# CT开放数据集

**URL:** <https://forum.beginner.center/t/topic/1898>\
**Category:** 🥼实践与临床\
**Created:** [2025 年4 月 16 日 11:05 UTC](https://forum.beginner.center/t/topic/1898 "2025-04-16T11:05:08Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![doggie](https://forum.beginner.center/user_avatar/forum.beginner.center/doggie/32/550_2.png) [@doggie](https://forum.beginner.center/u/doggie)\
**Post date:** [2025 年4 月 16 日 11:05 UTC](https://forum.beginner.center/t/topic/1898/1 "2025-04-16T11:05:08Z")

</div>

> **[Box](https://nihcc.app.box.com/v/DeepLesion)**

[DeepLesion](https://nihcc.app.box.com/v/DeepLesion) dataset contains 32,735 diverse lesions in 32,120 CT slices from 10,594 studies of 4,427 unique patients. Each lesion has a bounding box annotation on the key slice, which is derived from the longest diameter and longest perpendicular diameter. We annotated 5000 lesions with [MedSAM2](https://github.com/bowang-lab/MedSAM2) in a human-in-the-loop pipeline.  
[DeepLesion](https://nihcc.app.box.com/v/DeepLesion) 数据集包含来自 4,427 名独特患者的 10,594 项研究的 32,120 个 CT 切片中的 32,735 个不同病灶。每个病变在关键切片上都有一个边界框注释，该注释源自最长直径和最长垂直直径。我们在人机回环管道中用 [MedSAM2](https://github.com/bowang-lab/MedSAM2) 注释了 5000 个病灶。

# 资讯来源

- [wanglab/CT\_DeepLesion-MedSAM2 · Datasets at Hugging Face](https://huggingface.co/datasets/wanglab/CT_DeepLesion-MedSAM2)