Deep learning-based pulmonary tuberculosis recognition and diagnosis model, method and device and medium

A diagnostic model and deep learning technology, applied in the field of deep learning, can solve the problems of missing key details and wrong diagnosis, and achieve the effect of improving efficiency, application scope, and accuracy.

Pending Publication Date: 2022-02-01
山东健康医疗大数据有限公司
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Interpreting a patient's image data usually takes several minutes or even tens of minutes for one or more physicians, and it is still possible to miss some key details in the image, leading to wrong diagnosis

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
  • Deep learning-based pulmonary tuberculosis recognition and diagnosis model, method and device and medium
  • Deep learning-based pulmonary tuberculosis recognition and diagnosis model, method and device and medium
  • Deep learning-based pulmonary tuberculosis recognition and diagnosis model, method and device and medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0059] The tuberculosis identification and diagnosis model based on deep learning of the present invention includes three strategy models, namely Concatenation strategy model, Voting strategy model and Attention strategy model.

[0060] The Concatenation strategy model is used to extract features from the input image to obtain a feature map, and perform global average pooling on the feature map to obtain a feature vector, which is used to perform dimension reduction on the obtained feature vector to unify the dimensions of the feature vector and unify the dimensions. The spliced ​​feature vector is spliced ​​to obtain the spliced ​​feature vector, and the spliced ​​feature vector is used to identify and classify through the fully connected layer to obtain the diagnosis result.

[0061] The Voting strategy model is used to extract features from the input image to obtain a feature map, and perform global average pooling on the feature map to obtain a feature vector. For each dime...

Embodiment 2

[0072] The method for identifying and diagnosing pulmonary tuberculosis based on deep learning of the present invention comprises the following steps:

[0073] S100. Construct a pulmonary tuberculosis diagnostic model, the pulmonary tuberculosis diagnostic model is the deep learning-based pulmonary tuberculosis identification and diagnosis model disclosed in Example 1;

[0074] S200. Acquire tuberculosis X-ray images, each tuberculosis X-ray image corresponds to a diagnostic label;

[0075]S300. Construct a data set based on tuberculosis X-ray images and labels thereof, and divide the data set into a training set and a test set;

[0076] S400. Using the tuberculosis X-ray image in the training set as input and the diagnostic label as output, train the three strategy models in the tuberculosis identification and diagnosis model to obtain a trained tuberculosis identification and diagnosis model;

[0077] S500. Using the tuberculosis X-ray images in the test set as input, ident...

Embodiment 3

[0083] The device of the present invention includes: at least one memory and at least one processor; at least one memory for storing a machine-readable program; at least one processor for invoking the machine-readable program to execute the method disclosed in Embodiment 2.

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 deep-learning-based pulmonary tuberculosis recognition and diagnosis model, method and device and a medium, belongs to the technical field of deep learning, and aims to solve the technical problem of how to efficiently and accurately recognize and diagnose a pulmonary tuberculosis X-ray image. Three strategy models are included. The method comprises taking the pulmonary tuberculosis X-ray image in the training set as input, taking the diagnosis label as output, training the three strategy models, and obtaining a trained pulmonary tuberculosis identification and diagnosis model; taking the pulmonary tuberculosis X-ray image in the test set as input, performing identification and classification through the trained pulmonary tuberculosis identification and diagnosis model, and outputting a corresponding diagnosis result by each strategy model; for each strategy model, comparing an output diagnosis result with a related diagnosis label, and calculating the prediction accuracy of the strategy model; and selecting the strategy model with high accuracy as a target model, and identifying and classifying the to-be-detected pulmonary tuberculosis X-ray image through the target model.

Description

technical field [0001] The present invention relates to the technical field of deep learning, in particular to a model, method, device and medium for identifying and diagnosing pulmonary tuberculosis based on deep learning. Background technique [0002] With the rapid development of artificial intelligence technology, thanks to the explosive growth of computer computing power and data, the application of artificial intelligence technology combined with specific industries is also being rapidly implemented. And with the advancement of hospital informatization in the past ten years, the hospital has accumulated a large amount of patient data and precious image data, including tens of thousands of X-ray images of the lungs of patients. These lung X-ray images contain key pathological features of tuberculosis and have high application value. We need to explore and discover. Artificial intelligence technology can help us model the pathology of tuberculosis and perform rapid auxil...

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 Applications(China)
IPC IPC(8): G06K9/62G06N3/04G06N3/08G06V10/774G06V10/82
CPCG06N3/08G06N3/045G06F18/214
Inventor 杨高超孙承旭后永胜
Owner 山东健康医疗大数据有限公司
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