Presentation at INFORMS Healthcare 2023 on our work on rank-based compatibility. You can find a link to the post about the upcoming paper here.
View a copy of the presentation slides below.
Link to download presentation.
A recording of this presentation can be found here.
Abstract
Updating clinical machine learning models is necessary to maintain performance, but may cause compatibility issues, affecting user-model interaction. Current compatibility measures have limitations, especially where models generate risk-based rankings. We propose a new rank-based compatibility measure and loss function that optimizes discriminative performance while promoting good compatibility. We applied this to a mortality risk stratification study using MIMIC data, resulting in more compatible models while maintaining performance. These techniques provide new approaches for updating risk stratification models in clinical settings.
