BeSt-LeS: Benchmarking Stroke Lesion Segmentation using Deep Supervision

TL;DR

This paper benchmarks various deep learning models, including transformer-based and residual U-Net, for stroke lesion segmentation using the ATLAS v2.0 dataset, providing insights into their performance and statistical validation.

Contribution

It introduces a comprehensive benchmarking of 2D and 3D U-Net style models for stroke lesion segmentation, including the highest Dice scores and statistical analysis for model validation.

Findings

Highest Dice score of 0.583 with 2D transformer model

Highest Dice score of 0.504 with 3D residual U-Net

Statistical correlation between predicted and actual stroke volume

Abstract

Brain stroke has become a significant burden on global health and thus we need remedies and prevention strategies to overcome this challenge. For this, the immediate identification of stroke and risk stratification is the primary task for clinicians. To aid expert clinicians, automated segmentation models are crucial. In this work, we consider the publicly available dataset ATLAS $v2.0$ to benchmark various end-to-end supervised U-Net style models. Specifically, we have benchmarked models on both 2D and 3D brain images and evaluated them using standard metrics. We have achieved the highest Dice score of 0.583 on the 2D transformer-based model and 0.504 on the 3D residual U-Net respectively. We have conducted the Wilcoxon test for 3D models to correlate the relationship between predicted and actual stroke volume. For reproducibility, the code and model weights are made publicly available: https://github.com/prantik-pdeb/BeSt-LeS.