Auto-PET是一个自动化PET/CT器官分割数据集。该数据集首次在MICCAI 2022会议上发布,并在MICCAI 2023上进一步扩展。初始版本包含900个PET/CT扫描体数据,对应1014组PET-CT数据。截至2022年底新增150个体数据,2023年3月又增加200个。此次扩展包含一组新的FDG-PET-CT图像,形成了一个包含脑部掩模的综合性CT和PET数据集。提供的分割标签与LITS标准一致,数据来自慕尼黑路德维希-马克西米利安大学采集的FDG-PET/CT图像,为器官分割研究提供了宝贵资源。
全身维度 | 3D |
模态 | ct |
任务类型 | segmentation |
解剖结构 | 肿瘤 |
解剖区域 | 全身 |
类别数 | 1 |
数据量 | 1014训练集,200测试集 |
文件格式 | .nii.gz |
|--- Patient 1
|--- Study 1
|--- SUV.nii.gz (PET image in SUV)
|--- CTres.nii.gz (CT image resampled to PET)
|--- CT.nii.gz (Original CT image)
|--- SEG.nii.gz (Manual annotations of tumor lesions)
|--- PET.nii.gz (Original PET image as actictivity counts)
|--- Study 2 (Potential 2nd visit of same patient)
|--- ...
|--- Patient 2
|--- ...
统计类型 | 间距 (mm) | 尺寸 |
---|---|---|
最小值 | (2.036, 2.036, 3) |
(400, 400, 200) |
中位值 | (2.036, 2.036, 3) |
(400, 400, 326) |
最大值 | (2.036, 2.036, 3) |
(512, 512, 743) |
@article{gatidis2022whole,
title={A whole-body FDG-PET/CT Dataset with manually annotated Tumor Lesions},
author={Gatidis, Sergios and Hepp, Tobias and Fr{\"u}h, Marcel and La Foug{\`e}re, Christian and Nikolaou, Konstantin and Pfannenberg, Christina and Sch{\"o}lkopf, Bernhard and K{\"u}stner, Thomas and Cyran, Clemens and Rubin, Daniel},
journal={Scientific Data},
volume={9},
number={1},
pages={601},
year={2022},
publisher={Nature Publishing Group UK London}
}
@article{gatidis2023autopet,
title={The autoPET challenge: Towards fully automated lesion segmentation in oncologic PET/CT imaging},
author={Gatidis, Sergios and Fr{\"u}h, Marcel and Fabritius, Matthias and Gu, Sijing and Nikolaou, Konstantin and La Foug{\`e}re, Christian and Ye, Jin and He, Junjun and Peng, Yige and Bi, Lei and others},
year={2023}
}