一种基于超图驱动的超声心动图视频心脏解剖结构标志点探测系统及方法
The hypergraph-driven echocardiogram video cardiac anatomical landmark detection system solves the reliability and interpretability problems of echocardiogram video cardiac function quantification in existing technologies, realizes accurate quantification of cardiac function and multi-task adaptability, and is suitable for automated diagnosis of complex cardiac diseases.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- NORTHEAST FORESTRY UNIV
- Filing Date
- 2025-06-03
- Publication Date
- 2026-07-17
AI Technical Summary
Existing deep learning-based methods for quantifying cardiac function using echocardiography neglect computational reliability and clinical interpretability, especially in cases of high-dimensional features where they cannot effectively quantify cardiac function.
A hypergraph-driven echocardiographic video cardiac anatomical landmark detection system is adopted, which includes hypergraph node feature vector mapping, adaptive hypergraph structure construction, hypergraph dynamic system, landmark feature quantity mapping and bidirectional hypergraph spatiotemporal decoding module. The system captures the long-range correlation and temporal dependence between landmarks through hypergraph network to achieve cardiac function quantification.
It improves the accuracy and reliability of cardiac function quantification, enhances the model's multi-task adaptability and clinical interpretability, and is suitable for automated diagnosis of complex cardiac diseases.
Smart Images
Figure CN120635016B_ABST