Room: AAPM ePoster Library
Purpose: propose to develop deep learning models that can predict Pareto optimal dose distributions by using any given set of beam angles and patient anatomy, as input to train the deep neural networks. We implement and compare two deep learning networks that predict with two different beam configuration modalities.
Methods: used fluence map optimization to generate 500 IMRT plans for 70 patients that sampled the Pareto surface for each patient, for a total of 35,000 plans. We studied and compared two different models, Model I and Model II. Although they both used the same anatomical structures—including the planning target volume (PTV), organs at risk (OARs), and body—these models were designed with two different methods for representing beam angles. Model I directly uses beam angles as a second input to the network as a binary vector. Model II converts the beam angles into beam doses that are conformal to the PTV. We divided the 70 patients into 54 training, 6 validation, and 10 testing patients.
Results: training, the prediction time of each model is less than 1 second. Quantitatively, Model I’s prediction error of 0.043 (confirmation), 0.043 (homogeneity), 0.327 (R50), 2.80% (D95), 3.90% (D98), 0.6% (D50), 1.10% (D2) was lower than that of Model II, which obtained 0.076 (confirmation), 0.058 (homogeneity), 0.626 (R50), 7.10% (D95), 6.50% (D98), 8.40% (D50), 6.30% (D2). Model I also outperformed Model II in terms of the mean dose error and the max dose error on the PTV, bladder, rectum, left femoral head, and right femoral head.
Conclusion: will be able to use the variable-beam Pareto optimal framework to control the tradeoffs between the PTV and OAR weights, as well as the beam number and configurations in real time.
Funding Support, Disclosures, and Conflict of Interest: This study was supported by the National Institutes of Health (NIH) R01CA237269. Authors declare that there is no conflict of interest