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

Depth-learning-based eye-fundus image processing method, device and system

A fundus image and deep learning technology, which is applied in image data processing, image enhancement, image analysis, etc., can solve the problems of strong subjectivity of analysis results and high labor costs, and achieve unacceptability in solving time, reducing incompleteness, Analysis results are objective and accurate

Active Publication Date: 2017-02-15
北京新皓然软件技术有限责任公司
View PDF5 Cites 52 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The main problem to be solved by this application is to provide a method, device and system for processing fundus images based on deep learning. In the prior art, the analysis of fundus images is performed manually, and the analysis results are highly subjective and the labor cost is getting higher and higher. question

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
  • Depth-learning-based eye-fundus image processing method, device and system
  • Depth-learning-based eye-fundus image processing method, device and system
  • Depth-learning-based eye-fundus image processing method, device and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0036] like figure 1 Shown is the fundus image processing method based on deep learning in this embodiment, which is characterized in that it includes steps S100 to S400:

[0037] S100: Firstly, the fundus image is preprocessed, and the region of the fundus image is segmented and extracted;

[0038] S200: Then resampling the region image obtained by segmentation and extraction;

[0039] S300: Perform data expansion on the resampled regional image;

[0040] S400: Using a deep learning method to identify regional image features.

Embodiment 2

[0042] On the basis of Embodiment 1, wherein step S100 includes the following steps: First, in order to eliminate the differences between images caused by different illumination conditions and camera resolutions, calculate the color average value of the entire fundus image area, and subtract any pixel of the fundus image Go to the average color. Secondly, the multi-level parabola is used to describe the shape of the fundus blood vessels, and the centerline of the fundus blood vessels is identified. Then, the position of the optic disc region (the place where the blood vessels gather) is obtained by using the direction of the blood vessels, and the optic disc region is obtained by ellipse fitting. Taking the center of the optic disc as the origin, and expanding outward by 50 pixels from the origin to the farthest edge of the optic disc as the radius, the optic cup area and the periapapillary atrophic area (PPA) were extracted.

[0043] Preferably, step S200 includes the follow...

Embodiment 3

[0046] like figure 2 As shown, it is a flow chart of the training process of the convolutional neural network. First, the fundus image is preprocessed, and then resampling and data expansion are performed according to the method in Example 2. The trained convolutional neural network recognizes and recognizes the fundus image. analyze.

[0047] like figure 2 As shown, the convolutional neural network architecture includes 5 convolutional layers with weights and 2 fully connected layers, and the input layer is the image generated by resampling in the image preprocessing step. Connected to the input layer are 5 convolutional layers (Convolutional Layers). After the first and second convolutional layers are convoluted, the ReLUS (rectified linearunits) function is used for processing to speed up the training of the neural network; and then Local Response Normalization (Formula 1) is performed to prevent Overfitting, and finally Max Pooling (MaxPooling).

[0048] given Repr...

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 application discloses a depth-learning-based eye-fundus image processing method, device and system. The method comprises: eye-fundus image pretreatment is carried out and region segmentation extraction is carried out on the eye-fundus image; resampling is carried out on a regional image obtained by the segmentation extraction; data expansion is carried out on the regional image after resampling; and a regional image feature is identified by using a depth learning method. Therefore, an eye-fundus image can be analyzed automatically; the analysis result is objective and accurate; and the manpower cost is saved.

Description

technical field [0001] The present invention relates to the field of fundus image processing, in particular to a method and system for processing fundus images based on deep learning. Background technique [0002] The analysis of the fundus image is mainly performed manually in the prior art. Since the manual analysis is highly subjective, the accuracy and consistency of the manual analysis cannot be well guaranteed. At the same time, labor costs are getting higher and higher, so automatic analysis technology has great advantages in being more objective and saving labor costs. [0003] The concept of deep learning originated from the study of artificial neural networks. A multi-layer perceptron with multiple hidden layers is a deep learning structure. Deep learning combines low-level features to form more abstract high-level representation attribute categories or features to discover distributed feature representations of data. The concept of deep learning was proposed by...

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): G06T7/00G06T7/11
CPCG06T7/0012G06T2207/20081G06T2207/20084G06T2207/30041
Inventor 孟鑫
Owner 北京新皓然软件技术有限责任公司
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