Medical Physics Department, Centre François Baclesse, 14000 Caen, France.
Medical Physics Department, Centre François Baclesse, 14000 Caen, France; Université de Caen Normandie, CNRS, Normandie Université, ISTCT UMR6030, GIP CYCERON, F-14000 Caen, France.
Radiother Oncol. 2024 Nov;200:110483. doi: 10.1016/j.radonc.2024.110483. Epub 2024 Aug 17.
New radiotherapy machines such as Halcyon are capable of delivering dose-rate of 600 monitor-units per minute, allowing large numbers of patients treated per day. However, patient-specific quality assurance (QA) is still required, which dramatically decrease machine availability. Innovative artificial intelligence (AI) algorithms could predict QA result based on complexity metrics. However, no AI solution exists for Halcyon machines and the complexity metrics to be used have not been definitively determined. The aim of this study was to develop an AI solution capable of firstly determining the complexity indices to be obtained and secondly predicting patient-specific QA in a routine clinical setting.
Three hundred and eighteen beams from 56 patients with breast cancer were used. The seven complexity indices named Modulation-Complexity-Score (MCS), Small-Aperture-Score (SAS10), Beam-Area (BA), Beam-Irregularity (BI), Beam-Modulation (BM), Gantry and Collimator angles were used as input to the AI model. Machine learning (ML) and deep learning (DL) models using tensorflow were set up to predict DreamDose QA conformance.
MCS, BI, gantry and collimator angle are not correlated with QA compliance. Therefore, ML and DL models were trained using SAS10, BA and BM complexity indices. ROC analyses enabled to find best predicted probability threshold to increase specificity and sensitivity. ML models did not show satisfactory performance with an area under-the-curve (AUC) of 0.75 and specificity and sensitivity of 0.88 and 0.86. However, optimised DL model showed better performance with an AUC of 0.95 and specificity and sensitivity of 0.98 and 0.97.
The DL model demonstrated a high degree of accuracy in its predictions of the quality assurance (QA) results. Our online predictive QA-platform offers significant time savings in terms of accelerator occupancy and working time.
新的放疗设备，如 Halcyon，能够以每分钟 600 个监测单位的剂量率进行治疗，从而每天能够治疗大量患者。然而，仍然需要进行患者特异性的质量保证（QA），这会极大地降低机器的可用性。创新的人工智能（AI）算法可以根据复杂性指标预测 QA 结果。然而，目前还没有针对 Halcyon 机器的 AI 解决方案，也没有确定要使用的复杂性指标。本研究的目的是开发一种 AI 解决方案，能够首先确定要获得的复杂性指数，其次在常规临床环境中预测患者特异性 QA。
使用 56 例乳腺癌患者的 318 束射线。将七个复杂性指数，即调制复杂性评分（MCS）、小孔径评分（SAS10）、射束面积（BA）、射束不规则性（BI）、射束调制（BM）、转架和准直器角度，作为输入输入到 AI 模型中。使用 tensorflow 建立了机器学习（ML）和深度学习（DL）模型，以预测 DreamDose QA 的一致性。
MCS、BI、转架和准直器角度与 QA 一致性不相关。因此，使用 SAS10、BA 和 BM 复杂性指数训练 ML 和 DL 模型。ROC 分析确定了最佳预测概率阈值，以提高特异性和敏感性。ML 模型的性能并不令人满意，曲线下面积（AUC）为 0.75，特异性和敏感性分别为 0.88 和 0.86。然而，优化后的 DL 模型表现出更好的性能，AUC 为 0.95，特异性和敏感性分别为 0.98 和 0.97。
DL 模型在预测 QA 结果方面表现出高度的准确性。我们的在线预测 QA 平台在加速器占用和工作时间方面都节省了大量时间。
