We have located links that may give you full text access.
Artificial intelligence-based radiotherapy machine parameter optimization using reinforcement learning.
Medical Physics 2020 October 19
PURPOSE: To develop and evaluate a volumetric modulated arc therapy (VMAT) machine parameter optimization (MPO) approach based on deep-Q reinforcement learning (RL) capable of finding an optimal machine control policy using previous prostate cancer patient CT scans and contours, and applying the policy to new cases to rapidly produce deliverable VMAT plans in a simplified beam model.
METHODS: A convolutional deep-Q network was employed to control the dose rate and multileaf collimator of a C-arm linear accelerator model using the current dose distribution and machine parameter state as input. A Q-value was defined as the discounted cumulative cost based on dose objectives, and experience-replay RL was performed to determine a policy to minimize the Q-value. A two-dimensional network design was employed which optimized each opposing leaf pair independently while monitoring the corresponding dose plane blocked by those leaves. This RL approach was applied to CT and contours from 40 retrospective prostate cancer patients. The dataset was split into training (15 patients) and validation (5 patients) groups to optimize the network, and its performance was tested in an independent cohort of 20 patients by comparing RL-based dose distributions to conformal arcs and clinical intensity modulated radiotherapy (IMRT) delivering a prescription dose of 78 Gy in 40 fractions.
RESULTS: Mean ± SD execution time of the RL VMAT optimization was 1.5 ± 0.2 s per slice. In the test cohort, mean ± SD (P-value) planning target volume (PTV), bladder, and rectum dose were 80.5 ± 2.0 Gy (P < 0.001), 44.2 ± 14.6 Gy (P < 0.001), and 43.7 ± 11.1 Gy (P < 0.001) for RL VMAT compared to 81.6 ± 1.1 Gy, 51.6 ± 12.9 Gy, and 36.0 ± 12.3 Gy for clinical IMRT.
CONCLUSIONS: RL was applied to VMAT MPO using clinical patient contours without independently optimized treatment plans for training and achieved comparable target and normal tissue dose to clinical plans despite the application of a relatively simple network design originally developed for video-game control. These results suggest that extending a RL approach to a full three-dimensional beam model could enable rapid artificial intelligence-based optimization of deliverable treatment plans, reducing the time required for radiotherapy planning without requiring previous plans for training.
METHODS: A convolutional deep-Q network was employed to control the dose rate and multileaf collimator of a C-arm linear accelerator model using the current dose distribution and machine parameter state as input. A Q-value was defined as the discounted cumulative cost based on dose objectives, and experience-replay RL was performed to determine a policy to minimize the Q-value. A two-dimensional network design was employed which optimized each opposing leaf pair independently while monitoring the corresponding dose plane blocked by those leaves. This RL approach was applied to CT and contours from 40 retrospective prostate cancer patients. The dataset was split into training (15 patients) and validation (5 patients) groups to optimize the network, and its performance was tested in an independent cohort of 20 patients by comparing RL-based dose distributions to conformal arcs and clinical intensity modulated radiotherapy (IMRT) delivering a prescription dose of 78 Gy in 40 fractions.
RESULTS: Mean ± SD execution time of the RL VMAT optimization was 1.5 ± 0.2 s per slice. In the test cohort, mean ± SD (P-value) planning target volume (PTV), bladder, and rectum dose were 80.5 ± 2.0 Gy (P < 0.001), 44.2 ± 14.6 Gy (P < 0.001), and 43.7 ± 11.1 Gy (P < 0.001) for RL VMAT compared to 81.6 ± 1.1 Gy, 51.6 ± 12.9 Gy, and 36.0 ± 12.3 Gy for clinical IMRT.
CONCLUSIONS: RL was applied to VMAT MPO using clinical patient contours without independently optimized treatment plans for training and achieved comparable target and normal tissue dose to clinical plans despite the application of a relatively simple network design originally developed for video-game control. These results suggest that extending a RL approach to a full three-dimensional beam model could enable rapid artificial intelligence-based optimization of deliverable treatment plans, reducing the time required for radiotherapy planning without requiring previous plans for training.
Full text links
Related Resources
Trending Papers
Heart failure with preserved ejection fraction: diagnosis, risk assessment, and treatment.Clinical Research in Cardiology : Official Journal of the German Cardiac Society 2024 April 12
Proximal versus distal diuretics in congestive heart failure.Nephrology, Dialysis, Transplantation 2024 Februrary 30
World Health Organization and International Consensus Classification of eosinophilic disorders: 2024 update on diagnosis, risk stratification, and management.American Journal of Hematology 2024 March 30
Efficacy and safety of pharmacotherapy in chronic insomnia: A review of clinical guidelines and case reports.Mental Health Clinician 2023 October
Get seemless 1-tap access through your institution/university
For the best experience, use the Read mobile app
All material on this website is protected by copyright, Copyright © 1994-2024 by WebMD LLC.
This website also contains material copyrighted by 3rd parties.
By using this service, you agree to our terms of use and privacy policy.
Your Privacy Choices
You can now claim free CME credits for this literature searchClaim now
Get seemless 1-tap access through your institution/university
For the best experience, use the Read mobile app