A medical image segmentation method based on GPCS-TransUNet
By improving the GPCS-TransUNet model and utilizing GPGA, SLA, and LCSA modules to enhance feature interaction and expression, the shortcomings and redundancy of multi-scale modeling in U-shaped networks in medical image segmentation are solved, achieving higher-precision segmentation results.
CN122265302APending Publication Date: 2026-06-23GUILIN UNIVERSITY OF TECHNOLOGY
View PDF 0 Cites 0 Cited by
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- GUILIN UNIVERSITY OF TECHNOLOGY
- Filing Date
- 2026-03-23
- Publication Date
- 2026-06-23
AI Technical Summary
Technical Problem
Existing U-shaped networks in medical image segmentation suffer from insufficient multi-scale feature interaction modeling, inadequate key structure representation capabilities, and feature redundancy, which affect the model's performance when handling complex tasks.
Method used
A medical image segmentation network based on GPCS-TransUNet was designed. By introducing GPGA, SLA and LCSA modules, the feature interaction capability and feature representation quality are enhanced, and the model performance is optimized by skip connections and loss functions.
Benefits of technology
It significantly improves the accuracy and feature representation capabilities of medical image segmentation, especially its robustness and segmentation performance on complex image datasets.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN122265302A_ABST
Abstract
The application discloses a medical image segmentation method based on GPCS-TransUNet, and belongs to the medical image segmentation field.The method improves the TransUNet model, introduces a global perception gate aggregation attention in the encoder stage, enhances the interaction ability between different features, and effectively improves the shortcomings of the model when processing multi-scale feature interaction.In addition, a simplified linear attention module is introduced before the feature enters the Transformer module, which performs linear complexity context interaction and semantic reorganization on the feature, so that the feature has more stable semantic expression.Finally, in the decoder stage, a lightweight channel and spatial attention module is designed, which jointly models the channel and spatial attention, adaptively recalibrates the feature, strengthens the key semantic information and suppresses the redundant response, thereby improving the accuracy of the feature in semantic expression.In summary, the GPCS-TransUNet improves the accuracy and feature expression ability of medical image segmentation through multi-module collaborative optimization.
Need to check novelty before this filing date? Find Prior Art