一种多模态影像语义融合的病灶自动标注方法和系统
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- NANJING JINGSAN MEDICAL TECH CO LTD
- Filing Date
- 2025-07-01
- Publication Date
- 2026-07-17
AI Technical Summary
Existing medical imaging lesion annotation technologies rely on manual operation, which is inefficient, highly dependent on expertise, and inconsistent with the annotation results of different doctors. This makes it difficult to meet the needs of large-scale analysis, has poor generalization ability, and is difficult to adapt to multimodal signals and small lesions.
By employing a multimodal image semantic fusion method, pseudo-color images and fuzzy clustering techniques are used to locate abnormal signals. Combined with an NLP model, lesion annotation is performed, and an end-to-end analysis framework is constructed to achieve automatic lesion annotation and efficient diagnosis.
It achieves high-precision and standardized automatic lesion annotation, improves the accuracy and efficiency of diagnosis, reduces training resource consumption, is highly adaptable, and can quickly generate a large number of high-quality annotation results.
Smart Images
Figure CN120544202B_ABST