Molecular Imaging, Radiotherapy and Oncology (MIRO), IREC, UCLouvain, Brussels, Belgium.
Department of Oncology, Laboratory of Experimental Radiotherapy, KU Leuven, Leuven, Belgium.
Med Phys. 2023 Oct;50(10):6201-6214. doi: 10.1002/mp.16431. Epub 2023 May 4.
In cancer care, determining the most beneficial treatment technique is a key decision affecting the patient's survival and quality of life. Patient selection for proton therapy (PT) over conventional radiotherapy (XT) currently entails comparing manually generated treatment plans, which requires time and expertise.
We developed an automatic and fast tool, AI-PROTIPP (Artificial Intelligence Predictive Radiation Oncology Treatment Indication to Photons/Protons), that assesses quantitatively the benefits of each therapeutic option. Our method uses deep learning (DL) models to directly predict the dose distributions for a given patient for both XT and PT. By using models that estimate the Normal Tissue Complication Probability (NTCP), namely the likelihood of side effects to occur for a specific patient, AI-PROTIPP can propose a treatment selection quickly and automatically.
A database of 60 patients presenting oropharyngeal cancer, obtained from the Cliniques Universitaires Saint Luc in Belgium, was used in this study. For every patient, a PT plan and an XT plan were generated. The dose distributions were used to train the two dose DL prediction models (one for each modality). The model is based on U-Net architecture, a type of convolutional neural network currently considered as the state of the art for dose prediction models. A NTCP protocol used in the Dutch model-based approach, including grades II and III xerostomia and grades II and III dysphagia, was later applied in order to perform automatic treatment selection for each patient. The networks were trained using a nested cross-validation approach with 11-folds. We set aside three patients in an outer set and each fold consists of 47 patients in training, five in validation and five for testing. This method allowed us to assess our method on 55 patients (five patients per test times the number of folds).
The treatment selection based on the DL-predicted doses reached an accuracy of 87.4% for the threshold parameters set by the Health Council of the Netherlands. The selected treatment is directly linked with these threshold parameters as they express the minimal gain brought by the PT treatment for a patient to be indicated to PT. To validate the performance of AI-PROTIPP in other conditions, we modulated these thresholds, and the accuracy was above 81% for all the considered cases. The difference in average cumulative NTCP per patient of predicted and clinical dose distributions is very similar (less than 1% difference).
AI-PROTIPP shows that using DL dose prediction in combination with NTCP models to select PT for patients is feasible and can help to save time by avoiding the generation of treatment plans only used for the comparison. Moreover, DL models are transferable, allowing, in the future, experience to be shared with centers that would not have PT planning expertise.
在癌症治疗中，确定最有益的治疗技术是影响患者生存和生活质量的关键决策。目前，质子治疗（PT）与传统放疗（XT）的患者选择需要比较手动生成的治疗计划，这需要时间和专业知识。
我们开发了一种自动快速的工具，即 AI-PROTIPP（人工智能预测放射肿瘤治疗光子/质子指示），它可以定量评估每种治疗选择的益处。我们的方法使用深度学习（DL）模型直接预测给定患者的 XT 和 PT 的剂量分布。通过使用估计正常组织并发症概率（NTCP）的模型，即特定患者发生副作用的可能性，AI-PROTIPP 可以快速自动地提出治疗选择。
本研究使用了来自比利时圣卢克大学附属医院的 60 名口咽癌患者的数据库。为每位患者生成了 PT 计划和 XT 计划。使用剂量分布来训练两个剂量 DL 预测模型（每种模式一个）。该模型基于 U-Net 架构，这是一种目前被认为是剂量预测模型的最新技术的卷积神经网络。后来应用了荷兰基于模型的方法中使用的 NTCP 方案，包括 II 级和 III 级口干和 II 级和 III 级吞咽困难，以便为每位患者自动选择治疗。网络使用嵌套交叉验证方法进行训练，共 11 折。我们在一个外部集合中留出三个患者，每个折叠由 47 个训练患者、5 个验证患者和 5 个测试患者组成。这种方法使我们能够在 55 名患者（每个测试乘以折叠次数的五名患者）上评估我们的方法。
基于 DL 预测剂量的治疗选择达到了荷兰卫生委员会设定的阈值参数的 87.4%的准确性。所选治疗与这些阈值参数直接相关，因为它们表示 PT 治疗为患者带来的最小收益，以使患者适合接受 PT 治疗。为了验证 AI-PROTIPP 在其他情况下的性能，我们调整了这些阈值，所有考虑的情况下的准确性都高于 81%。预测和临床剂量分布的患者平均累积 NTCP 差异非常相似（差异小于 1%）。
AI-PROTIPP 表明，在结合 NTCP 模型使用 DL 剂量预测为患者选择 PT 是可行的，并可以通过避免仅用于比较的治疗计划的生成来节省时间。此外，DL 模型是可转移的，允许与没有 PT 计划专业知识的中心共享经验。
