
Medicine and Health
Peeking into a black box, the fairness and generalizability of a MIMIC-III benchmarking model
E. Röösli, S. Bozkurt, et al.
This study explores critical issues of bias and fairness in a MIMIC-III trained model for predicting in-hospital mortality, revealing significant concerns for Black and publicly insured ICU patients. Conducted by Eliane Röösli, Selen Bozkurt, and Tina Hernandez-Boussard, this research emphasizes the necessity for robust fairness assessments in healthcare model evaluations.
Playback language: English
Related Publications
Explore these studies to deepen your understanding of the subject.