Multimodal HIE Lesion Segmentation in Neonates: A Comparative Study of Loss Functions

TL;DR

This study compares various loss functions for neonatal HIE lesion segmentation in MRI, demonstrating that combining region-based and boundary-aware losses improves accuracy with limited data.

Contribution

It introduces and evaluates compound loss functions tailored for HIE lesion segmentation, showing their superiority over standalone losses.

Findings

Compound losses outperform standalone loss functions.

Tversky-HausdorffDT Loss achieves highest Dice scores.

Dice-Focal-HausdorffDT minimizes Mean Surface Distance.

Abstract

Segmentation of Hypoxic-Ischemic Encephalopathy (HIE) lesions in neonatal MRI is a crucial but challenging task due to diffuse multifocal lesions with varying volumes and the limited availability of annotated HIE lesion datasets. Using the BONBID-HIE dataset, we implemented a 3D U-Net with optimized preprocessing, augmentation, and training strategies to overcome data constraints. The goal of this study is to identify the optimal loss function specifically for the HIE lesion segmentation task. To this end, we evaluated various loss functions, including Dice, Dice-Focal, Tversky, Hausdorff Distance (HausdorffDT) Loss, and two proposed compound losses -- Dice-Focal-HausdorffDT and Tversky-HausdorffDT -- to enhance segmentation performance. The results show that different loss functions predict distinct segmentation masks, with compound losses outperforming standalone losses. Tversky-HausdorffDT Loss achieves the highest Dice and Normalized Surface Dice scores, while Dice-Focal-HausdorffDT Loss minimizes Mean Surface Distance. This work underscores the significance of task-specific loss function optimization, demonstrating that combining region-based and boundary-aware losses leads to more accurate HIE lesion segmentation, even with limited training data.