MTNeuro

MTNeuro是一个多任务神经影像基准，基于体积的、微米分辨率X射线显微断层成像图像，覆盖了老鼠大脑的大量丘脑皮层部分，包括多个皮层和皮下区域。该数据集提供了多种方式来读取关于大脑结构和架构的信息，并生成了多种预测挑战，评估了多种监督和自监督模型用于脑区预测和像素级微结构语义分割。

MTNeuro is a multi-task neuroimaging benchmark based on volumetric, micrometer-resolution X-ray microtomography images, covering a large thalamocortical portion of the mouse brain, including multiple cortical and subcortical regions. This dataset provides multiple approaches to accessing information about brain structure and architecture, and has created diverse prediction challenges for evaluating various supervised and self-supervised models for brain region prediction and pixel-level microstructural semantic segmentation.