颅内动脉瘤自动分割方法和系统
By using the parallel design of artery feature encoders and aneurysm feature encoders and a depth-separable dilated convolution module, the problem of insufficient segmentation accuracy of intracranial aneurysms in the prior art is solved, achieving more accurate aneurysm segmentation and supporting the diagnosis and treatment of aneurysm disease.
CN120471933BActive Publication Date: 2026-07-17HANGLOK-TECH CO LTD
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- HANGLOK-TECH CO LTD
- Filing Date
- 2025-04-03
- Publication Date
- 2026-07-17
AI Technical Summary
Technical Problem
Existing technologies struggle to effectively utilize the contextual information of intracranial arterial structures and aneurysms, resulting in limited segmentation accuracy for intracranial aneurysms.
Method used
An intracranial aneurysm segmentation network designed in parallel with artery feature encoders and aneurysm feature encoders achieves accurate aneurysm segmentation through multi-scale feature extraction via adaptive feature modulation and attention computation, combined with a depthwise separable dilated convolution module.
Benefits of technology
It improves the accuracy of intracranial aneurysm segmentation and can assist in the diagnosis and treatment of aneurysm diseases, such as rupture risk assessment, interventional path planning, and surgical navigation.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN120471933B_ABST
Abstract
本发明公开了一种颅内动脉瘤自动分割方法和系统，方法包括：构建脑血管分割预训练网络并获得预训练模型权重、构建颅内动脉瘤分割网络；动脉瘤分割网络由双分支动脉‑动脉瘤编码器和动脉瘤解码器组成，构建分组自适应查询注意力机制，利用编码器的动脉特征增强动脉瘤特征表达，并减小动脉结构对动脉瘤分割的干扰；在动脉瘤解码器的深度监督框架中构建深度可分离空洞卷积模块，通过三个并行的深度可分离空洞卷积分支捕获多尺度特征。对颅内动脉结构与动脉瘤的上下文关系进行建模，迁移预训练的血管特征指导动脉瘤分割，减小依赖预分割动脉结构可能带来的不稳定性，实现对多种尺寸动脉瘤的精准分割。
Need to check novelty before this filing date? Find Prior Art