Disease classification method fusing multi-modal features based on integrated learning and equipment

An integrated learning and disease classification technology, applied in character and pattern recognition, instruments, computer parts, etc., can solve the problems of insufficient use of medical imaging data mining, only considering single-modal image features, and insufficient classification accuracy. Good self-learning ability, improve classification accuracy and model generalization ability, and avoid the effect of prediction errors

Pending Publication Date: 2022-01-14
SHANGHAI UNIV OF MEDICINE & HEALTH SCI +1
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The existing mining and utilization of medical imaging data is not enough. Although the research on the computer-aided classification model of diseases has achieved certain results, there are still the following deficiencies: (1) The characteristic mode is single
Most of the current models are based on independent mode analysis, only considering single-mode image features, and the classification accuracy is insufficient
(2) Only use a single classifier model, but there is no one algorithm that is always better than other algorithms. Based on different modal features, there are different optimal classifiers, and a robust and robust Pa Kinson's disease classification model still to be examined

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Disease classification method fusing multi-modal features based on integrated learning and equipment
  • Disease classification method fusing multi-modal features based on integrated learning and equipment

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0037] This embodiment provides a disease classification method based on integrated learning and fusion of multi-modal features. This embodiment applies it to the classification of early Parkinson's disease (Parkinson's Disease, PD), including the following steps:

[0038] Step 1: Acquire multimodal magnetic resonance images and clinical text information of the same subject obtained on the same device, and perform data preprocessing.

[0039] Multimodal magnetic resonance images include T1-weighted magnetic resonance imaging (T1-WI), diffusion tensor imaging (DTI) and resting functional magnetic resonance imaging (resting functional magnetic resonance imaging, rs-fMRI) three modality image data. The clinical text information selects pre-clinical non-motor disorder score data. Corresponding image preprocessing is performed on the three modal magnetic resonance imaging data, and data normalization preprocessing is performed on the clinical text information.

[0040] (1) T1-WI ...

Embodiment 2

[0060] This embodiment also provides a disease classification device based on integrated learning and fusion of multi-modal features, including data screening and preprocessing modules, feature extraction modules, adaptive base classifier selection modules and meta-learner integrated learning modules, wherein data screening And the preprocessing module is used to obtain T1-weighted magnetic resonance images, diffusion tensor images, resting state functional magnetic resonance images and pre-clinical non-movement scores, and perform corresponding image preprocessing on the three modal magnetic resonance imaging data, Perform data normalization preprocessing on clinical text information; the feature extraction module is used to extract the whole brain morphological feature map of each modality image, and calculate the morphological feature values ​​in the brain area of ​​interest, and use the double-sample t-test statistics The clinical text information with inter-group differenc...

Embodiment 3

[0062] The present invention also provides an electronic device, including one or more processors, memory, and one or more programs stored in the memory, one or more programs include a method for performing the integrated learning based on the method described in Embodiment 1. Instructions for Disease Classification Methods Fused with Multimodal Features.

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention relates to a disease classification method fusing multi-modal features based on integrated learning and equipment. The classification method comprises the following steps: obtaining multi-modal magnetic resonance imaging and clinical text information of a same object on same equipment, and carrying out data preprocessing; extracting a whole-brain morphological feature map under each mode from the preprocessed multi-modal magnetic resonance imaging, correspondingly calculating image feature values in a brain region of interest, and extracting the clinical text information with inter-group differences to form text feature values; taking the image feature value of each mode and the text feature values as input of a corresponding optimal base classifier to obtain a plurality of rough classification results; and fusing the plurality of rough classification results to obtain a final classification result. Compared with the prior art, the invention has the advantages of high accuracy, high robustness and the like.

Description

technical field [0001] The invention relates to the technical field of computer processing based on medical images, in particular to a disease classification method and equipment based on integrated learning and fusion of multi-mode features. Background technique [0002] With the development of medical imaging technology, the data detection of various imaging methods, including magnetic resonance (Magnetic Resonance, MR) imaging, diffusion tensor imaging (Diffusion Tensor Imaging, DTI), etc., to reveal the situation of the sampler more completely , has become a development trend to improve the detection accuracy. [0003] The current mining and utilization of medical imaging data is not enough. Although the research on the computer-aided classification model of diseases has achieved certain results, there are still the following deficiencies: (1) The characteristic mode is single. Most of the current models are based on independent mode analysis, only considering single-mo...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06V10/764G06V10/80G06K9/62
CPCG06F18/2413G06F18/24155G06F18/2411G06F18/254G06F18/24
Inventor 黄钢聂生东杨一风
Owner SHANGHAI UNIV OF MEDICINE & HEALTH SCI
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap