event-icon
Description

Decision-making in health is complex, especially when considering individual differences. Machine learning (ML) and patient generated data can identify patterns and insights at the individual level, where human cognition falls short. We applied attributable components analysis (ACA) to type 2 diabetes self-monitoring data to identify patterns of association between nutrition and blood glucose control, and found that ACA offers more robust uncertainty estimates than linear regression. We discuss the implications for ML-driven decision support systems.

Learning Objective: Understand the challenges in applying machine learning (ML) to patient-generated data, specifically considering applications for ML-driven clinical decisions support systems.

Authors:

Elliot Mitchell (Presenter)
Columbia University

Lena Mamykina, Columbia University
Matthew Levine, California Institute of Technology
Esteban Tabak, Courant Institute of Mathematical Sciences
David Albers, University of Colorado

Presentation Materials:

Tags