Dataset Information

Beam Angle Optimization for Double-Scattering Proton Delivery Technique Using an Eclipse Application Programming Interface and Convolutional Neural Network.

ABSTRACT: To automatically identify optimal beam angles for proton therapy configured with the double-scattering delivery technique, a beam angle optimization method based on a convolutional neural network (BAODS-Net) is proposed. Fifty liver plans were used for training in BAODS-Net. To generate a sequence of input data, 25 rays on the eye view of the beam were determined per angle. Each ray collects nine features, including the normalized Hounsfield unit and the position information of eight structures per 2° of gantry angle. The outputs are a set of beam angle ranking scores (S _beam) ranging from 0° to 359°, with a step size of 1°. Based on these input and output designs, BAODS-Net consists of eight convolution layers and four fully connected layers. To evaluate the plan qualities of deep-learning, equi-spaced, and clinical plans, we compared the performances of three types of loss functions and performed K-fold cross-validation (K = 5). For statistical analysis, the volumes V_27Gy and V_30Gy as well as the mean, minimum, and maximum doses were calculated for organs-at-risk by using a paired-samples t-test. As a result, smooth-L1 loss showed the best optimization performance. At the end of the training procedure, the mean squared errors between the reference and predicted S _beam were 0.031, 0.011, and 0.004 for L1, L2, and smooth-L1 loss, respectively. In terms of the plan quality, statistically, Plan_BAO has no significant difference from Plan_Clinic (P >.05). In our test, a deep-learning based beam angle optimization method for proton double-scattering treatments was developed and verified. Using Eclipse API and BAODS-Net, a plan with clinically acceptable quality was created within 5 min.

SUBMITTER: Cheon W

PROVIDER: S-EPMC8476903 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Beam angle optimization (BAO) by far remains an important and challenging problem in external beam radiation therapy treatment planning. Conventional BAO algorithms discussed in previous studies all focused on photon-based therapies. Impact of BAO on proton therapy is important while proton therapy increasingly receives great interests. This study focuses on potential benefits of BAO on intensity-modulated proton therapy (IMPT) that recently began available to clinical cancer treatment.The authors have developed a novel uncertainty incorporated BAO algorithm for IMPT treatment planning in that IMPT plan quality is highly sensitive to uncertainties such as proton range and setup errors. A linear programming was used to optimize robust intensity maps to scenario-based uncertainties for an incident beam angle configuration. Unlike conventional intensity-modulated radiation therapy with photons (IMXT), the search space for IMPT treatment beam angles may be relatively small but optimizing an IMPT plan may require higher computational costs due to larger data size. Therefore, a deterministic local neighborhood search algorithm that only needs a very limited number of plan objective evaluations was used to optimize beam angles in IMPT treatment planning.Three prostate cancer cases and two skull base chordoma cases were studied to demonstrate the dosimetric advantages and robustness of optimized beam angles from the proposed BAO algorithm. Two- to four-beam plans were optimized for prostate cases, and two- and three-beam plans were optimized for skull base cases. By comparing plans with conventional two parallel-opposed angles, all plans with optimized angles consistently improved sparing at organs at risks, i.e., rectum and femoral heads for prostate, brainstem for skull base, in either nominal dose distribution or uncertainty-based dose distributions. The efficiency of the BAO algorithm was demonstrated by comparing it with alternative methods including simulated annealing and genetic algorithm. The numbers of IMPT plan objective evaluations required were reduced by up to a factor of 5 while the same optimal angle plans were converged in selected comparisons.Uncertainty incorporated BAO may introduce pronounced improvement of IMPT plan quality including dosimetric benefits and robustness over uncertainties, based on the five clinical studies in this paper. In addition, local search algorithms may be more efficient in finding optimal beam angles than global optimization approaches for IMPT BAO.

Project description:PurposeTo compare the dosimetric performances of intensity-modulated proton therapy (IMPT) plans generated with two different beam angle configurations (the Right-Left oblique posterior beams and the Superior-Inferior oblique posterior beams) for the treatment of distal esophageal carcinoma in the presence of uncertainties and interplay effect.Methods and materialsTwenty patients' IMPT plans were retrospectively selected, with 10 patients treated with the R-L oblique posterior beams (Group R-L) and the other 10 patients treated with the S-I oblique posterior beams (Group S-I). Patients in both groups were matched by their clinical target volumes (CTVs-high and low dose levels) and respiratory motion amplitudes. Dose-volume-histogram (DVH) indices were used to assess plan quality. DVH bandwidth was calculated to evaluate plan robustness. Interplay effect was quantified using four-dimensional (4D) dynamic dose calculation with random respiratory starting phase of each fraction. Normal tissue complication probability (NTCP) for heart, liver, and lung was calculated, respectively, to estimate the clinical outcomes. Wilcoxon signed-rank test was used for statistical comparison between the two groups.ResultsCompared with plans in Group R-L, plans in Group S-I resulted in significantly lower liver Dmean and lung V30Gy[RBE] with slightly higher but clinically acceptable spinal cord Dmax . Similar plan robustness was observed between the two groups. When interplay effect was considered, plans in Group S-I performed statistically better for heart Dmean and V30Gy[RBE] , lung Dmean and V5Gy[RBE] , and liver Dmean , with slightly increased but clinically acceptable spinal cord Dmax . NTCP for liver was significantly better in Group S-I.ConclusionsIMPT plans in Group S-I have better sparing of liver, heart, and lungs at the slight cost of spinal cord maximum dose protection, and are more interplay-effect resilient compared to IMPT plans in Group R-L. Our study supports the routine use of the S-I oblique posterior beams for the treatments of distal esophageal carcinoma.

Dataset Information

Beam Angle Optimization for Double-Scattering Proton Delivery Technique Using an Eclipse Application Programming Interface and Convolutional Neural Network.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets