医疗影像数据分析方法、装置、存储介质、电子设备及产品
By acquiring image feature description information and multi-level image data features, and using multi-level feature domain models and AI algorithms to match intelligent agents, the target AI algorithm is automatically determined, which solves the problems of heterogeneity and standardization of medical image data and improves the accuracy and reliability of analysis.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHANGHAI MEDICAL IMAGE INSIGHTS INTELLIGENT TECHNOLOGY CO LTD
- Filing Date
- 2026-03-16
- Publication Date
- 2026-07-17
AI Technical Summary
Medical image data is heterogeneous in origin, and its completeness and standardization are difficult to guarantee. Existing algorithm matching methods have high maintenance costs and poor scalability, making it difficult to execute AI tasks quickly and accurately in real medical environments.
By acquiring image feature description information and multi-level image data features, and using a pre-built multi-level feature domain model and AI algorithm to match intelligent agents, the target AI algorithm is automatically determined for medical AI analysis.
It enables the rapid and accurate identification of suitable AI algorithms, improving the accuracy and reliability of medical image data analysis and reducing the demand for computing resources.
Smart Images
Figure CN121885111B_ABST