ASCVD) recommend the use of risk stratification models to identify patients most likely to benefit from cholesterol-lowering and other therapies. These models have differential performance across race and gender groups with inconsistent behavior across studies, potentially resulting in an inequitable distribution of beneficial therapy. In this work, we leverage adversarial learning and a large observational cohort extracted from electronic health records (EHRs) to develop a "fair" ASCVD risk prediction model with reduced variability in error rates across groups. We empirically demonstrate that our approach is capable of aligning the distribution of risk predictions conditioned on the outcome across several groups simultaneously for models built from high-dimensional EHR data. We also discuss the relevance of these results in the context of the empirical trade-off between fairness and model performance.
Complete list of metadatas

https://hal.inria.fr/hal-02388730
Contributor : Adrien Coulet <>
Submitted on : Monday, December 2, 2019 - 9:58:55 AM
Last modification on : Thursday, May 14, 2020 - 11:38:32 AM

Links full text

Identifiers

Citation

Stephen Pfohl, Ben Marafino, Adrien Coulet, Fatima Rodriguez, Latha Palaniappan, et al.. Creating Fair Models of Atherosclerotic Cardiovascular Disease Risk. AIES '19 - Proceedings of the 2019 AAAI/ACM Conference on AI, Ethics, and Society, Jan 2019, Honolulu, United States. pp.271-278, ⟨10.1145/3306618.3314278⟩. ⟨hal-02388730⟩

Share

Metrics

Record views

66