CT image-based interpretable pulmonary tuberculosis classification network identification method

A CT image, classification network technology, applied in neural learning methods, character and pattern recognition, biological neural network models, etc., can solve problems such as missed diagnosis, misdiagnosis, and difficulty in identification.

Active Publication Date: 2021-07-09
ZHEJIANG UNIV
View PDF4 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In CT images, pulmonary tuberculosis has the characteristics of polymorphism, multiple sites, multiple nodules, and cavities, and the mixture of various forms makes it difficult to identify, and medical

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
  • CT image-based interpretable pulmonary tuberculosis classification network identification method
  • CT image-based interpretable pulmonary tuberculosis classification network identification method
  • CT image-based interpretable pulmonary tuberculosis classification network identification method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0040] Embodiment 1, the interpretable pulmonary tuberculosis classification network recognition method based on CT images, such as Figure 1-8 shown, including the following steps:

[0041] Step 1. Get the image and preprocess it

[0042] Obtain a CT (Computed Tomography, CT for short) image of the lungs of the person to be tested (such as a patient), as the original CT image, and input it to the host computer for processing;

[0043] In order to reduce the influence of irrelevant factors in the detection of pulmonary tuberculosis, reduce the calculation amount of the feature extraction network in the detection of pulmonary tuberculosis, and reduce the training time, the original CT image is first preprocessed to extract the lung parenchyma, such as figure 2 , using the method of iterative threshold method;

[0044] Step 1.1, image preprocessing

[0045] 1) Binarize the original CT image, that is, set a global initial threshold T, which can be based on the maximum gray valu...

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 an interpretable pulmonary tuberculosis classification network identification method based on a CT image, and the method comprises the following steps: obtaining a CT image of the lung of a to-be-detected person, and carrying out the normalization processing and zooming of a pulmonary parenchyma image; establishing and training an interpretable pulmonary tuberculosis classification network, wherein the interpretable pulmonary tuberculosis classification network is established by taking a residual network as a basic network and adding a Dense thought and an improved attention mechanism; and sending the lung parenchyma image which is normalized and scaled to the size of 512 * 512 into the trained interpretable pulmonary tuberculosis classification network, and obtaining and outputting a classification result of the pulmonary tuberculosis of the to-be-detected person and a corresponding class activation thermodynamic diagram in an upper computer. According to the CT image of the to-be-tested person, the classification result of the pulmonary tuberculosis of the to-be-tested person can be obtained, and the corresponding class activation thermodynamic diagram can be output.

Description

technical field [0001] The invention relates to the fields of computer-aided diagnosis and treatment and image recognition, in particular to an interpretable pulmonary tuberculosis classification network recognition method based on CT images. Background technique [0002] Currently, tuberculosis is one of the major infectious diseases threatening health, and its typical manifestation is pulmonary tuberculosis. Tuberculosis is a chronic infectious disease with a long history of endangering human health. It was the most serious epidemic in the early 20th century and even caused millions of deaths around the world. At present, my country is one of the countries with a high burden of tuberculosis epidemics. Because tuberculosis is highly contagious and lethal, early auxiliary diagnosis of tuberculosis can help doctors find early tuberculosis patients, conduct early diagnosis and treatment, and reduce infection rate and mortality, so it has very important clinical significance. ...

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): G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06N3/045G06F18/214G06F18/241
Inventor 金心宇马云龙余晨洋金昀程
Owner ZHEJIANG 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

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap