一种用于烧伤伤情评估的多任务图像分割方法及系统
By employing a multi-task learning convolutional neural network segmentation model, combined with a shared feature encoder and attention mechanism, the accuracy problem of burn depth and area assessment was solved, achieving high-precision, real-time segmentation of burn images and improving the automation and accuracy of burn injury assessment.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- XI AN JIAOTONG UNIV
- Filing Date
- 2024-04-08
- Publication Date
- 2026-07-17
AI Technical Summary
Existing intelligent diagnostic systems for assessing burn depth and area suffer from low accuracy, cumbersome operation, difficulty in handling irregularly shaped and cross-site burn areas, and lack of detailed burn location and distribution information.
A convolutional neural network segmentation model based on multi-task learning is adopted, which combines a shared feature encoder and human body segmentation branches and burn wound segmentation branches. High-precision segmentation of burn images is achieved through attention mechanism and self-attention module, realizing joint assessment of burn depth and area.
It achieves fully automatic, high-precision, and real-time segmentation of burn images, providing a more comprehensive and individualized assessment of burn injuries, and improving patients' treatment outcomes and recovery expectations.
Smart Images
Figure CN118229708B_ABST