Unlock instant, AI-driven research and patent intelligence for your innovation.

Pulmonary nodule screening method based on neural network

A technology of neural network and screening method, which is applied in the field of neural network and thoracic and pulmonary nodule screening, can solve the problems of model difficulty, poor effectiveness, and category imbalance training problems, and achieve the effect of weakening the influence

Active Publication Date: 2022-07-05
SICHUAN UNIV
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] The purpose of the present invention is to provide a method for screening pulmonary nodules based on a neural network, which solves the existing technology due to poor effectiveness when extracting features from X-rays, and fails to effectively solve the training problem caused by the imbalance of categories, and can Solve the problem that the constructed model is still difficult to be used in clinical practice due to the above reasons

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
  • Pulmonary nodule screening method based on neural network
  • Pulmonary nodule screening method based on neural network
  • Pulmonary nodule screening method based on neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0049] The embodiment of the present invention proposes a method for screening pulmonary nodules based on neural network, the flowchart of which is shown in figure 1 , wherein the method includes the following steps: firstly, the preparation of chest image data; secondly, the construction and training of the main feature extraction network; then, the construction of the cross-weighted cross-entropy function; then, the approximation of the AUC performance index; then, the online training and handling of special cases; finally, localization of thoracic lung nodules based on weakly supervised learning.

[0050] In this example, the ImageNet dataset and the trained InceptionV3 model are used as the main feature extraction network to train an end-to-end X-ray diagnostic model, and a cross-weighted cross-entropy function is constructed to solve the problem of class imbalance, so as to guide the training to be carried out effectively and prevent training. The majority of samples are ...

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 pulmonary nodule screening method based on neural network, which belongs to the field of neural network and thoracic pulmonary nodule screening. The present invention can solve the problem of poor effectiveness in extracting features in X-rays in the prior art, and can not effectively solve the training problem caused by category imbalance, and can solve the problem that the constructed model is still difficult to use due to the above reasons. on clinical issues. To this end, the present invention includes: preparation of chest image data; construction and training of main feature extraction network; construction of cross-weighted cross-entropy function approximation of AUC performance indicators; online training and processing in special cases; learning based on weak supervision Localization of thoracic pulmonary nodules.

Description

technical field [0001] The invention relates to the field of neural network and thoracic pulmonary nodule screening, in particular to a pulmonary nodule screening method based on neural network. Background technique [0002] Chest X-ray (chest x-ray imaging) is a rapid and less invasive technique that produces images of the interior of the chest and is widely used to screen for various chest pulmonary nodules, including cardiac hypertrophy, pneumonia, lung cancer Wait. In clinical practice, even very experienced experts still need to carefully analyze the imaging in the film, but due to the huge number of patients, the scarcity of professional doctors and the uneven geographical distribution, this has brought trouble to the majority of patients. The automatic diagnosis of thoracic and pulmonary nodules from X-ray images by the method of deep neural network is a very meaningful auxiliary medical method. It is of great significance to alleviate the medical imbalance and impro...

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): G06V10/46G06V10/774G06V10/82G06K9/62G06N3/04G06N3/08G16H30/20
Inventor 章毅王成弟郭际香李为民徐修远邵俊张海仙李经纬周尧宋璐佳
Owner SICHUAN UNIV