What question did this study set out to answer?

To develop an Enhanced UNet3D architecture for accurate 3-D dose distribution prediction in VMAT lung cancer treatment.

April 26, 2026Bioengineering0 citationsOpen Access

Deep-Learning-Based 3D Dose Distribution Prediction for VMAT Lung Cancer Treatment Using an Enhanced UNet3D Architecture with Composite Loss Functions

PMPhilip Chung Yin MakHong Kong Polytechnic University LKLuoyi KongHong Kong Polytechnic University LCLawrence Wing Chi ChanHong Kong Polytechnic University

Key Points

To develop an Enhanced UNet3D architecture for accurate 3-D dose distribution prediction in VMAT lung cancer treatment.
Used a retrospective dataset of 170 VMAT plans for model training and validation.
Implemented Enhanced Combined Loss (ECLoss) for improved dose prediction and voxel-level imbalance handling.
Evaluated model performance on a test set of 14 plans for accuracy and computational efficiency.
Achieved a Mean Absolute Error (MAE) of 0.238 ± 0.075 Gy on the test set.
Demonstrated a structural similarity index measure (SSIM) of 0.970 ± 0.005.
Provided near-real-time inference at approximately 0.5 seconds per patient, reducing computational resource needs.

Abstract

The high complexity of radiation therapy for lung cancer necessitates effective planning of advanced treatments such as Volumetric Modulated Arc Therapy (VMAT) by radiation oncologists. The current VMAT treatment planning process typically involves extensive manual interaction and a time-consuming, trial-and-error, iterative approach that requires planners’ experience. This can lead to varying levels of plan quality. To improve the quality of radiotherapy treatment plans quickly and accurately, this research presents a new architecture, Enhanced UNet3D, to generate three-dimensional (3-D) dose distributions for lung cancer patients. Enhanced UNet3D utilises a symmetric encoder–decoder architecture with residual connections and a target region-attention module to achieve high accuracy in dose shaping within the PTV. A new composite objective function, Enhanced Combined Loss (ECLoss), that includes both SharpLoss, a structure-aware DVH-guided loss, and 3D gradient regularisation, has been developed to address voxel-level class imbalance and achieve realistic spatial dose falloff. This research utilised a retrospective dataset of 170 VMAT plans to train and validate the proposed model. On the test set (n = 14), the model demonstrated exceptional overall accuracy, with a Mean Absolute Error (MAE) of 0.238 ± 0.075 Gy and a structural similarity index measure (SSIM) of 0.970 ± 0.005. Moreover, the model can perform near-real-time inference at approximately 0.5 s per patient, representing a significant reduction in computational resources compared to other architectures. Therefore, these results demonstrate that the Enhanced UNet3D model with ECLoss is a clinically feasible tool for the rapid evaluation and quality assurance of radiotherapy treatment plans and may reduce the need for manual trial-and-error in VMAT workflows.

Ask AI

Helpful

Bookmark

View Full Paper