Department of Radiation Oncology, Cancer Center, West China Hospital, Sichuan University, Chengdu, China.
Abdominal Oncology Ward, Cancer Center, West China Hospital, Sichuan University, Chengdu, China.
J Appl Clin Med Phys. 2024 Oct;25(10):e14482. doi: 10.1002/acm2.14482. Epub 2024 Aug 9.
Radiotherapy has been crucial in prostate cancer treatment. However, manual segmentation is labor intensive and highly variable among radiation oncologists. In this study, a deep learning based automated contouring model is constructed for clinical target volumes (CTVs) of intact and postoperative prostate cancer.
Computed tomography (CT) data sets of 197 prostate cancer patients were collected. Two auto-delineation models were built for radical radiotherapy and postoperative radiotherapy of prostate cancer respectively, and each model included CTVn for pelvic lymph nodes and CTVp for prostate tumors or prostate tumor beds.
In the radical radiotherapy model, the volumetric dice (VD) coefficient of CTVn calculated by AI, was higher than that of the one delineated by the junior physicians (0.85 vs. 0.82, p = 0.018); In the postoperative radiotherapy model, the quantitative parameter of CTVn and CTVp, counted by AI, was better than that of the junior physicians. The median delineation time for AI was 0.23 min in the postoperative model and 0.26 min in the radical model, which were significantly shorter than those of the physicians (50.40 and 45.43 min, respectively, p < 0.001). The correction time of the senior physician for AI was much shorter compared with that for the junior physicians in both models (p < 0.001).
Using deep learning and attention mechanism, a highly consistent and time-saving contouring model was built for CTVs of pelvic lymph nodes and prostate tumors or prostate tumor beds for prostate cancer, which also might be a good approach to train junior radiation oncologists.
放射治疗在前列腺癌治疗中至关重要。然而，手动分割对于放射肿瘤学家来说既费力又高度可变。在这项研究中，构建了一种基于深度学习的自动勾画模型，用于勾画完整前列腺癌和前列腺癌术后的临床靶区（CTV）。
收集了 197 例前列腺癌患者的计算机断层扫描（CT）数据集。分别为根治性放疗和前列腺癌术后放疗构建了 2 个自动勾画模型，每个模型均包括盆腔淋巴结CTVn 和前列腺肿瘤或前列腺肿瘤床的 CTVp。
在根治性放疗模型中，AI 计算的 CTVn 体积 Dice 系数（VD）高于初级医师勾画的 CTVn（0.85 与 0.82，p=0.018）；在术后放疗模型中，AI 勾画的 CTVn 和 CTVp 的定量参数优于初级医师。AI 在术后模型中的勾画中位时间为 0.23 分钟，在根治性模型中的勾画中位时间为 0.26 分钟，均显著短于医师的勾画时间（分别为 50.40 分钟和 45.43 分钟，p<0.001）。在两个模型中，高级医师对 AI 的修正时间均明显短于初级医师（p<0.001）。
使用深度学习和注意力机制，为前列腺癌盆腔淋巴结和前列腺肿瘤或前列腺肿瘤床的 CTV 构建了一个高度一致且节省时间的勾画模型，这也可能是培训初级放射肿瘤学家的一种良好方法。
