BeSt-LeS: Benchmarking Stroke Lesion Segmentation using Deep Supervision
This work provides a benchmark for clinicians to aid in stroke identification and risk stratification, but it is incremental as it applies existing methods to a public dataset.
The authors benchmarked U-Net style models for automated stroke lesion segmentation on the ATLAS v2.0 dataset, achieving a highest Dice score of 0.583 on 2D transformer-based models and 0.504 on 3D residual U-Nets.
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.