Zang Lele, Liu Jing, Zhang Huiqi, Zhu Shitao, Zhu Mingxuan, Wang Yuqin, Kang Yaxin, Chen Jihong, Xu Qin
Department of Gynecology, Clinical Oncology School of Fujian Medical University, Fujian Cancer Hospital, Fuzhou, 350011, Fujian, China.
Department of Radiation Oncology, Fujian Cancer Hospital, Clinical Oncology School of Fujian Medical University, Fuzhou, 350011, Fujian, China.
Sci Rep. 2025 Mar 24;15(1):10152. doi: 10.1038/s41598-025-94431-1.
This study developed and evaluated an automatic segmentation model based on the Mamba framework (AM-UNet) for rapid and precise delineation of high-risk clinical target volume (HRCTV) and organs at risk (OARs) in cervical cancer brachytherapy. Using 694 CT scans from 179 cervical cancer patients, the performance of five models (AM-UNet, UNet, DeepLab V3, UNETR and nnU-Net) was compared. The models were assessed using the Dice similarity coefficient (DSC), 95% Hausdorff distance (HD95), and dose-volume index (DVI). AM-UNet achieved mean DSCs of 0.862, 0.937, 0.823, and 0.725 for HRCTV, bladder, rectum, and sigmoid, respectively. Subjective evaluations showed 93.07% of AM-UNet predicted HRCTV were rated as clinically acceptable or needing minor adjustments, with no unacceptable cases. Dosimetric differences between AM-UNet-generated and manually delineated contours were within 1%, highlighting its potential for improving clinical workflows in brachytherapy.
本研究开发并评估了一种基于曼巴框架的自动分割模型（AM-UNet），用于在宫颈癌近距离放疗中快速、精确地勾勒高危临床靶区（HRCTV）和危及器官（OARs）。使用来自179例宫颈癌患者的694份CT扫描图像，比较了五个模型（AM-UNet、UNet、DeepLab V3、UNETR和nnU-Net）的性能。使用骰子相似系数（DSC）、95%豪斯多夫距离（HD95）和剂量体积指数（DVI）对模型进行评估。AM-UNet对HRCTV、膀胱、直肠和乙状结肠的平均DSC分别达到0.862、0.937、0.823和0.725。主观评估显示，AM-UNet预测的HRCTV中有93.07%被评为临床可接受或只需进行小调整，无不可接受的情况。AM-UNet生成的轮廓与手动勾勒的轮廓之间的剂量差异在1%以内，突出了其在改善近距离放疗临床工作流程方面的潜力。
