Segmentation method of oct image layer based on random forest and compound activity curve

A random forest, image layer technology, applied in the field of medical image processing algorithms, can solve the problems of layer segmentation failure, blurred boundaries, large changes in retinal layer structure, etc.

Active Publication Date: 2020-08-11
SUZHOU UNIV
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This results in low contrast and blurred boundaries between retinal layers in OCT images, and also highly variable retinal layer structure
Therefore, layer segmentation may fail using traditional surface detection methods such as image search algorithms, and meanwhile, effusion segmentation using traditional methods such as region growing may also easily leak into the neighborhood

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
  • Segmentation method of oct image layer based on random forest and compound activity curve
  • Segmentation method of oct image layer based on random forest and compound activity curve
  • Segmentation method of oct image layer based on random forest and compound activity curve

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0114] This embodiment is based on the OCT image layer segmentation method of random forest and composite activity curve, including:

[0115] Such as figure 1 As shown, training OCT image features to train a random forest classifier, when training the random forest classifier, the retinal OCT image of central serous retinopathy is divided into 8 regions; region 1: nerve fiber layer; region 2: ganglion cell layer ;Region 3: inner plexiform layer; region 4: inner core layer; region 5: outer plexiform layer; region 6: outer nuclear layer + external membrane + sample area; Area + retinal pigment epithelium / Bruch; area 8 (class 8): vitreous body + choroid; the upper surface layer Surface1 of area 1, the upper surface Surface2 of area 2... the upper surface of area 8, Surface1 is referred to as SF1, and the same is true for Surface2 SF2 for short...Surface8 for SF8 for short, see for details figure 1 shown.

[0116] Specific methods for OCT image layer segmentation include:

[0...

Embodiment 2

[0125] This embodiment is based on the OCT image layer segmentation method of random forest and compound activity curve, on the basis of embodiment 1, obtains OCT image feature training random forest classifier specifically includes:

[0126] The retinal OCT images of central serous retinopathy were divided into 8 regions; region 1: nerve fiber layer; region 2: ganglion cell layer; region 3: inner plexiform layer; region 4: inner inner layer; region 5: outer plexiform layer layer; area 6: outer nuclear layer + external membrane + sample area; area 7: ellipsoidal area + outer photoreceptor segmental layer + staggered area + retinal pigment epithelium / Bruch; area 8 (class8): vitreous + choroid; area The upper surface layer Surface1 of 1, the upper surface Surface2 of area 2...the upper surface of area 8, Surface1 is referred to as SF1, and similarly, Surface2 is referred to as SF2...Surface8 is referred to as SF8, for details, see figure 1 The specific division shown is not limi...

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 layer segmentation method based on random forest and compound activity curve, which is designed for accurate segmentation of retinal layer and effusion. The OCT image layer segmentation method based on random forest and compound activity curve of the present invention, comprises: obtains OCT image feature training random forest classifier; Composite activity curve algorithm obtains final SF1; Extract 24 features of OCT image, use random forest classifier , the OCT image layer segmentation method based on random forest and compound activity curve in the present invention has simple operation and accurate detection result. It overcomes the problems of low recognition rate and poor segmentation effect of existing OCT image segmentation algorithms for lesions.

Description

technical field [0001] The invention belongs to the field of medical image processing algorithms, in particular to an OCT image layer segmentation method based on random forests and compound activity curves. Background technique [0002] Central serous retinopathy is a severe and complex retinal disease that can easily lead to blindness. Central serous retinopathy mainly manifests as the accumulation of subretinal serous fluid in the staggered area, which may also lead to retinal pigment epithelial detachment. In addition to serous accumulation, retinal pigment epithelial detachment may occur subserously or near the center of the macula. This fluid causes swelling of the retinal layers, which may suddenly change in thickness and optical strength. Thus, central serous retinopathy is a common disorder of the macula, which is responsible for central vision. Therefore, the quantitative analysis of central serous retinopathy has very important significance in the research of r...

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 Patents(China)
IPC IPC(8): G06T7/11G06K9/62
CPCG06T7/11G06T2207/30096G06T2207/30041G06F18/24323
Inventor 向德辉陈新建
Owner SUZHOU 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