BraTS2023

该数据集名为BraTS2023，包含了580位患者的磁共振成像（MRI）数据，重点关注T1、T2和FLAIR三种类型的图像。数据集被划分为训练集（500位患者）、验证集（40位患者）和测试集（40位患者）。所有图像均被调整至240x240的尺寸，并归一化至[0, 1]的范围。训练集中包含了40,000张图像。该数据集的任务是多模态磁共振图像合成。

This dataset, named BraTS2023, contains magnetic resonance imaging (MRI) data from 580 patients, focusing on T1-weighted, T2-weighted, and FLAIR images. The dataset is split into three subsets: training set (500 patients), validation set (40 patients), and test set (40 patients). All images have been resized to 240×240 pixels and normalized to the range [0, 1]. The training set includes 40,000 images. The core task of this dataset is multimodal magnetic resonance image synthesis.