PhiSeg - Segmentation on medical images - model presented at MICCAI 2019

Segmentation can be a difficult task on medical images as is shown by the fact that even experts disagree to some extent over where the boundaries for different tissue begin.

by Kristine Haberer

Dr. Baumgartner, former CVL member, explored this topic with other colleagues and proposed a probabilistic hierarchical model that allows to sample from the distribution of segmentations. This work facilitates principled uncertainty estimation in medical image segmentation and has been presented at MICCAI 2019.

external pagePaper
external pageCode


 

JavaScript has been disabled in your browser