Paper on inherently interpretable multi-label classifiers using counterfactuals accepted to MIDL 2023!

Our research on using class-specific counterfactuals to create an inherently interpretable multi-label classifier led by MLMIA lab member Susu Sun has been accepted to the Medical Imaging with Deep Learning (MIDL) conference, which will be held in Nashville, Tennessee this year.

A pre-print can be found here: arxiv.org/abs/2303.00500

More information can be found in this Twitter thread.