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 Experiment Video

Updated: May 28, 2026

Three-Dimensional Reconstruction for the Whole Lung with Early Multiple Pulmonary Nodules
07:53

Three-Dimensional Reconstruction for the Whole Lung with Early Multiple Pulmonary Nodules

Published on: October 13, 2023

Deep-Learning-Based 3D Dose Distribution Prediction for VMAT Lung Cancer Treatment Using an Enhanced UNet3D

Philip Chung Yin Mak1,2, Luoyi Kong1, Lawrence Wing Chi Chan1

  • 1Department of Health Technology and Informatics, The Hong Kong Polytechnic University, Hong Kong SAR, China.

Bioengineering (Basel, Switzerland)
|May 27, 2026
PubMed
Summary

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

Artificial Intelligence for Computer-Aided Detection in Biomedical Applications.

Bioengineering (Basel, Switzerland)·2026
Same author

Exploring Sarcopenic Obesity in the Cancer Setting: Insights from the National Health and Nutrition Examination Survey on Prognosis and Predictors Using Machine Learning.

Bioengineering (Basel, Switzerland)·2025
Same author

Kolmogorov-Arnold Network Model Integrated with Hypoxia Risk for Predicting PD-L1 Inhibitor Responses in Hepatocellular Carcinoma.

Bioengineering (Basel, Switzerland)·2025
Same author

Contrast-Enhanced Ultrasound-Based Radiomics for the Prediction of Axillary Lymph Nodes Status in Breast Cancer.

Cancer reports (Hoboken, N.J.)·2024
Same author

Comparative Analysis of Repeatability in CT Radiomics and Dosiomics Features under Image Perturbation: A Study in Cervical Cancer Patients.

Cancers·2024
Same author

Evaluating the Diagnostic Performance of Large Language Models on Complex Multimodal Medical Cases.

Journal of medical Internet research·2024

This study introduces Enhanced UNet3D, an AI model that accurately generates 3D radiation dose distributions for lung cancer patients undergoing Volumetric Modulated Arc Therapy (VMAT). The model significantly speeds up treatment planning, improving efficiency and plan quality.

Area of Science:

  • Medical Physics
  • Artificial Intelligence in Radiation Oncology
  • Radiotherapy Treatment Planning

Background:

  • Volumetric Modulated Arc Therapy (VMAT) for lung cancer requires complex, time-consuming manual planning by radiation oncologists.
  • Current VMAT planning involves iterative trial-and-error, leading to variable plan quality and high computational demands.
  • There is a need for automated, accurate, and efficient methods to improve VMAT treatment plan quality and reduce planning time.

Purpose of the Study:

  • To develop and validate an AI-based model, Enhanced UNet3D, for rapid and accurate generation of 3D dose distributions in lung cancer VMAT.
  • To introduce a novel composite objective function, Enhanced Combined Loss (ECLoss), for improved dose shaping and spatial falloff.
  • To assess the clinical feasibility and efficiency of the Enhanced UNet3D model in VMAT treatment planning workflows.
Keywords:
3D U-NetDVHVMATattentioncomposite lossdeep learningdose predictionlung cancerradiotherapy planning

More Related Videos

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

Related Experiment Videos

Last Updated: May 28, 2026

Three-Dimensional Reconstruction for the Whole Lung with Early Multiple Pulmonary Nodules
07:53

Three-Dimensional Reconstruction for the Whole Lung with Early Multiple Pulmonary Nodules

Published on: October 13, 2023

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

Main Methods:

  • Developed Enhanced UNet3D, a 3D convolutional neural network with an encoder-decoder architecture, residual connections, and attention modules.
  • Implemented a new composite objective function (ECLoss) combining SharpLoss (DVH-guided) and 3D gradient regularization.
  • Trained and validated the model on a retrospective dataset of 170 lung cancer VMAT plans.

Main Results:

  • The Enhanced UNet3D model achieved high accuracy in dose distribution prediction, with a Mean Absolute Error (MAE) of 0.238 ± 0.075 Gy and SSIM of 0.970 ± 0.005 on the test set.
  • Near-real-time inference was achieved at approximately 0.5 seconds per patient, drastically reducing computational time.
  • The model demonstrated effective dose shaping within the planning target volume (PTV) and realistic spatial dose falloff.

Conclusions:

  • The Enhanced UNet3D model with ECLoss provides a clinically feasible solution for rapid evaluation and quality assurance of lung cancer VMAT plans.
  • This AI-driven approach can significantly reduce the need for manual trial-and-error in VMAT planning, enhancing efficiency.
  • The model shows potential to improve the consistency and quality of radiotherapy treatment plans, benefiting patient care.