MSLesSeg: baseline and benchmarking of a new Multiple Sclerosis Lesion Segmentation dataset

MSLesSeg is a publicly accessible MRI dataset aimed at advancing research in Multiple Sclerosis (MS) lesion segmentation. It includes 115 MRI scans from 75 patients, featuring T1, T2, and FLAIR sequences. The dataset is enriched with supplementary clinical data gathered from multiple sources. Expert-validated annotations provide high-quality lesion segmentation labels, offering a reliable, human-labeled benchmark for evaluating segmentation algorithms.MSLesSeg 是一款公开可获取的磁共振成像（Magnetic Resonance Imaging, MRI）数据集，旨在推动多发性硬化（Multiple Sclerosis, MS）病灶分割领域的研究工作。该数据集包含来自75名患者的115次MRI扫描，涵盖T1、T2与液体衰减反转恢复（Fluid-Attenuated Inversion Recovery, FLAIR）序列。数据集还补充了多源采集的附属临床数据。经专家验证的标注提供了高质量的病灶分割标签，可为分割算法的性能评估提供可靠的人工标注基准。