Generalized Wasserstein Dice Loss, Test-time Augmentation, and Transformers for the BraTS 2021 challenge

TL;DR

This paper presents a robust ensemble approach using diverse models, a generalized Wasserstein Dice loss, and test-time augmentation to improve brain tumor segmentation accuracy across different MRI scanners in the BraTS 2021 challenge.

Contribution

It introduces a novel ensemble strategy, incorporates transformers into U-Net, and demonstrates the effectiveness of a generalized Wasserstein Dice loss for brain tumor segmentation.

Findings

Ensemble of seven 3D U-Nets achieves 89.4% dice score.

Generalized Wasserstein Dice loss outperforms other loss functions.

Test-time augmentation enhances robustness and speed.

Abstract

Brain tumor segmentation from multiple Magnetic Resonance Imaging (MRI) modalities is a challenging task in medical image computation. The main challenges lie in the generalizability to a variety of scanners and imaging protocols. In this paper, we explore strategies to increase model robustness without increasing inference time. Towards this aim, we explore finding a robust ensemble from models trained using different losses, optimizers, and train-validation data split. Importantly, we explore the inclusion of a transformer in the bottleneck of the U-Net architecture. While we find transformer in the bottleneck performs slightly worse than the baseline U-Net in average, the generalized Wasserstein Dice loss consistently produces superior results. Further, we adopt an efficient test time augmentation strategy for faster and robust inference. Our final ensemble of seven 3D U-Nets with test-time augmentation produces an average dice score of 89.4% and an average Hausdorff 95% distance of 10.0 mm when evaluated on the BraTS 2021 testing dataset. Our code and trained models are publicly available at https://github.com/LucasFidon/TRABIT_BraTS2021.