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

White blood cell multi-classification identification method based on deep residual network

A technology of white blood cell and identification method, which is applied in the field of multi-classification and identification of white blood cells based on deep residual network, can solve the problems of not very good effect and insufficient detailed report of detection data, and achieves the effect of fast running speed and high accuracy.

Inactive Publication Date: 2018-03-09
HANGZHOU DIANZI UNIV
View PDF0 Cites 24 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

At present, there are mainly three-classification technology and five-classification technology based on white blood cell analyzers on the market, but it can only detect three main categories or five main categories, which makes the test data report not detailed enough and the effect is not very good

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
  • White blood cell multi-classification identification method based on deep residual network
  • White blood cell multi-classification identification method based on deep residual network
  • White blood cell multi-classification identification method based on deep residual network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0020] The present invention will be further described below in conjunction with the accompanying drawings.

[0021] Such as Figure 1-3 As shown, the white blood cell multi-classification recognition method based on the deep residual network, the specific implementation steps are as follows:

[0022] Step 1. Use the training set data to train the deep residual network architecture, and use the test set to verify the accuracy of the model generated by the deep residual network architecture, discard the model with poor performance on the test set, and save the performance on the test set better model. Such as process figure 1 In the "Train WBCnet" step unit;

[0023] Step 2. Image data acquisition: collect batches of multi-category white blood cell microscopic images (RGB white blood cell images), such as figure 1 In the "Image Acquisition" step unit;

[0024] Step 3. Data cleaning: mainly use methods such as cropping, flipping, PCA, and adding random noise to perform da...

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 white blood cell multi-classification identification method based on deep residual network. The invention provides a network configuration based on deep learning convolutionnerve network technology, and the residual design concept is merged in the network configuration. Strange white blood cell images can be identified and classified by using quite good fitting model generated by the network configuration. A multi-classification network configuration capable of conducting end-to-end learning can be designed by means of convolution nerve network in deep learning having residual configuration, and a classification model having quite good fitting can be trained based on the configuration. As the running speed of the configuration is fast, the fitting model of the training data can be rapidly obtained. The multi-classification model can conduct accurate identification and classification on strange white blood cell microscopic images. The white blood cell multi-classification identification method has quite a high accuracy in the field of leukemia cell recognition and classification.

Description

technical field [0001] The invention belongs to the field of leukemia cell classification, and in particular relates to a multi-classification recognition method for white blood cells based on a deep residual network. Background technique [0002] Leukemia has been the disease with the highest morbidity and fatality rate of malignant tumors in the world for many years in a row. At present, it mainly relies on blood routine examination and statistics of the proportion of various types of white blood cells to determine leukemia, and this operation is heavily dependent on the clinical experience of doctors. and level of awareness. Doctors have different levels of cognition of various types of leukemia, and their ability to judge the type of white blood cells is also different, so sometimes there are serious errors in judging leukemia and its type. There are many different types of leukemia, but the treatment required for each type is also different from disease to disease. At...

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
IPC IPC(8): G06K9/62G06N3/04
CPCG06N3/045G06F18/243
Inventor 姜明程柳张旻汤景凡杨智聪杜炼
Owner HANGZHOU DIANZI UNIV
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