轻量多期分割框架用于子宫内膜癌临床靶区勾画
研究纳入175例接受术后盆腔放疗的子宫内膜癌患者,构建融合增强与非增强CT特征的轻量深度学习框架NCLNet自动勾画临床靶区。NCLNet的DSC为0.871±0.027,ASSD为0.878±0.265 mm,参数量仅5.7 M;医生修改自动勾画靶区的平均时间为2.9 min,修改体积占比3.61%。
要点
- NCLNet在175例子宫内膜癌术后盆腔放疗患者中实现CTV自动勾画,DSC为0.871±0.027。
- NCLNet参数量5.7 M,低于nnUNet的31.0 M,DSC 0.871高于nnUNet的0.860。
- 医生修改自动勾画CTV平均仅需2.9 min,修改体积占比3.61%。
核心结论
- NCLNet可为子宫内膜癌术后盆腔放疗患者生成高质量的临床靶区自动勾画。
研究设计与样本
- 研究构建多期分割数据集,纳入175例接受术后盆腔放疗的子宫内膜癌患者,包含增强CT和非增强CT图像。
- 提出基于深度学习的NCLNet框架,在轻量网络结构中融合非增强与增强CT特征,并采用边界感知多期学习策略优化。
- 采用DSC、ASSD及新提出的轮廓DSC(CDSC)评价预测外轮廓精度,并由三名医生修改预测CTV以评估临床实用性。
主要发现
- NCLNet的DSC为0.871±0.027,ASSD为0.878±0.265 mm,参数量5.7 M,FLOPS为639.1G。
- nnUNet的DSC为0.860±0.027,ASSD为0.920±0.286 mm,参数量31.0 M,FLOPs为643.8G。
- 自动勾画CTV的平均修改时间为2.9 min,修改体积百分比为3.61%。
- 8 mm厚度的CDSC为0.853±0.030,与专家临床修改时间的相关性高于DSC和ASSD。
临床怎么用
- 该框架可用于子宫内膜癌术后盆腔放疗CTV的自动勾画,减少医生修改时间与修改体积。
- 临床使用时医生仍需对自动勾画结果进行修改确认,平均修改时间约2.9 min。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
BACKGROUND: The accurate delineation of the clinical target volume (CTV) is a critical step in precision radiotherapy for endometrial cancer (EC). Multi-phase CT provides more information for delineating the CTV. Our study aims to establish an innovative method for the specific delineation of CTV using multi-phase CT. METHODS: Our multi-phase delineation datasets comprise 175 patients who received postoperative pelvic radiotherapy. These datasets include images of contrast-enhanced computed tomography (CECT) and non-contrast-enhanced computed tomography (NECT). Additionally, we introduce a novel framework for automatically segmenting the CTV using a deep learning model. The key component of our framework is the NCLNet, which fuses features from NECT and CECT within a Lightweight Network structure. This structure is optimized using a boundary-aware multi-phase learning strategy that we propose. In addition to the dice similarity coefficient (DSC) and the average symmetric surface distance (ASSD), we propose a novel contour dice similarity coefficient (CDSC) metric to evaluate the accuracy of the predictive outer contour. Three physicians modified the predictive CTV to assess the clinical utility of the proposed method. RESULTS: The NCLNet achieves a DSC of 0.871 ± 0.027 and an ASSD of 0.878 ± 0.265 mm, with lower computation complexity (5.7 M parameters and FLOPS was 639.1G). In comparison, the widely used nnUNet attains a DSC of 0.860 ± 0.027 and an ASSD of 0.920 ± 0.286 mm, while requiring significantly more parameters (31.0 M) and similar FLOPs (643.8G). The overall evaluation against several benchmarks demonstrates better or comparable performance relative to methods with higher complexity. The average modification time and the modification volume percentage of the automatically delineated CTV were only 2.9 min and 3.61%, respectively. The CDSC of 8 mm thickness was 0.853 ± 0.030, demonstrating higher correlation with the clinical modification time of experts than both DSC and ASSD. CONCLUSIONS: The NCLNet generates high-quality automatic delineation of the CTV for postoperative pelvic radiotherapy in patients with EC.
引用
- Qu A, Zhu L, Xiong W, et al. Discerning clinical target volume of endometrial cancer via a lightweight multi-phase delineation framework. Radiat Oncol. 2026. PMID: 41680891. doi:10.1186/s13014-026-02800-5
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。