Knowledge-based planning algorithm for lung SBRT with robust Bayesian stochastic frontier analysis and missing data management.

PURPOSE: A knowledge-based planning technique is developed based on Bayesian stochastic frontier analysis. A novel missing data management is applied in order to handle missing organs-at-risk and work with a complete dataset. METHODS: Geometric metrics are used to predict DVH metrics for lung SBRT with a retrospective database of 299 patients. In total, 16 DVH metrics were predicted for the main bronchus, heart, esophagus, spinal cord PRV, great vessels, and chest wall. The predictive model is tested on a test group of 50 patients. RESULTS: Mean difference between the observed and predicted values ranges between 1.5 ± 1.9 Gy and 4.9 ± 5.3 Gy for the spinal cord PRV D0.35cc and the main bronchus D0.035cc, respectively. CONCLUSIONS: The missing data model implanted in the predictive model is robust in the estimation of the parameters. Bayesian stochastic frontier analysis with missing data management can be used to predict DVH metrics for lung SBRT treatment planning.

Stochastic frontier analysis as knowledge-based model to improve sparing of organs-at-risk for VMAT-treated prostate cancer.

Stochastic frontier analysis (SFA) is used as a novel knowledge-based technique in order to develop a predictive model of dosimetric features from significant geometric parameters describing a patient morphology. 406 patients treated with VMAT for prostate cancer were analyzed retrospectively. Cases were divided into three prescription-based groups. Seven geometric parameters are extracted to characterize the relationship between the organs-at-risk (bladder and rectum) with the planning volume (PTV). In total, 37 dosimetric parameters are tested for these two OARs. SFA allows the determination of the minimum achievable dose to the OAR based on the geometric parameters. Stochastic frontiers are determined with a maximum likelihood estimation technique. The SFA model was tested using validation cohort (30 patients with prescribed dose between 60 and 70 Gy) where 77% (23 out of 30) of the predicted DVHs present a 5% or less dose deterioration for the bladder and rectum with the planned DVH. SFA can be used in EBRT planning as a predictive model based on anatomical features of previously treated plans.

A stochastic frontier analysis for enhanced treatment quality of high-dose-rate brachytherapy plans.

The purpose of the present study is to develop patient specific unbiased quality control (QC) models for high dose rate (HDR) brachytherapy plans. The proposed models are based on the stochastic frontier analysis formalism, a method of economic modeling. They act as a QC tool by predicting before the treatment planning process starts, the dosimetric coverage achievable for a HDR brachytherapy prostate plan. The geometric parameters considered in developing the models were: patient clinical target volume (CTV), organs at risk (OAR) volume, the bidirectional Hausdorff distance between CTV and OARs, and a fourth parameter measuring the catheters degree of non-parallelism within the target volume. Dosimetry parameters of interest are V100 for the CTV, V75 (bladder, rectum) and D10 (urethra). Results show that the built models can provide valuable information on the personalization of the optimization process based on the patient geometric parameters. The impact on the quality plan due to the planner's experience variability and judgment can be reduced by using those models, since the planner will attempt to achieve dosimetric parameters predicted by the models. Furthermore, the models provide information on the better trade-off between the target volume coverage and OARs sparing that can be achieved, regardless of the planner's experience; the latter being achieved by moving each plan at least around their respective frontier for V100, V75 and D10. The shortfall of the dosimetric parameters values computed by the treatment planning system (TPS) from those predicted by the models for a proportion of plans in the dataset reveals that optimized plans from a TPS, even clinically acceptable, are not necessarily the best that could be achieved. These represent 83% of plans in the training set for the target volume coverage (V100), ∼50% for the bladder (V75) and ∼72% for the urethra (D10).