Abstract:Plan quality variability is a known problem in clinical IMRT. The objective of this work was to develop a method to efficiently train accurate predictive DVH (pDVH) models for post-operative endometrial cancer patients when plan quality variations are present in the training cohort. A previously developed framework to predict achievable OAR DVHs that correlates expected doses to voxel distances from a PTV surface was used to create pDVH models for post-operative endometrial cancer patients. A random sample of 20 clinically treated IMRT plans with identical clinical objectives was used to train raw pDVH models for rectum, bladder, and sigmoid colon. A sum of residuals (SR) analysis quantifying the integrated difference between the clinical DVHs and the pDVHs showed larger plan quality variation than seen in previously modeled sites (prostate, head-and-neck). This cohort was used to test a method to train pDVH models that accurately predict OAR DVHs without replanning every sub-optimal patient. Initial training plans were ranked using the raw pDVH model's mean SR for rectum, bladder, and sigmoid. The 5 worst ranked plans were replanned, improving OAR DVHs while maintaining PTV V95% > 100% and V115% < 0% per institutional standards. Replan_25% pDVH models were trained with the 5 best ranked plans and the 5 replanned outliers. The entire 20 patient cohort was replanned as a benchmark. These replans were also used to create a replan_100% pDVH model. An institutional metric for plan quality (V40) was used to quantify clinical gains in rectum, bladder, and sigmoid. All pDVH models were compared to the replan sample using dV40 = V40(replan) - V40(pred) to assess clinical significance and mean SR to quantify DVH prediction accuracy. As shown in the Table, the average reduction in V40 between the clinical plan and the replan demonstrates large clinical improvements. The raw pDVH models underestimated achievable V40 compared to the more accurate replan_100% and replan_25% pDVH models which demonstrated statistically identical dV40 predictions across all organs. Comparable SR values between replan_25% and replan_100% exhibit equivalent model performance.Poster Viewing Abstract 3099; TableResults for V40, dV40, and mean SR for rectum, bladder, and sigmoid colonOrganV40(orig)-V40(replan)(mean +/− SD)dV40 (mean +/− SD)SR (mean +/− SD)Raw modelReplan_25% modelReplan_100% modelRaw modelReplan_25% modelReplan_100% modelRectum8.3% +/− 8.4%−4.2% +/− 5.1%2.8% +/− 4.8%2.1% +/− 4.5%−0.044 +/− 0.0400.005 +/− 0.0410.018 +/− 0.044Bladder14.2% +/− 12.3%−3.0% +/− 8.5%−2.1% +/− 9.2%−0.1% +/− 8.6%−0.037 +/− 0.0400.003 +/− 0.0380.007 +/− 0.039Sigmoid10.5% +/− 12.4%−6.9% +/− 8.0%2.1% +/− 7.6%1.2% +/− 7.4%−0.027 +/− 0.042−0.016 +/− 0.0440.044 +/− 0.047 Open table in a new tab The results of this study validate an efficient method to obtain accurate predictions of near-optimal OAR DVHs when large plan quality variations are present in the training sample.

Effect of Different Volume Dose Calculators on Configuring a DVH-Estimation Model for Knowledge-Based Lung SABR Planning

Impact of inconsistent dose calculation algorithms on modeling and use of automated treatment planning for lung stereotactic body radiotherapy

The Benefits Evaluation Of Abdominal Deep Inspiration Breath Hold Based On Knowledge-Based Radiotherapy Treatment Planning For Left-Sided Breast Cancer

Robustness of a Pretrained Dose-Volume Histogram Estimation Model on Knowledge-Based Planning of Other Types

A dosimetric evaluation of knowledge-based VMAT planning with simultaneous integrated boosting for rectal cancer patients.

Modeling Organs at Risk Dose Sparing in Head and Neck Tomotherapy and Imrt Plans: A Comparison Study

Difference in target dose distributions between Acuros XB and collapsed cone convolution/superposition and the impact of the tumor locations in clinical cases of stereotactic ablative body radiotherapy for lung cancer

A feasibility study of local adaptation of Lung SBRT RapidPlan commercial model

SU-F-T-348: the Impact of Model Library Population on RapidPlan Based Dose-Volume Histograms (dvhs) Prediction for Rectal Cancer Patients Treated with Volumetric-Modulated Radiotherapy (VMAT)

A feasibility and dosimetric study of configuring a Varian RapidPlan model based on Elekta cases

The Model Training of Knowledge-Based Radiotherapy Treatment Planning and Its Preliminary Application

Efficient Method to Train Pdvh Models with Plan Quality Variation Present in the Training Cohort

Study of PlanIQ Feasibility DVH Usage to Improve Cervical Cancer Treatment Plan

Efficacy Of Epid-Based In Vivo Dosimetry And Calibrated Cbct Images For A Timely Lung Cancer Replanning

Knowledge-based Semi-Automated Optimization of Intensity-Modulated Radiotherapy Plans Using a Volume Modulated Arc Therapy-Configured Model

Applying a RapidPlan model trained on a technique and orientation to another: a feasibility and dosimetric evaluation.

Dosimetric Comparison of Different Algorithms in Stereotactic Body Radiation Therapy (SBRT) Plan for Non-Small Cell Lung Cancer (NSCLC)

Dosimetry and efficiency comparison of knowledge-based and manual planning using volumetric modulated arc therapy for craniospinal irradiation

Implementation of a Knowledge-Based Treatment Planning Model for Cardiac-Sparing Lung Radiation Therapy

Enhancing dosimetric practices through knowledge-based predictive models: a case study on VMAT prostate irradiation

Clinical Evaluation Of Rtog 0813 Dosimetric Criteria For Lung Sabr Using Iplan Monte Carlo Algorithm