Researchers developed a fast, accurate method combining Monte Carlo simulation and deep learning to generate EPID transmission dose data for radiation therapy quality assurance.
Key Details
- 1A new framework merges GPU-accelerated Monte Carlo (ARCHER) simulations with a SUNet deep learning model for denoising EPID dose data.
- 2Tested with IMRT lung cancer cases, denoising improved SSIM from 0.61 to 0.95 and gamma passing rate from 48.47% to 89.10% for low-particle data.
- 3At 1×10⁷ simulated particles, the method achieved an SSIM of 0.96, GPR of 94.35%, and a processing time of only 1.88 seconds.
- 4The approach enables rapid, patient-specific QA necessary for online adaptive radiation therapy (ART).
- 5Denoised dose images maintain clinical details while reducing noise/graininess, supporting practical deployment in routine workflows.
Why It Matters
This innovation addresses a major bottleneck in clinical radiotherapy QA by dramatically accelerating Monte Carlo-based dose verification while preserving high accuracy. It enables wider implementation of adaptive and real-time therapies, ultimately improving patient safety and workflow efficiency.
Source
EurekAlert
Related News
•EurekAlert
Framework Guides Digital Pathology and AI Integration for Labs
A new lifecycle framework outlines practical steps for sustainable digital pathology and AI program implementation in clinical labs.
•EurekAlert
Recent AI Advances in Digital Breast Pathology: Models, Explainability, and Applications
AI is rapidly transforming breast pathology by improving diagnostic accuracy, workflow efficiency, and precision medicine through advances in deep learning and multimodal models.
•EurekAlert
HKUST Releases Misalignment-Resistant GenAI for Virtual Pathology Staining
HKUST researchers created a generative AI tool that enables high-fidelity virtual staining in histopathology even with imperfectly aligned training image pairs.