How We Won the ISLES'24 Challenge by Preprocessing
This work addresses stroke diagnosis and treatment by improving segmentation accuracy, though it is incremental as it focuses on preprocessing enhancements.
The paper tackled the problem of stroke lesion segmentation from CT scans by developing a preprocessing pipeline, achieving a mean test Dice score of 28.5 with a standard deviation of 21.27 in the ISLES'24 challenge.
Stroke is among the top three causes of death worldwide, and accurate identification of stroke lesion boundaries is critical for diagnosis and treatment. Supervised deep learning methods have emerged as the leading solution for stroke lesion segmentation but require large, diverse, and annotated datasets. The ISLES'24 challenge addresses this need by providing longitudinal stroke imaging data, including CT scans taken on arrival to the hospital and follow-up MRI taken 2-9 days from initial arrival, with annotations derived from follow-up MRI. Importantly, models submitted to the ISLES'24 challenge are evaluated using only CT inputs, requiring prediction of lesion progression that may not be visible in CT scans for segmentation. Our winning solution shows that a carefully designed preprocessing pipeline including deep-learning-based skull stripping and custom intensity windowing is beneficial for accurate segmentation. Combined with a standard large residual nnU-Net architecture for segmentation, this approach achieves a mean test Dice of 28.5 with a standard deviation of 21.27.