A method for establishing a model of eosinophil proportion using pathological pictures of nasal polyps

A technology for eosinophils and nasal polyps, which is applied in the field of establishing eosinophil proportion model using pathological pictures of nasal polyps, can solve the problems of lack of accuracy and timeliness of auxiliary diagnosis system, measurement deviation, and high time cost

Active Publication Date: 2022-06-03
THE THIRD AFFILIATED HOSPITAL OF SUN YAT SEN UNIV +1
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In addition, due to different experiences of different doctors, or the same doctor at different times, the sampling estimates obtained by randomly selecting the field of view are also different, that is, artificial random sampling counts may also have measurement bias
[0004] Statistics of the proportion of eosinophils to inflammatory cells on the entire slide specimen can make a more accurate diagnosis and avoid sampling errors. However, pathologists need to spend 2 to 4 hours to fully count a slide specimen, and the time cost is extremely high
[0005] The clinical diagnosis of chronic sinusitis depends entirely on the experience of pathologists at this stage, and lacks an objective, highly accurate and timely auxiliary diagnosis system

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
  • A method for establishing a model of eosinophil proportion using pathological pictures of nasal polyps
  • A method for establishing a model of eosinophil proportion using pathological pictures of nasal polyps
  • A method for establishing a model of eosinophil proportion using pathological pictures of nasal polyps

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

file is saved.

Next, call the "multiresolutionimageinterface" library (https: / /

github.com / computationalpathologygroup / ASAP), generate a large mask map according to xml, the mask map is

The schematic diagram of the mask, the pixel value corresponding to the white area (outlined area) of the mask image is 1, and the pixels corresponding to other black areas are

A value of 0, the resolution of the entire mask image is the same as the WSI image, the digital pathology slide (whole slide image).

[0029] The WSI was then read with the open source "openslide" library (https: / / openslide.org / ), and the corresponding large

mask map, and cut the WSI and the corresponding large mask map according to the set resolution such as 512*512, respectively, and get

The small pathological map and the small mask map, and the positions of the small pathological map and the small mask map are in one-to-one correspondence.

Finally, according to the small picture resolution of setting, with sli...

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 method for establishing a proportion model of eosinophils by using pathological pictures of nasal polyps, comprising the following steps: data preparation, making nasal polyps into glass slides, and then obtaining WSI pictures by scanning with a digital pathological instrument, and cutting the WSI pictures , to obtain small pathological pictures; all small pathological pictures are divided into training set data and test set data according to the set ratio; the eosinophil proportion model is established, and the model parameters are obtained by using the Inception V3 model and its training on the ImageNet data set. Remove the last fully connected layer FC of this model, and add a fully connected layer FC with only one neuron in it, without using any activation function, set the loss function to use the mean square error MSE, and set the learning rate lr. The present invention is applied in the pathological auxiliary diagnosis system of chronic sinusitis nasal polyps, and the proportion of eosinophils on pathological pictures can be obtained quickly and accurately through learning and training.

Description

A method for establishing a model of eosinophil proportion using pathological pictures of nasal polyps technical field The present invention relates to the technical field of medical treatment means, be specifically related to a kind of utilizing nasal polyp pathological picture to establish Techniques for the method of the eosinophilic proportion model. Background technique Chronic rhinosinusitis (CRS) can be clinically divided into two types: without nasal polyps (CRSsNP) and with nasal polyps (CRSwNP). kind. Chronic rhinosinusitis with nasal polyps (CRSwNP) is subdivided into eosinophilic nasal polyps (eCRSwNP) and non-eosinophilic Two subtypes of granulocytic nasal polyps (neCRSwNP). Eosinophilic nasal polyps (eCRSwNP) sensitive to hormone therapy non-eosinophilic nasal polyps (neCRSwNP) are susceptible to macrolide antibiotic therapy. clinically, yes On how to define eosinophilic nasal polyps (eCRSwNP) and non-eosinophilic nasal polyps (neCRSwNP), disease ...

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): G16H30/40G16H50/20G16H50/50G06N3/08G06N3/04
CPCG16H50/50G16H50/20G16H30/40G06N3/08G06N3/045
Inventor 杨钦泰韩蓝青任勇吴庆武陈健宁邓慧仪孙悦奇袁联雄王玮豪郑瑞洪海裕孔维封黄雪琨袁田邱惠军李权黄桂芳叶俊杰王伦基
Owner THE THIRD AFFILIATED HOSPITAL OF SUN YAT SEN 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