Alzheimer's disease classification method based on multi-modal hypergraph convolutional neural network

A convolutional neural network and Alzheimer's disease technology, applied in the field of intelligent medical computer-aided diagnosis, can solve problems such as complex structure and complex pair relationship, and achieve the effect of shortening time, high classification effect, and small data dimension

Pending Publication Date: 2021-05-28
WUXI TAIHU UNIV +1
View PDF0 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In the real world, the relationship between subjects is much more complex than the paired relationship, and the object of CNN research is a regular spatial structure. Medical images such as brain images have a more complex structure.

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
  • Alzheimer's disease classification method based on multi-modal hypergraph convolutional neural network
  • Alzheimer's disease classification method based on multi-modal hypergraph convolutional neural network
  • Alzheimer's disease classification method based on multi-modal hypergraph convolutional neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0077] In accordance with the implementation of the training phase, the examples are done for the data in the standard dataset ADNI database. The data of subjects with three modalities of MRI, PET and CSF at the same time were selected for the experiment, and only the data collected at the fiducial points of these subjects were selected. In the ADNI database, there are 202 subjects with the above three modalities at the same time, including 51 with Alzheimer's disease (AD), 52 in the normal control group (Normal Control, NC), mild Cognitive Impairment (MCI) has 99. MCI was further divided into 56 patients who did not convert to Alzheimer's disease after 18 months (ie MCI non-converters, MCI-nc), and 56 who converted to Alzheimer's disease after 18 months (ie MCI converters). , MCI-c), was 43. The demographic information of these subjects is shown in Table 1:

[0078] Table 1: ADNI Demographic Information of 202 Subjects with Three Modalities: MRI, PET, and CSF

[0079] ...

Embodiment 2

[0087] In order to verify that the classification effect of multi-modal features is better than that of single-modal features, the present invention also uses different single modalities for classification, compares with the comprehensive multi-modal classification, and finally calculates the classification effects of various modalities and draws them as Table 4, Table 5, Table 6. The feature extraction method and process when using unimodal classification is the same as that of multimodality, and the final classifier uses unimodal TSK fuzzy classifier. It can be seen from the table that no matter what kind of classification task, the multimodal method achieves better results than the single modality. In the classification task of MCI vs. NC, the classification result of combining multiple modalities is 10% higher than the highest result in single-modality classification, reaching 86.83%, and the classification effect of multiple modalities in several other classification task...

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 belongs to the technical field of intelligent medical computer-aided diagnosis application, and relates to an Alzheimer's disease classification method based on a multi-modal hypergraph convolutional neural network. The method comprises a training stage and a using stage, wherein the training stage comprises an initial multi-modal feature construction model, a deep multi-modal feature extraction model and final Alzheimer disease classification. The initial multi-modal construction model uses a K-nearest neighbor and hypergraph theory to construct a hypergraph for each modal, and the initial multi-modal construction model is obtained; in order to improve the effectiveness of the multi-modal features, based on the initial multi-modal hypergraph data, deep learning is carried out by using the hypergraph-based graph convolutional neural network to construct the deep multi-modal features, and compared with the original multi-modal features, the multi-modal features subjected to deep feature extraction have smaller data dimensions and a higher classification effect.

Description

technical field [0001] The invention belongs to the technical field of intelligent medical computer aided diagnosis application, and relates to an Alzheimer's disease classification method based on a multimodal hypergraph convolutional neural network. [0002] technical background [0003] Alzheimer's Disease (AD) is the main cause of Alzheimer's disease. According to the survey, about 40 million people worldwide suffer from Alzheimer's disease, and this number is expected to double every 20 years and is expected to reach 2050. In 2019, there will be 115 million people with Alzheimer's disease, or about 1 in 85 older people will have Alzheimer's disease on average. The proportion of elderly people in developing countries is lower than that in developed countries in Europe and the United States, but the proportion of Alzheimer's disease is higher than that in Western Europe and the United States. [0004] Alzheimer's disease is a progressive and slow degeneration of brain fun...

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): G16H50/20G06K9/62G06K9/32G06K9/46G06N3/04G06N3/08G06F16/901
CPCG16H50/20G06F16/9024G06N3/04G06N3/08G06V10/25G06V10/40G06F18/254G06F18/24
Inventor 邓赵红曹营利姚晓峰王士同
Owner WUXI TAIHU UNIV
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