深度学习放疗剂量预测:单中心622例多瘤种临床洞察
本研究纳入单中心622例多瘤种计划数据,用U-Net、Flex-Net和Highres-Net预测剂量,以归一化剂量差异(NDD)和剂量体积直方图(DVH)指标定量比较,并由放射肿瘤科医生对测试集做定性评估。结果显示鼻咽癌、乳腺癌和术后宫颈癌靶区NDD低于3%,测试集(n=123)中53.7%预测被判定临床可接受、32.5%需小幅调整,采用「最佳选择」策略将临床接受度提升至62.6%。
要点
- 鼻咽癌、乳腺癌和术后宫颈癌靶区预测与临床剂量相关性好,NDD低于3%。
- U-Net在宫颈根治性放疗、Flex-Net在乳腺/直肠/术后宫颈癌、Highres-Net在鼻咽癌/食管癌中准确度最高。
- 测试集53.7%预测临床可接受、32.5%需小幅调整,「最佳选择」策略使接受度升至62.6%。
核心结论
- 自动剂量预测可为快速计划生成提供可靠起点,「最佳选择」策略通过整合各模型优势提升预测准确度并有望改善多瘤种临床效率。
研究设计与样本
- 单中心纳入622例多瘤种计划数据,覆盖鼻咽癌(n=29)、食管癌(n=82)、左乳癌(n=107)、右乳癌(n=95)、根治性放疗宫颈癌(n=84)、术后宫颈癌(n=122)和直肠癌(n=103)。
- 使用U-Net、Flex-Net和Highres-Net生成剂量预测,数据按训练60%、验证20%、测试20%划分。
- 定量比较采用归一化剂量差异(NDD)和剂量体积直方图(DVH)指标,并在测试集上由放射肿瘤科医生做定性评估。
主要发现
- 预测剂量与临床剂量相关性良好,鼻咽癌、乳腺癌和术后宫颈癌靶区NDD低于3%。
- 定性评估显示U-Net在宫颈根治性放疗、Flex-Net在乳腺/直肠/术后宫颈癌、Highres-Net在鼻咽癌/食管癌中准确度最高。
- 测试集(n=123)中53.7%预测被判定临床可接受、32.5%需小幅调整。
- 「最佳选择」策略整合三种模型优势,将临床接受度提升至62.6%。
临床怎么用
- 可将自动剂量预测作为快速计划生成的起点,并按瘤种选用准确度最高的模型。
- 对预测结果采用「最佳选择」策略整合多模型优势,以提高临床接受度。
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
PURPOSE: Accurate pre-treatment dose prediction is essential for efficient radiotherapy planning. Although deep learning models have advanced automated dose distribution, comprehensive multi-tumor analyses remain scarce. This study assesses deep learning models for dose prediction across diverse tumor types, combining objective and subjective evaluation methods. METHODS AND MATERIALS: We included 622 patients with planning data across various tumor sites: nasopharyngeal carcinoma (n = 29), esophageal carcinoma (n = 82), left-sided breast carcinoma (n = 107), right-sided breast carcinoma (n = 95), cervical carcinoma treated with radical radiotherapy (n = 84), postoperative cervical carcinoma (n = 122), and rectal carcinoma (n = 103). Dose predictions were generated using U-Net, Flex-Net, and Highres-Net models, with data split into training (60%), validation (20%), and testing (20%) sets. Quantitative comparisons used normalized dose difference (NDD) and dose-volume histogram (DVH) metrics, and qualitative assessments by radiation oncologists were performed on the testing set. RESULTS: Predicted and clinical doses correlated well, with NDD values under 3% for tumor targets in nasopharyngeal, breast, and postoperative cervical cancer. Qualitative assessments revealed that U-Net, Flex-Net, and Highres-Net achieved the highest accuracy in cervical radical, breast/rectal/postoperative cervical, and nasopharyngeal/esophageal cancers, respectively. Among the test cases (n = 123), 53.7% were deemed clinically acceptable and 32.5% required minor adjustments. The "Best Selection" approach, combining strengths of all three models, raised clinical acceptance to 62.6%. CONCLUSION: This study demonstrates that automated dose prediction can provide a robust starting point for rapid plan generation. Leveraging model-specific strengths through the "Best Selection" approach enhances prediction accuracy and shows potential to improve clinical efficiency across multiple tumor types.
引用
- Hou Z, Qin L, Gu J, et al. Deep learning-powered radiotherapy dose prediction: clinical insights from 622 patients across multiple sites tumor at a single institution. Radiat Oncol. 2025. PMID: 40390053. doi:10.1186/s13014-025-02634-7
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。