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

OCT image-based Alzheimer's disease risk prediction method and system and medium

A prediction method and image technology, which is applied in image enhancement, image analysis, image data processing, etc., can solve problems such as difficult to meet the accuracy requirements of Alzheimer's disease risk prediction, low extraction accuracy, and complicated operation, and achieve Reduced overhead and receptive field, simple operation process, and high robustness

Active Publication Date: 2021-09-17
PING AN TECH (SHENZHEN) CO LTD
View PDF5 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The existing methods for extracting retinal quantitative indicators based on OCT images are mainly manual and semi-automatic extraction. However, both methods have the defects of complex operation and low extraction accuracy, and the results are difficult to meet the requirements for the risk of Alzheimer's disease. Forecasting Accuracy Requirements

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
  • OCT image-based Alzheimer's disease risk prediction method and system and medium
  • OCT image-based Alzheimer's disease risk prediction method and system and medium
  • OCT image-based Alzheimer's disease risk prediction method and system and medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0068] This embodiment implements a method for predicting the risk of Alzheimer's disease based on OCT images, such as figure 1 shown, including the following steps:

[0069] S101. Input the fundus OCT image to the trained retinal segmentation model to obtain the segmentation mask of the retinal region;

[0070] S102. Input the fundus OCT image to the trained choroidal segmentation model to obtain a segmentation mask of the choroidal region;

[0071] S103, detecting the position of the fovea of ​​the macula;

[0072] S104. Calculate the distance between the retinal segmentation mask at the fovea position and the upper and lower boundaries of the choroidal segmentation mask, and obtain the retinal thickness and choroidal thickness at the corresponding position;

[0073] S105. Input the omentum thickness, the choroid thickness, and the age and gender information of the person who took the fundus OCT image into the optimized multi-index Logistic regression model to obtain the A...

Embodiment 2

[0084] This embodiment implements a method for predicting the risk of Alzheimer's disease based on OCT images, including:

[0085] In the first step, the fundus OCT image is input to the trained retinal segmentation model to obtain the segmentation mask of the retinal region.

[0086] Specifically, the steps of training the retinal segmentation model are: obtaining OCT image training samples; configuring the encoding layer, bottleneck layer, decoding layer and classifier parameters of the retinal segmentation model; initializing the number of iterations of the retinal segmentation model; defining the A loss function of the retinal segmentation model; the OCT image training sample is input into the retinal segmentation model for training; when the training reaches a preset number of iterations, the training is terminated.

[0087] Obtain OCT image training samples; configure the encoding layer, bottleneck layer, decoding layer and classifier parameters; initialize the number of...

Embodiment 3

[0100] This embodiment implements a prediction system for the risk of Alzheimer's disease based on OCT images, such as figure 2 shown, including:

[0101] An obtaining module 301, configured to obtain a segmentation mask of the retinal region and a segmentation mask of the choroidal region;

[0102] A detection module 302, configured to detect the position of the fovea of ​​the macula;

[0103] Calculating the thickness module 303, for calculating the distance between the upper and lower boundaries of the macular fovea position retinal segmentation mask and the choroidal segmentation mask, to obtain the retinal thickness and choroidal thickness at the corresponding position;

[0104] The grade prediction module 304 is used to input the omentum thickness, the choroid thickness and the age and gender information of the person who took the fundus OCT image into the optimized multi-index Logistic regression model to obtain the risk grade of Alzheimer's disease.

[0105] Wherein...

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 an OCT image-based Alzheimer's disease risk prediction method and system and a medium. The OCT image-based Alzheimer's disease risk prediction method comprises the following steps that a fundus OCT image is input into a trained retina segmentation model to obtain a segmentation mask of a retina region; a fundus OCT image is input into a trained choroid segmentation model to obtain a segmentation mask of a choroid region; the position of a macular central recess is detected; the distance between the upper and lower boundaries of the retina segmentation mask and the choroid segmentation mask at the macular central recess is calculated to obtain the retina thickness and the choroid thickness at the corresponding positions; and the omentum thickness, the choroid thickness and the age and gender information of the person taking the fundus OCT image are input into an optimized multi-index Logistic regression model to obtain the disease risk level of the Alzheimer's disease. The retina and choroid segmentation and thickness measurement are carried out by adopting a Unet network structure, the accuracy is high, a multi-factor disease risk prediction model is constructed, and a more reliable result for predicting the Alzheimer's disease patient can be provided.

Description

technical field [0001] The present application relates to the technical field of Alzheimer's disease prediction, and more specifically, the present application relates to a method, system and medium for predicting the risk of Alzheimer's disease based on OCT images. Background technique [0002] Alzheimer's disease (Alzheimer's Disease, AD) is a progressive dementia symptom caused by the degeneration of the nervous system. The clinical manifestations of Alzheimer's disease are mainly memory impairment, cognitive dysfunction, mental symptoms, and personality and behavioral abnormalities. Its onset is slow and insidious, which can cause a series of neuropsychiatric symptoms and seriously affect the elderly. Patient health and quality of life. With the continuous aging of my country's population, the incidence of AD is increasing year by year. [0003] However, with the emergence of drugs to improve cognitive function, early treatment of AD has become very important. At pres...

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): A61B3/10A61B3/12A61B3/14A61B5/00G06K9/46G06K9/62G06N3/04G06N3/08G06T7/00G06T7/11G16H50/30
CPCA61B3/102A61B3/12A61B3/14A61B5/4088A61B5/7264A61B5/7267A61B5/7275G06T7/0012G06T7/11G06N3/08G16H50/30G06T2207/10101G06T2207/20076G06T2207/20081G06T2207/20084G06T2207/30041G06N3/047G06N3/048G06N3/045G06F18/241G06F18/2415Y02T10/40
Inventor 张成奋吕彬王关政吕传峰
Owner PING AN TECH (SHENZHEN) CO LTD
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