Auditing Learned Associations in Deep Learning Approaches to Extract Race and Ethnicity from Clinical Text

We present an approach to audit the learned associations of models trained to identify RE information in clinical text by measuring the concordance between model-derived salient features and manually identified RE-related spans of text. We show that while models perform well on the surface, there exist concerning learned associations and potential for future harms from RE-identification models if left unaddressed.PMID:38222422 | PMC:PMC10785932

Source: AMIA Annual Symposium Proceedings - January 15, 2024 Category: Bioinformatics Authors: Oliver J Bear Don't Walk Iv Adrienne Pichon Harry Reyes Nieva Tony Sun Jaan Altosaar Karthik Natarajan Adler Perotte Peter Tarczy-Hornoch Dina Demner-Fushman No émie Elhadad Source Type: research