Automatic IMRT planning via static field fluence prediction (AIP-SFFP): a deep learning algorithm for real-time prostate treatment planning.
Researchers
Journal
Modalities
Models
Abstract
To develop a Deep Learning (DL) based algorithm, Automatic IMRT Planning via Static Field Fluence Prediction (AIP-SFFP), for automated prostate IMRT planning with real-time planning efficiency.
AIP-SFFP generates a prostate IMRT plan through predictions of fluence maps using the patient anatomy. No inverse planning is required. AIP-SFFP centralizes a custom-build deep learning neuro network for fluence map prediction. Predictions are imported to a commercial treatment planning system for dose calculation and plan generation. AIP-SFFP was demonstrated for prostate IMRT simultaneously-integrated-boost (SIB) planning (58.8Gy/70Gy to PTV58.8Gy/PTV70Gyin 25 fx). Training data was generated from 106 patients using a knowledge-based planning (KBP) plan generator. Two types of 2D projection images were designed to represent structures’ sizes and locations, and a total of 8 projections were utilized to describe targets and organs-at-risk. Projections at 9 template beam angles were stacked as inputs for AI training. 14 patients were used as independent tests. The generated test plans were compared with the plans from the KBP training plan generator and clinic practice.
After normalization (PTV70GyV70Gy=95%), all 14 AI plans met institutional criteria. The coverage of PTV58.8Gyin AI plans was comparable to KBP and Clinic plans without statistical significance. BODY D1ccand rectum D0.1ccof AI plans were slightly higher (<1Gy) compared to KBP and Clinic plans; in contrast, bladder D1ccand other rectum and bladder low dose in AI plans were slightly improved without clinical relevance. The overall isodose distribution in AI plans was comparable with KBP plans and clinical plans. AIP-SFFP generated each test plan within 20 seconds including prediction and dose calculation.
AIP-SFFP was successfully developed for prostate IMRT planning. AIP-SFFP demonstrated good overall plan qualities and real-time efficiency. Holding great promises, AIP-SFFP will be investigated for immediate clinical application.
© 2020 Institute of Physics and Engineering in Medicine.