3D Segmentation Networks for Excessive Numbers of Classes: Distinct Bone Segmentation in Upper Bodies

TL;DR

This paper presents a novel 3D segmentation network capable of automatically distinguishing over 125 different bones in CT scans, addressing challenges of many-label segmentation in medical imaging.

Contribution

It introduces specific modifications to network architecture, loss functions, and data augmentation techniques for effective many-label 3D segmentation.

Findings

Successfully segments over 125 bones simultaneously

Demonstrates robustness in complex many-label segmentation tasks

Provides insights into training deep networks with extensive label sets

Abstract

Segmentation of distinct bones plays a crucial role in diagnosis, planning, navigation, and the assessment of bone metastasis. It supplies semantic knowledge to visualisation tools for the planning of surgical interventions and the education of health professionals. Fully supervised segmentation of 3D data using Deep Learning methods has been extensively studied for many tasks but is usually restricted to distinguishing only a handful of classes. With 125 distinct bones, our case includes many more labels than typical 3D segmentation tasks. For this reason, the direct adaptation of most established methods is not possible. This paper discusses the intricacies of training a 3D segmentation network in a many-label setting and shows necessary modifications in network architecture, loss function, and data augmentation. As a result, we demonstrate the robustness of our method by automatically segmenting over one hundred distinct bones simultaneously in an end-to-end learnt fashion from a CT-scan.