PreCT-160K

该数据集是迄今为止从多种来源收集的最大医疗预训练数据集，确保了训练过程中数据格式的统一性。此外，该数据集还经过了彻底的预处理，以解决体积间在大小、间隔和强度上的差异。其任务旨在为医疗图像分析进行预训练。

This is the largest medical pre-training dataset collected from diverse sources to date, ensuring the uniformity of data formats during the training process. Furthermore, this dataset has undergone rigorous preprocessing to resolve discrepancies in size, spacing, and intensity among different medical volumes. The pre-training task based on this dataset is specifically designed for medical image analysis.