Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

A five-classification method for leukocytes based on deep learning

A technology of deep learning and white blood cells, which is applied in the field of medical image processing, can solve the problems of describing the local features of uncelled images, not being able to check which type the wrong cells belong to, and not being able to use the next detection for damaged cells.

Active Publication Date: 2018-10-23
CHINA JILIANG UNIV +1
View PDF5 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Disadvantages: staining the cell nucleic acid destroys the structure of the cell, making the damaged cell unusable for the next detection, and it is impossible to check which type the wrong cell belongs to
[0006] Disadvantages of the above technology: the features taken are based on global considerations, and the local features of the cell image are not described

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 five-classification method for leukocytes based on deep learning
  • A five-classification method for leukocytes based on deep learning
  • A five-classification method for leukocytes based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0054] The present invention is further described below in conjunction with embodiment.

[0055] The present invention provides a method for five classifications of white blood cells based on deep learning. First, use simple color component relationships and morphological operations to find out the circumscribed rectangle of the area where the white blood cells are located as the positioning frame of the white blood cells to detect the white blood cells from the microscope picture; then use Particle features and SVM identify basophils, eosinophils and other cells; then use convolutional neural network to automatically extract the features of the remaining cell pictures, and use random forest to realize neutrophils, lymphocytes, monocytes Three categories.

[0056] like figure 1 As shown, the white blood cell classification method based on deep learning of the present invention includes the following steps:

[0057] (1) White blood cell detection

[0058] (1.1) The microscop...

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 belongs to the field of medical image processing, and relates to a five-classification technology for leukocytes in a human peripheral blood cell image, particularly to a leukocyte five-classification method based on deep learning. According to the leukocyte five-classification method, leukocytes are detected from a microscope image by using simple color-component relationships and morphological operations, basophils and eosinophils are identified by using particle characteristics and SVM, the remaining cell image characteristics are automatically extracted by using a convolutional neural network, and finally the remaining three-classification is achieved by using random forest. With the leukocyte five-classification method of the present invention, the errors caused by segmentation in the traditional method can be avoided, the leukocyte five-classification problem can be effectively solved, and the cells in different databases can achieve the good classification results.

Description

technical field [0001] The invention belongs to the field of medical image processing, and relates to a five-classification technology for white blood cells in human peripheral blood cell images, in particular to a five-classification method for white blood cells based on deep learning. Background technique [0002] Leukocytes in the blood are very important to the immune function of the human body. The number and percentage of various types of leukocytes in the blood are different under normal and diseased conditions. Doctors can judge the type of disease and the severity of the disease based on these important basic data This is of great value for the study of blood diseases in medical diagnosis, so it is very meaningful to study the differential count of white blood cells. With the continuous development of computer and artificial intelligence technology, cell image analysis has become an important auxiliary tool for clinical diagnosis, pathological analysis and treatment...

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): G01N15/10
CPCG01N15/10G01N2015/1006
Inventor 赵建伟张敏淑曹飞龙周正华冯爱明楚建军
Owner CHINA JILIANG 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
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