Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox

Are we using appropriate segmentation metrics? Identifying correlates of human expert perception for CNN training beyond rolling the DICE coefficient

Mar 10, 2021
Florian Kofler, Ivan Ezhov, Fabian Isensee, Fabian Balsiger, Christoph Berger, Maximilian Koerner, Johannes Paetzold, Hongwei Li, Suprosanna Shit, Richard McKinley, Spyridon Bakas, Claus Zimmer, Donna Ankerst, Jan Kirschke, Benedikt Wiestler, Bjoern H. Menze

Share this with someone who'll enjoy it:

In this study, we explore quantitative correlates of qualitative human expert perception. We discover that current quality metrics and loss functions, considered for biomedical image segmentation tasks, correlate moderately with segmentation quality assessment by experts, especially for small yet clinically relevant structures, such as the enhancing tumor in brain glioma. We propose a method employing classical statistics and experimental psychology to create complementary compound loss functions for modern deep learning methods, towards achieving a better fit with human quality assessment. When training a CNN for delineating adult brain tumor in MR images, all four proposed loss candidates outperform the established baselines on the clinically important and hardest to segment enhancing tumor label, while maintaining performance for other label channels.

   Access Paper Source

Share this with someone who'll enjoy it: