Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

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

Quantifying cardiac deformable image registration accuracy and its dosimetric variability for 4D dose accumulation in stereotactic arrhythmia radioablation.

Physics in medicine and biology·2026
Same author

Failure modes and effects analysis of LINAC-based stereotactic arrhythmia radioablation categorized by segmental targets.

Radiation oncology (London, England)·2026
Same author

A non-invasive end-to-end intelligent assistance system for breast ultrasound.

Nature communications·2026
Same author

Exploring Artificial Intelligence in Orthopedic Surgery: A Review of Perception, Decision, and Execution Systems.

Sensors (Basel, Switzerland)·2026
Same author

Precise Localization for Anatomo-Physiological Hallmarks of the Cervical Spine by Using Neural Memory Ordinary Differential Equation.

International journal of neural systems·2024
Same author

Automated Quality Assessment of Medical Images in Echocardiography Using Neural Networks with Adaptive Ranking and Structure-Aware Learning.

International journal of neural systems·2024

Related Experiment Video

Updated: Mar 21, 2026

Dynamic Lung Tumor Tracking for Stereotactic Ablative Body Radiation Therapy
08:17

Dynamic Lung Tumor Tracking for Stereotactic Ablative Body Radiation Therapy

Published on: June 7, 2015

16.3K

Enhancing Exploration and Exploitation in Tumor Treatment Through Action-Guided Deep Reinforcement Learning.

Chengrong Yu1, Zhonglian Wei2, Yuncheng Shen3

  • 1National Pilot School of Software, Yunnan University, Kunming 650091, P. R. China.

International Journal of Neural Systems
|March 20, 2026
PubMed
Summary
This summary is machine-generated.

This study introduces action-guided deep reinforcement learning (AgDRL) for automated radiation therapy planning. The novel approach enhances optimization efficiency and exploration diversity, outperforming existing platforms.

Keywords:
DRLInverse treatment planningaction–state space

More Related Videos

Targeting Neuronal Fiber Tracts for Deep Brain Stimulation Therapy Using Interactive, Patient-Specific Models
14:14

Targeting Neuronal Fiber Tracts for Deep Brain Stimulation Therapy Using Interactive, Patient-Specific Models

Published on: August 12, 2018

9.4K
Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
04:09

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

Published on: October 10, 2018

8.9K

Related Experiment Videos

Last Updated: Mar 21, 2026

Dynamic Lung Tumor Tracking for Stereotactic Ablative Body Radiation Therapy
08:17

Dynamic Lung Tumor Tracking for Stereotactic Ablative Body Radiation Therapy

Published on: June 7, 2015

16.3K
Targeting Neuronal Fiber Tracts for Deep Brain Stimulation Therapy Using Interactive, Patient-Specific Models
14:14

Targeting Neuronal Fiber Tracts for Deep Brain Stimulation Therapy Using Interactive, Patient-Specific Models

Published on: August 12, 2018

9.4K
Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
04:09

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

Published on: October 10, 2018

8.9K

Area of Science:

  • Medical Physics
  • Artificial Intelligence
  • Radiotherapy

Background:

  • Inverse treatment planning optimizes radiation dose delivery for precise tumor targeting and healthy tissue sparing.
  • Manual parameter adjustments are labor-intensive and time-consuming.
  • Deep reinforcement learning (DRL) offers automated planning but often struggles with continuous state-action spaces, leading to inefficient optimization.

Purpose of the Study:

  • To develop an innovative action-guided DRL (AgDRL) approach for automated inverse treatment planning.
  • To enhance exploration and exploitation in DRL for radiation therapy planning by leveraging reward-guided actions.
  • To improve the efficiency and effectiveness of automated inverse planning compared to existing methods.

Main Methods:

  • Implemented an action-guided DRL (AgDRL) approach incorporating both exploitation (high-reward actions) and exploration (low-reward actions).
  • Utilized reward-guided actions to guide optimal action adjustment for exploitation and training resets for exploration.
  • Evaluated the method using DRL metrics (reward gains) and clinical measurements (dose-volume histograms, DVHs) on a rectal cancer dataset.

Main Results:

  • The AgDRL approach significantly improved optimization efficiency through high-reward strategies.
  • Enhanced exploration diversity was achieved via low-reward strategies, broadening the search within the latent state space.
  • The proposed AgDRL method consistently outperformed the MatRad treatment planning optimization platform in quantitative and qualitative experiments.

Conclusions:

  • Action-guided DRL (AgDRL) provides an effective automated solution for inverse treatment planning.
  • The AgDRL strategy successfully balances exploration and exploitation, leading to superior optimization performance.
  • This approach holds significant promise for improving the efficiency and precision of radiation therapy planning.