Multi-modal fusion recognition system and device for mycobacterium pulmonary granuloma infection

By designing a multi -mode fusion recognition system, combining pathological images, CT images and clinical information, using self -attention mechanism and Transformer model to solve the problem of identifying the infection of lung granulation and swelling in the existing technology, realized the realization, and achieved it. Quickly and accurately identify and avoid the complexity and high cost in traditional detection methods.

CN119360167AActive Publication Date: 2025-01-24CHINA JAPAN FRIENDSHIP HOSPITAL
View PDF 3 Cites 2 Cited by

Patent Information

Application Number
CN202411393728.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-08
Publication Date
2025-01-24
Estimated Expiration
2044-10-08

Smart Images

  • Figure CN119360167A_ABST
    Figure CN119360167A_ABST
Patent Text Reader

Abstract

The invention provides a multi-modal fusion recognition system and device for mycobacterium pulmonary granuloma infection, and relates to the technical field of multi-modal fusion recognition, and the system mainly comprises a pathological image feature extraction module, a CT image feature extraction module, a clinical information feature extraction module and a feature fusion classification module. The pathological image feature extraction module comprises a preprocessing unit, a segmentation unit and a pathological feature unit; the feature fusion classification module comprises a self-attention mechanism module and is used for inputting the pathological features, the CT image features and the clinical information features into the self-attention mechanism module and outputting an identification result of mycobacterium pulmonary granuloma infection; according to the scheme, the pulmonary granuloma disease information of various modals such as pathology, imaging and clinic is comprehensively fused, and mycobacterium pulmonary granuloma infection can be accurately, simply, conveniently and economically identified.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Lung CT image classification system based on domain knowledge and parallel separable convolution Swin Transform

    CN117058448A

  • System for assessing risk of pulmonary nodules

    WO2022110278A1

  • Multi-modal brain tumor image segmentation method based on self-supervised learning

    WO2024108522A1

Cited By

  • Lung injury intelligent detection system based on lung image and pathological data

    CN120260893A

  • Intelligent lung injury detection system based on lung imaging and pathological data

    CN120260893B