Digital pathological image feature extraction system and method
By combining homogeneous and heterogeneous neural networks in pathological images, the problem of incomplete extraction of pathological images is solved, and the accuracy of tumor diagnosis is improved.
CN120374996APending Publication Date: 2025-07-25SHENZHEN INST OF ADVANCED TECH CHINESE ACAD OF SCI
Patent Information
- Application Number
- CN202510246656.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-28
- Publication Date
- 2025-07-25
Smart Images
Figure CN120374996A_ABST
Abstract
The invention relates to biomedical images, in particular to a digital pathological image feature extraction system and method, solves the problem that pathological image features cannot be fully extracted in the prior art, and performs feature fusion on tumor infiltration information features and tumor spatial heterogeneity information features extracted from pathological images. The fusion result is used as the pathological image features of the pathological image, and the obtained pathological image features can reflect the pathological image features more comprehensively and can be used for tumor subtype classification tasks, survival prediction tasks and the like.
Need to check novelty before this filing date? Find Prior Art
Citation Information
Cited By
Pathological image segmentation method and system based on graph neural network and cross-region fusion
CN120876516A
Pathological image segmentation method and system based on graph neural network and cross-region fusion
CN120876516B