Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Brain age prediction model and method based on attention mechanism and bilinear fusion

A prediction model and attention technology, applied in the field of information, can solve the problems of poor prediction effect, cumbersome data processing, loss of feature information, etc., and achieve the effect of enriching feature expression ability and accurate brain age prediction.

Active Publication Date: 2020-09-08
NORTHWEST UNIV(CN)
View PDF6 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

But using this specific feature extraction method, there will be a loss of feature information, because these features may not be specifically designed to extract brain age-related information
These traditional methods require cumbersome early data processing and poor forecasting results

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
  • Brain age prediction model and method based on attention mechanism and bilinear fusion
  • Brain age prediction model and method based on attention mechanism and bilinear fusion
  • Brain age prediction model and method based on attention mechanism and bilinear fusion

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0048] Such as figure 1 As shown, this embodiment provides a brain age prediction model based on bilinear fusion and attention mechanism, including:

[0049] The preprocessing module is used to preprocess the original brain MRI data set to obtain the gray matter image X as the model input image;

[0050] 3D CNN feature extraction module, the feature extraction module includes 4 L1 layer to L4 layer with the same structure sub-layer; for image feature extraction of the model input image input in the 3D CNN feature extraction module, and the 3D CNN The feature matrix X output by the batch-normalization layer of the last L4 layer 4 as image features;

[0051] Bilinear fusion processing module, used for feature matrix X 4 for processing, the X 4 Perform transposition to obtain a new matrix B, the formula for defining B is: B=X 4 T ·X 4 , where B is the feature after bilinear fusion, X 4 T for X 4 The transpose matrix;

[0052] The attention value acquisition module is u...

Embodiment 2

[0058] Such as figure 1 As shown, the present invention discloses a method for constructing a brain age prediction model based on bilinear fusion and attention mechanism, a brain age prediction model based on bilinear fusion and attention mechanism (ie 3D CNN combined with bilinear fusion and Attention mechanism model) construction method comprises the following steps:

[0059] Step 1. Preprocess the original brain MRI data set to obtain the gray matter image X (121×145×121) as the model input image;

[0060] Specifically, the original brain MRI data set is first divided into a training set and a test set; and the original image MRI in the original brain MRI data set is generated as a gray matter image with a size of 121×145×121 as a model input image;

[0061] Step 2, first construct a 3D CNN feature extraction module, which contains 4 L1 to L4 layer blocks with the same structural sublayer;

[0062] Then input the model input image in step 1 to the 3D CNN feature extractio...

Embodiment 3

[0078] The invention also discloses a brain age prediction method based on bilinear fusion and attention mechanism, the method first preprocesses the original brain MRI data set, and obtains a gray matter image X (121×145×121) as a model input Image pairs to establish 3D CNN combined with bilinear fusion and attention mechanism model for training and testing to obtain predicted brain age f(x m ), using the mean square error MSE as the objective function:

[0079]

[0080] M represents the number of samples in the training set, y m Indicates the exact age of the tag, f(x m ) represents predicted brain age. The optimization strategy used is Adam to update parameters and minimize the objective function.

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 a brain age prediction model and method based on an attention mechanism and bilinear fusion. The method comprises the steps of extracting features of an image through a 3D CNNnetwork; then, in order to better enrich the feature expression capability of each pixel point, carrying out bilinear fusion processing on the extracted image features; meanwhile, due to the fact thatan attention mechanism can capture key feature information influencing the brain age and reduces the attention degree of the feature information irrelevant to the brain age, inputting the image features obtained after bilinear fusion processing into an Attention layer, and obtaining the attention weight of the images; obtaining the attention of the images based on the attention weight; and finally, sending the image features to a full connection layer, and performing brain age regression prediction. According to the invention, the accuracy of brain age prediction is improved by using a methodof combining the attention mechanism and bilinear fusion.

Description

technical field [0001] The invention belongs to the field of information technology, relates to image processing, data processing, and medical image processing, in particular to a brain age prediction model and method based on bilinear fusion and attention mechanism. Background technique [0002] Brain age is an important indicator of health because abnormal brain age size can lead to cognitive impairment and risk of neurodegenerative diseases. As a result, a growing number of researchers are attempting to study the interplay between brain aging and disease, and using neuroimaging techniques such as brain magnetic resonance imaging (MRI) to explore biomarkers in individual brains known as "brain age." , and confirmed that brain age has important value in neuroscience and clinical medicine. Brain age deviation is the deviation from the predicted normal aging of a healthy brain. When the predicted brain age is greater than the individual's chronological age, there is a risk ...

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): G16H50/30G06K9/62G06N3/04G06N3/08
CPCG16H50/30G06N3/08G06N3/045G06F18/25
Inventor 李展樊青晨王凯凯毋婷婷彭进业赵国英杨溪温超
Owner NORTHWEST UNIV(CN)
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com