Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

The Association Between Active Aging and Health-Related Quality of Life.

Geriatrics (Basel, Switzerland)·2026

Same author

A Transformer-Based Machine Learning Framework for Risk Stratification of Left Bundle Branch Block After Transcatheter Aortic Valve Replacement.

Diagnostics (Basel, Switzerland)·2026

Same author

Evaluating Curvature-Induced Variation in Deep Learning-Based Beamforming for Flexible Transducers in Ultrasound-Guided Radiation Therapy.

Bioengineering (Basel, Switzerland)·2026

Same author

Correlation between changes in substructure diffusion tensor imaging and neurocognitive outcomes for pediatric brain tumor survivors.

Neuro-oncology advances·2026

Same author

Prospective operational feasibility, safety, and workflow of magnetic resonance-guided tracking technology for interstitial gynecologic brachytherapy.

Brachytherapy·2026

Same author

Characterization of the cell division-associated peptidoglycan amidase AmiA of <i>Chlamydia trachomatis</i>.

Journal of bacteriology·2026

Same journal

Correction to "On the shape of the radiation survival curve in tumor spheroids: The role of oxygen heterogeneity".

Medical physics·2026

Same journal

Multi-view constrained semi-supervised vertebra detection for 3D ultrasound spine volume.

Medical physics·2026

Same journal

Accuracy of quantitative <sup>177</sup>Lu SPECT/CT imaging: A systematic review.

Medical physics·2026

Same journal

Physics-constrained dual-domain network for CBCT reconstruction from orthogonal X-rays in gynecologic radiotherapy.

Medical physics·2026

Same journal

Decomposition-based harmonization for quantitative PET imaging across scanners and radiotracers.

Medical physics·2026

Same journal

Development and evaluation of an in vivo dose-based monitoring system for electron FLASH radiation therapy.

Medical physics·2026

See all related articles

Search research articles

Related Experiment Video

Updated: Dec 5, 2025

Author Spotlight: Improving Radiation Therapy Access with Radiation Planning Assistant

Author Spotlight: Improving Radiation Therapy Access with Radiation Planning Assistant

Published on: October 6, 2023

Artificial intelligence-based radiotherapy machine parameter optimization using reinforcement learning.

William Thomas Hrinivich¹, Junghoon Lee¹

¹Department of Radiation Oncology and Molecular Radiation Sciences, Johns Hopkins University, Baltimore, MD, 21287, USA.

Medical Physics

|October 18, 2020

Summary

This summary is machine-generated.

This study introduces a novel reinforcement learning (RL) approach for optimizing volumetric modulated arc therapy (VMAT) machine parameters. The AI model rapidly generates high-quality VMAT plans for prostate cancer, comparable to traditional methods.

Keywords:

artificial intelligence deep-Q learning optimization reinforcement learning treatment planning volumetric modulated arc therapy

More Related Videos

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

Proton Therapy Delivery and Its Clinical Application in Select Solid Tumor Malignancies

Proton Therapy Delivery and Its Clinical Application in Select Solid Tumor Malignancies

Published on: February 6, 2019

Related Experiment Videos

Last Updated: Dec 5, 2025

Author Spotlight: Improving Radiation Therapy Access with Radiation Planning Assistant

Author Spotlight: Improving Radiation Therapy Access with Radiation Planning Assistant

Published on: October 6, 2023

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

Proton Therapy Delivery and Its Clinical Application in Select Solid Tumor Malignancies

Proton Therapy Delivery and Its Clinical Application in Select Solid Tumor Malignancies

Published on: February 6, 2019

Area of Science:

Medical Physics
Radiation Oncology
Artificial Intelligence in Medicine

Background:

Volumetric Modulated Arc Therapy (VMAT) is a complex radiotherapy technique requiring precise machine parameter optimization.
Current VMAT planning can be time-consuming, necessitating efficient automated solutions.
Deep reinforcement learning (RL) offers a potential avenue for automating and optimizing treatment planning.

Purpose of the Study:

To develop and evaluate a deep-Q reinforcement learning (RL) based machine parameter optimization (MPO) approach for VMAT.
To enable rapid generation of deliverable VMAT plans for prostate cancer using a simplified beam model.
To assess the feasibility of applying RL to optimize VMAT without prior treatment plans for training.

Main Methods:

A convolutional deep-Q network was utilized to control linear accelerator parameters (dose rate, multileaf collimator) based on dose distribution and machine state.
Experience-replay RL was performed to minimize a Q-value defined by cumulative dose objective costs.
A 2D network optimized opposing leaf pairs independently, applied to CT scans and contours from 40 prostate cancer patients.

Main Results:

The RL VMAT optimization achieved a mean execution time of 1.5 ± 0.2 seconds per slice.
In the test cohort, RL VMAT demonstrated comparable planning target volume (PTV) doses (80.5 ± 2.0 Gy) to clinical IMRT (81.6 ± 1.1 Gy).
RL VMAT showed improved bladder doses (44.2 ± 14.6 Gy) and comparable rectum doses (43.7 ± 11.1 Gy) versus IMRT (51.6 ± 12.9 Gy and 36.0 ± 12.3 Gy, respectively).

Conclusions:

Reinforcement learning (RL) successfully optimized VMAT machine parameters using clinical data without pre-optimized plans.
The RL approach achieved target and normal tissue dose distributions comparable to clinical intensity-modulated radiotherapy (IMRT) plans.
Extending RL to a 3D model holds promise for rapid, AI-driven radiotherapy plan optimization, reducing planning time.