Sylvain Faisan

Calibration methods that ignore logit shifts can lead to misleading confidence estimates, but translation-invariant calibrators provide a solution that enhances reliability in semantic segmentation.

Seemingly strong segmentation models can fail at clinically critical tumor-vessel interfaces, highlighting the need for uncertainty-aware AI in pancreatic cancer staging.

TwinTrack transforms ambiguous medical image segmentation by calibrating model outputs to reflect the true consensus of expert annotators.