Automatic segmentation of the pulmonary lobes with a 3D u-net and optimized loss function

Bianca Lassen-Schmidt, Alessa Hering, Stefan Krass, Hans Meine

Keywords: pulmonary lobes, lung lobes, segmentation, deep learning, CNN, 3D U-net

TL;DR: A weighted loss function improves pulmonary lobe segmentation with a u-net.

Abstract: Fully-automatic lung lobe segmentation is challenging due to anatomical variations, pathologies, and incomplete fissures. We trained a 3D u-net for pulmonary lobe segmentation on 49 mainly publically available datasets and introduced a weighted Dice loss function to emphasize the lobar boundaries. To validate the performance of the proposed method we compared the results to two other methods. The new loss function improved the mean distance to 1.46 mm (compared to 2.08 mm for simple loss function without weighting).

Track: short paper

Paper Type: methodological development

Presentation Upload Agreement: I agree that my presentation material (videos and slides) will be made public.