Alzheimer's disease classification and prediction system based on multi-task learning

A multi-task learning and prediction system technology, applied in the field of Alzheimer's disease classification and prediction system, can solve the problem of not being able to judge whether individuals with mild cognitive impairment will transform into Alzheimer's disease, and achieve improved classification efficiency and accuracy, avoiding processing, and reducing the need for prior knowledge

Active Publication Date: 2020-08-04
HARBIN INST OF TECH
View PDF5 Cites 14 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0009] The purpose of the present invention is to solve the problem that the existing classification system of Alzheimer's disease can only distinguish Alzheimer's disease patients, healthy individuals and individuals with mild cognitive impairm...

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 and prediction system based on multi-task learning
  • Alzheimer's disease classification and prediction system based on multi-task learning
  • Alzheimer's disease classification and prediction system based on multi-task learning

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0033] Specific implementation mode 1: In this implementation mode, a multi-task learning-based Alzheimer's disease classification and prediction system includes:

[0034] A multi-task learning-based Alzheimer's disease classification and prediction system according to the present invention is realized through the following technical solutions, including:

[0035] Structural magnetic resonance imaging (sMRI) preprocessing to reduce the effects of bias, noise and intensity in raw T1-weighted images and to unify image specifications;

[0036] Clinical indicator selection and feature extraction, that is, selecting clinical indicators for preprocessing and extracting feature vectors;

[0037] Disease classification and prediction model construction, that is, building each functional neural network module, adjusting the parameters of each module and building an overall neural network model, using the training sample set to train the disease classification and prediction model; usin...

specific Embodiment approach 2

[0047] Specific embodiment 2: The difference between this embodiment and specific embodiment 1 is that the image processing main module is used to collect head images, preprocess the collected head images, obtain preprocessed images, and The processed image is input into the training main module and the detection main module; the specific process is:

[0048] Obtain high-quality T1-weighted structural magnetic resonance imaging of the subject's head at the baseline time point (the time of the first visit, called the baseline time point) (Baseline) (high quality means that the patient's head position in the picture has no obvious shift , and the picture has no obvious artifacts; T1 weighting is a form of MRI image imaging);

[0049] Using the SPM tool (Statistical Parametric Mapping Toolbox) to perform skull stripping on structural magnetic resonance imaging;

[0050] Use non-rigid registration algorithms, including affine transformation and nonlinear transformation, etc., to ...

specific Embodiment approach 3

[0055] Embodiment 3: The difference between this embodiment and Embodiment 1 or 2 is that the clinical index processing main module is used to select the clinical index, obtain the feature vector of the clinical index, and input the feature vector of the clinical index into the training main module. module and detection main module; and the specific process is:

[0056] 1), selected clinical indicators;

[0057] 2), preprocessing the selected clinical indicators;

[0058] 3) Constructing a long-short-term memory neural network autoencoder (LSTM autoencoder) to extract the characteristics of longitudinal data of neuropsychological tests;

[0059] The clinical data includes baseline data and longitudinal data, and the features of the normalized baseline data and neuropsychological test longitudinal data are used as the input of Alzheimer's disease classification and prediction models.

[0060] Clinical indicators mainly include demographic information, genetic information and ...

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 discloses an Alzheimer's disease classification and prediction system based on multi-task learning, and relates to an Alzheimer's disease classification and prediction system. The objective of the invention is to solve the problem that an existing Alzheimer's disease classification system cannot judge whether a mild cognitive impairment individual will be transformed into Alzheimer'sdisease or not. The system comprises a an image processing main module, a clinical index processing main module, a neural network main module, a training main module and a detection main module; wherein the image processing main module is used for acquiring a head image, preprocessing the acquired head image to obtain a preprocessed image, and inputting the preprocessed image into the training main module and the detection main module; wherein the clinical index processing main module is used for selecting clinical indexes, acquiring feature vectors of the clinical indexes and inputting the feature vectors of the clinical indexes into the training main module and the detection main module; and the neural network main module is used for building an Alzheimer's disease classification and prediction model. The system is applied to the technical field of intelligent medical detection.

Description

technical field [0001] The invention belongs to the technical field of intelligent medical detection, and in particular relates to an Alzheimer's disease classification and prediction system. Background technique [0002] Alzheimer's disease (AD) is a chronic neurodegenerative disease and another major health threat after cardiovascular and cerebrovascular diseases and cancer. Alzheimer's disease is triggered by damage to nerve cells in areas of the brain associated with memory, and its symptoms are mainly manifested in memory impairment. Mild cognitive impairment (Mild Cognitive Impairment, MCI) is a transitional state between normal state and Alzheimer's disease, its prevalence is four times that of Alzheimer's disease, but the memory deficit it causes is relatively Symptoms are mild. Approximately 10% to 16% of people with mild cognitive impairment develop Alzheimer's disease each year. Therefore, early diagnosis and intervention of mild cognitive impairment play an im...

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
IPC IPC(8): G06K9/62G16H50/20G06N3/04G06N3/08
CPCG16H50/20G06N3/08G06N3/045G06F18/241G06F18/214
Inventor 李明磊罗浩李翔李款蒋宇辰尹珅
Owner HARBIN INST OF TECH
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