Evaluation of dose-volume histogram prediction for organ-at risk and planning target volume based on machine learning

Abstract

The purpose of this work is to evaluate the performance of applying patient dosimetric information induced by individual uniform-intensity radiation fields in organ-at risk (OAR) dose-volume histogram (DVH) prediction, and extend to DVH prediction of planning target volume (PTV). Ninety nasopharyngeal cancer intensity-modulated radiation therapy (IMRT) plans and 60 rectal cancer volumetric modulated arc therapy (VMAT) plans were employed in this study. Of these, 20 nasopharyngeal cancer cases and 15 rectal cancer cases were randomly selected as the testing data. The DVH prediction was performed using two methods. One method applied the individual dose-volume histograms (IDVHs) induced by a series of fields with uniform-intensity irradiation and the other method applied the distance-to-target histogram and the conformal-plan-dose-volume histogram (DTH + CPDVH). The determination coefficient R2 and mean absolute error (MAE) were used to evaluate DVH prediction accuracy. The PTV DVH prediction was performed using the IDVHs. The PTV dose coverage was evaluated using D98, D95, D1 and uniformity index (UI). The OAR dose was compared using the maximum dose, V30 and V40. The significance of the results was examined with the Wilcoxon signed rank test. For PTV DVH prediction using IDVHs, the clinical plan and IDVHs prediction method achieved mean UI values of 1.07 and 1.06 for nasopharyngeal cancer, and 1.04 and 1.05 for rectal cancer, respectively. No significant difference was found between the clinical plan results and predicted results using the IDVHs method in achieving PTV dose coverage (D98,D95,D1 and UI) for both nasopharyngeal cancer and rectal cancer (p-values ≥ 0.052). For OAR DVH prediction, no significant difference was found between the IDVHs and DTH + CPDVH methods for the R2, MAE, the maximum dose, V30 and V40 (p-values ≥ 0.087 for all OARs). This work evaluates the performance of dosimetric information of several individual fields with uniform-intensity radiation for DVH prediction, and extends its application to PTV DVH prediction. The results indicated that the IDVHs method is comparable to the DTH + CPDVH method in accurately predicting the OAR DVH. The IDVHs method quantified the input features of the PTV and showed reliable PTV DVH prediction, which is helpful for plan quality evaluation and plan generation.

Organ‐at‐risk dose prediction using a machine learning algorithm: Clinical validation and treatment planning benefit for lung SBRT

“…One research group used two different methods, one based on the uniform field doses of nine equidistant IMRT beams and the other utilizing geometric information about relative target and OAR positions. 10 With similar results for the two methods the study found mean prediction errors across OARs of between 1% to 8% for nasopharyngeal cancer cases, and 2% to 6% for rectal cancer cases, compared to 0.3% to 3.7% for 50 Gy in five fractions and 0.3% to 7.6% for 54 Gy in three fractions in our study. Another approach also focusing on patients with nasopharyngeal cancer adopted multiple linear fitting of DVH metrics from various sub‐regions of OARs based on how far away each sub‐region is from the target.…”

“…A recent study evaluated two different methods of DVH prediction for the treatment of nasopharyngeal cancer and rectal cancer. 10 The authors found that prediction of specific dose‐volume histogram (DVH) points was achievable within 5% accuracy for most of the studied OARs.…”

“…Dose prediction strategies strive to estimate patient‐specific OAR doses without the need to generate a treatment plan, facilitating quick and efficient discussions about achievable plan quality and trade‐offs to consider. A recent study evaluated two different methods of DVH prediction for the treatment of nasopharyngeal cancer and rectal cancer 10 . The authors found that prediction of specific dose‐volume histogram (DVH) points was achievable within 5% accuracy for most of the studied OARs.…”

Knowledge‐based planning for the radiation therapy treatment plan quality assurance for patients with head and neck cancer

“…This could be remedied by using methods that can predict DVH for both target volumes and OARs. 34 , 35 Recent studies have also proposed methods to predict voxel‐based 3D dose distributions, 36 , 37 , 38 which are additional promising tools for plan quality control and assurance. Our future work may include building RapidPlan models from reoptimized plans according to multiple physician reviews such as ones collected in this study.…”

Machine Learning for Future Subtyping of the Tumor Microenvironment of Gastro-Esophageal Adenocarcinomas

“…Interestingly, ML may also be used to propose risk factors for anastomotic leakage after esophagectomy [120]. Other attempts included using DL to identify optimal dosing of radiotherapy in GEA or defining the optimal target volume and organs at risk [121][122][123][124][125]. A dedicated analysis by Rahman et.…”