Runtime freezing: dynamic class loss for multi-organ 3D segmentation

Conference item

Segmentation has become a crucial pre-processing step to many refined downstream tasks, and particularly so in the medical domain. Even with recent improvements in segmentation models, many segmentation tasks remain difficult. When multiple organs are segmented simultaneously, difficulties are due not only to the limited availability of labelled data, but also to class imbalance. In this work we propose dynamic class-based loss strategies to mitigate the effects of highly imbalanced training data. We show how our approach improves segmentation performance on a challenging Multi-Class 3D Abdominal Organ dataset.

Authors: Willoughby, J; Voiculescu, I

• Publication status: Published
• Peer review status: Peer reviewed
• Publisher: IEEE
• Host title: Proceedings of the 21st IEEE International Symposium on Biomedical Imaging (ISBI 2024)
• Pages: 1-4
• Publication date: 2024-08-22
• Acceptance date: 2024-02-02
• Event title: 21st IEEE International Symposium on Biomedical Imaging (ISBI 2024)
• Event location: Athens, Greece
• Event website: https://biomedicalimaging.org/2024/
• Event start date: 2024-05-27
• Event end date: 2024-05-30
• DOI: 10.1109/ISBI56570.2024.10635571
• EISSN: 1945-8452
• ISSN: 1945-7928
• EISBN: 979-8-3503-1333-8
• ISBN: 979-8-3503-1334-5
• Language: English