EP-2199: Developing Head-and-Neck IMRT Organ-at-risk Sparing Decisions Support System using Model Tree
Year: 2018
Type: conference-abstract
Abstract: Purpose or ObjectiveTo develop a system that captures head-and-neck (HN) intensity modulated radiation therapy (IMRT) cases with single-lateral and bi-lateral parotid sparing decisions using model tree. Material and Methods Seventy-three HN IMRT cases were included in this study, with 45 bilateral parotid sparing (90 spared parotids) and 28 single parotid sparing (28 spared parotids) cases.The model tree training: the combined set of 80 bilaterally and 23 single-laterally spared parotids.Baseline model training: 1. bi-lateral sparing model was trained using 80 bilaterally spared parotids 2. single-lateral sparing model was trained using 23 single-laterally spared parotids.3. non-differentiated model was trained using the combined set of 80 bi-laterally and 23 single-laterally spared parotids Model evaluation: The remaining 10 bi-laterally spared parotids and 5 single-laterally spared parotids were used to validate the model tree and the three baseline models.The experiment was repeated 20 times using bootstrap.The Weighted Sum of Residual (WSR) was used to evaluate the accuracy of dose-volume histogram (DVH) prediction.The difference between predicted and the clinically planned parotid D50% was assessed. ResultsThe mean WSR of the validation bi-lateral cases was 0.021, 0.034, 0.026 and 0.022 for the bi-lateral model, the single-lateral model, the non-differentiated model and model tree, respectively.Model tree predicted similarly as the bi-lateral model based on the mean WSR while single-lateral model predicted unfavorably for the bi-lateral cases.For the single-lateral validation cases, the mean WSR was -0.007, 0.010, -0.001 and 0.004 for the bi-lateral model, the single-lateral model, the nondifferentiated model and model tree, respectively.The mean D50% difference of the bi-lateral model, the single-lateral model, the nondifferentiated model and model tree were -0.40Gy, -2.00Gy, -1.12Gy and -0.87Gy for bi-lateral cases, and 1.63Gy, -0.13Gy, 0.89Gy and 0.60Gy for single-lateral cases.Model tree improved prediction accuracy for both bi-lateral and single-lateral cases over non-differentiated model. ConclusionAn organ-at-risk sparing decision support system was proposed for HN treatment planning.Results showed the deficiency of using a single regression model for HN IMRT modeling because different parotid sparing decisions were involved in clinical cases.Results suggest that the model tree is effective in modeling HN IMRT cases with different parotid sparing decisions.The decision support system could aid make sparing decisions for HN patients based on patient specific anatomy.
Source: Radiotherapy and Oncology
Language: English
Subfield: Radiation
Field: Physics and Astronomy
Domain: Physical Sciences
Open Access status: hybrid
Estimated APC paid: $3,770
OpenAlex ID: W2809344782