Institute for Onco-Physics, Albert Einstein College of Medicine, Bronx, NY, USA.
Department of Radiation Oncology, Montefiore Medical Center, Bronx, NY, USA.
J Appl Clin Med Phys. 2022 Jun;23(6):e13609. doi: 10.1002/acm2.13609. Epub 2022 Apr 23.
To quantify the clinical performance of a machine learning (ML) algorithm for organ-at-risk (OAR) dose prediction for lung stereotactic body radiation therapy (SBRT) and estimate the treatment planning benefit from having upfront access to these dose predictions.
ML models were trained using multi-center data consisting of 209 patients previously treated with lung SBRT. Two prescription levels were investigated, 50 Gy in five fractions and 54 Gy in three fractions. Models were generated using a gradient-boosted regression tree algorithm using grid searching with fivefold cross-validation. Twenty patients not included in the training set were used to test OAR dose prediction performance, ten for each prescription. We also performed blinded re-planning based on OAR dose predictions but without access to clinically delivered plans. Differences between predicted and delivered doses were assessed by root-mean square deviation (RMSD), and statistical differences between predicted, delivered, and re-planned doses were evaluated with one-way analysis of variance (ANOVA) tests.
ANOVA tests showed no significant differences between predicted, delivered, and replanned OAR doses (all p ≥ 0.36). The RMSD was 2.9, 3.9, 4.3, and 1.7Gy for max dose to the spinal cord, great vessels, heart, and trachea, respectively, for 50 Gy in five fractions. Average improvements of 1.0, 1.4, and 2.0 Gy were seen for spinal cord, esophagus, and trachea max doses in blinded replans compared to clinically delivered plans with 54 Gy in three fractions, and 1.8, 0.7, and 1.5 Gy, respectively, for the esophagus, heart and bronchus max doses with 50 Gy in five fractions. Target coverage was similar with an average PTV V100% of 94.7% for delivered plans compared to 97.3% for blinded re-plans for 50 Gy in five fractions, and respectively 98.4% versus 99.2% for 54 Gy in three fractions.
This study validated ML-based OAR dose prediction for lung SBRT, showing potential for improved OAR dose sparing and more consistent plan quality using dose predictions for patient-specific planning guidance.
定量评估机器学习（ML）算法在预测肺癌立体定向体部放疗（SBRT）中危及器官（OAR）剂量方面的临床性能，并估算提前获取这些剂量预测信息的治疗计划获益。
使用包含 209 例既往接受肺部 SBRT 治疗患者的多中心数据训练 ML 模型。研究了两种处方剂量水平，分别为 50 Gy 分 5 次和 54 Gy 分 3 次。采用梯度提升回归树算法，通过五重交叉验证进行网格搜索生成模型。使用 20 例未纳入训练集的患者进行 OAR 剂量预测性能测试，每个处方剂量各 10 例。我们还基于 OAR 剂量预测进行了盲法重新计划，但不参考临床实施的计划。通过均方根偏差（RMSD）评估预测剂量与实际剂量之间的差异，并采用单因素方差（ANOVA）检验评估预测剂量、实际剂量和重新计划剂量之间的统计学差异。
ANOVA 检验结果显示，预测剂量、实际剂量和重新计划剂量之间无显著差异（均 p≥0.36）。50 Gy 分 5 次处方时，脊髓、大血管、心脏和气管的最大剂量预测值与实际值之间的 RMSD 分别为 2.9、3.9、4.3 和 1.7 Gy。54 Gy 分 3 次处方时，盲法重新计划可使脊髓、食管和气管最大剂量分别平均改善 1.0、1.4 和 2.0 Gy，而 50 Gy 分 5 次处方时，食管、心脏和支气管最大剂量分别平均改善 1.8、0.7 和 1.5 Gy。对于 50 Gy 分 5 次处方，临床实施计划和盲法重新计划的靶区体积（PTV）V100%分别为 94.7%和 97.3%；而对于 54 Gy 分 3 次处方，靶区体积 V100%分别为 98.4%和 99.2%。
本研究验证了基于 ML 的肺部 SBRT 中 OAR 剂量预测，显示出通过剂量预测进行个体化计划指导，在改善 OAR 剂量保护和提高计划质量方面具有潜在应用价值。
