ISLES’24 (Ischemic Stroke Lesion Segmentation Challenge 2024)

ISLES’24 – Ischemic Stroke Lesion Segmentation Challenge 2024 A real-world, longitudinal, multimodal acute-stroke dataset: acute admission CT (NCCT, CTA, 4D CT-perfusion + derived perfusion maps), follow-up MRI (DWI/ADC), and clinical tabular data, with the final infarct lesion as the segmentation target. This repository mirrors the public TRAINING set only (149 cases). The 98-case ISLES’24 test set is withheld by the organizers for challenge scoring on Grand Challenge and is… See the full description on the dataset page:

Source: Hugging Face Hub (MedOtter/isles2024). Metadata imported from the dataset’s Hub tags.