A Deep Learning-Based Intelligent Radiotherapy Quality Assessment Method and System
The intelligent radiotherapy quality assessment method based on deep learning, utilizing deep residual neural networks and regional attention weighting mechanisms, overcomes the shortcomings of traditional methods in handling complex and changing data in three-dimensional spatial topology, achieving higher assessment accuracy and adaptability. In particular, it significantly improves the robustness and accuracy of the results in dose distribution calculations among complex anatomical regions.
CN122091128APending Publication Date: 2026-05-26SUPERACCURACY SCIENCE & TECHNOLOGY CO LTD
View PDF 0 Cites 0 Cited by
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SUPERACCURACY SCIENCE & TECHNOLOGY CO LTD
- Filing Date
- 2026-02-13
- Publication Date
- 2026-05-26
Smart Images
Figure CN122091128A_ABST
Abstract
This invention discloses a deep learning-based intelligent radiotherapy quality assessment method and system. The intelligent radiotherapy quality assessment method includes: 1) constructing a deep residual neural network structure and performing supervised training based on training samples to obtain a deep learning model for radiotherapy quality assessment; 2) acquiring dose distribution maps and corresponding anatomical structure region images of the radiotherapy plan and performing preprocessing to obtain a registration image dataset; 3) constructing a standardized dose distribution vector based on the registration image dataset; 4) constructing a corresponding graph structure based on the standardized dose distribution vector and constructing a multi-scale graph structure set; 5) constructing a region attention weighting mechanism and performing sparse reconstruction to obtain a weighted embedding feature vector representing the spatial topological relationship between each node of the graph structure; 6) inputting the weighted embedding feature vector representing the spatial topological relationship between each node of the graph structure into the deep learning model for radiotherapy quality assessment to generate a radiotherapy quality assessment result.
Need to check novelty before this filing date? Find Prior Art