Zhang Hantao, Guo Weidong, Wan Shouhong, Zou Bingbing, Wang Wanqin, Qiu Chenyang, Liu Kaige, Jin Peiquan, Yang Jiancheng
School of Computer Science and Technology, University of Science and Technology of China, Hefei, China.
Institute of Artificial Intelligence, Hefei Comprehensive National Science Center, Hefei, China.
Commun Med (Lond). 2025 Jul 1;5(1):256. doi: 10.1038/s43856-025-00953-0.
Rectal cancer segmentation in CT is crucial for timely diagnosis. Despite promising methods, challenges remain due to the rectum's complex anatomy and the lack of a comprehensive annotated dataset.
A total of 33,024 slice pairs from 398 rectal cancer patients in a new source center are enrolled into our dataset, named CARE Dataset, with pixel-level annotations for both normal and cancerous rectum tissue. We split it into 317 cases for training and 81 for testing. Additionally, we introduce a segmentation model, U-SAM, which, to the best of our knowledge, is a novel approach designed to handle the complex anatomy of the rectum by incorporating prompt information. Segmentation performance for both normal and cancerous rectum was evaluated using Intersection-over-Union (IoU), Dice Coefficient (Dice), and Normalized Surface Distance (NSD). With the assistance of 46 clinical practitioners, an observer study is conducted to benchmark the U-SAM with human performance and evaluate its clinical applicability. The original new source 398 CT scans and our code are openly available for research.
Our method achieves Dice of 71.23% for normal rectum and 76.38% for rectal tumor, with IoU of 55.32% and 61.78%, and NSD values of 83.63% and 58.59%, respectively, surpassing state-of-the-art methods. The observer study validates that U-SAM can produce diagnostic results comparable to those of highly experienced doctors in just 3 seconds of inference time (with about 5 minutes for prompt acquisition) in clinical settings.
The proposed U-SAM offers an efficient and reliable method for segmenting rectal cancer and normal tissue, significantly reducing time in clinical settings and effectively assisting radiologists. We believe this initial exploration in CT-based rectal cancer segmentation will be instrumental for future diagnosis.
CT 中的直肠癌分割对于及时诊断至关重要。尽管有一些很有前景的方法，但由于直肠解剖结构复杂且缺乏全面标注的数据集，挑战依然存在。
来自一个新来源中心的 398 例直肠癌患者的总共 33024 个切片对被纳入我们的数据集，命名为 CARE 数据集，其中正常和癌性直肠组织均有像素级标注。我们将其分为 317 例用于训练，81 例用于测试。此外，我们引入了一种分割模型 U-SAM，据我们所知，这是一种通过整合提示信息来处理直肠复杂解剖结构的新颖方法。使用交并比（IoU）、骰子系数（Dice）和归一化表面距离（NSD）评估正常和癌性直肠的分割性能。在 46 名临床医生的协助下，进行了一项观察者研究，将 U-SAM 与人类表现进行基准对比，并评估其临床适用性。原始的新来源 398 例 CT 扫描图像和我们的代码可供公开研究使用。
我们的方法在正常直肠上的骰子系数为 71.23%，在直肠肿瘤上为 76.38%，IoU 分别为 55.32%和 61.78%，NSD 值分别为 83.63%和 58.59%，超过了现有最先进的方法。观察者研究证实，在临床环境中，U-SAM 在仅 3 秒的推理时间（提示获取约需 5 分钟）内就能产生与经验丰富的医生相当的诊断结果。
所提出的 U-SAM 为直肠癌和正常组织的分割提供了一种高效可靠的方法，显著减少了临床环境中的时间，并有效协助放射科医生。我们相信，这种基于 CT 的直肠癌分割的初步探索将对未来的诊断起到重要作用。
