ESTRO 2020 Abstract book
S124 ESTRO 2020
clinic. We are working to expand use of the system to other disease sites. OC-0222 Deep Neural Network and Transfer Learning for DVH prediction in VMAT prostate treatments E.M. Ambroa Rey 1 , J. Pérez-Alija 2 , P. Gallego 2 1 Consorci Sanitari de Terrassa, Medical Physics Unit- Radiation Oncology, Terrassa, Spain ; 2 Hospital de la Santa Creu i Sant Pau, Medical Physics, Barcelona, Spain Purpose or Objective Volumetric modulated arc therapy (VMAT) has been used widely to provide highly conformal plans. However, treatment planning has increased in complexity and has become a time-consuming process. The purpose of this work is to establish a convolutional neural network (CNN) model for the prediction of rectum and bladder dose-volume histograms (DVH) in prostate patients treated with a VMAT technique. Material and Methods A total of 145 patients with intermediate or high-risk prostate cancer treated with simultaneous integrated boost (SIB) were selected for this study. The prescribed dose was 50.4 Gy and 70Gy in 28 fractions to the pelvic and prostate area, respectively. Data were split into two sets: 120 and 25 patients, respectively. Besides, the first set was partitioned in training, validation, and test, each with 100, 10, and 10 patients. The second set was used for final validation. We use transfer learning in combination with a VGG-16 network. VGG-16 is a deep CNN pre-trained with the ImageNet image dataset (1.2 million natural images of 1000 object categories). We dropped the fully-connected layers from the VGG-16 and added a new fully-connected neural network. The inputs for the CNN were a 2D image of the volumes contoured in the CT. We only retained the geometrical information of every CT. The outputs were the corresponding dose maps. Rectum and bladder DVHs were computed for each patient summing up all the dose- volume information in every slice. To ensure the quality of the data, we selected all potential outliers and proceeded to re-optimize them. The already trained CNN was tested using the second test of patients (25). All patients were re-optimized by the same operator unfamiliar with the results of the prediction. A confusion matrix was used to report the number of false positives, false negatives, true positives, and true negatives. Results Figure I shows the clinically approved, replanned, and predicted bladder and rectum DVHs for three representative cases. The results demonstrated that for most cases the actual DVH, either clinically approved or replanned, was within the DVH predicted by our model.
Results Among all the treated SBRT ECHO plans, 400 were for 24Gy in a single fraction, 1165 for 27Gy in three fractions, and the rest for various prescriptions doses with varied fractionations. Most plans were for paraspinal tumors with 174, 762 and 426 in cervical, thoracic and lumbosacral spine respectively. The median PTV size was 84 cc (range 7 - 633). The median time to produce one ECHO plan was 64 minutes (range 11-340), largely dependent on the field sizes. Over 90% of cases required just one run to produce a clinically accepted plan, the rest required additional run of ECHO with parameter tweak for physician special requests. All plans produced met or bettered the institutional clinical criteria. Excellent target coverage was achieved with PTV V100% averaged 93.0% ± 3.1% and PTV V95% averaged 98.2% ± 1.9%. ECHO plans were also highly conformal with Paddick Conformity Index averaged 0.86 ± 0.06. All ECHO plans were delivered after passing intuitional quality assurance process. We are currently using ECHO to generate over 80 SBRT plans a week in our clinic.
Conclusion We successfully implemented a constrained hierarchial optimization method in our clinic for automated SBRT planning. ECHO has achieved the expected goals of producing consistently high quality clinical plans, in a reasonable time, that push normal tissue sparing as much as possible while respecting disease treatment goals. This has further resulted in an improved clinical workflow and shorter times between simulation and treatment in our
Made with FlippingBook - Online magazine maker