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

Pulmonary nodule texture feature extraction system and method based on body local direction three-valued mode

A local three-value pattern and texture feature technology, applied in character and pattern recognition, image data processing, computer parts and other directions, can solve the problem of lung nodule recognition rate, texture information loss and other problems, to improve efficiency, reduce Feature dimension, the effect of improving accuracy

Active Publication Date: 2019-07-05
HARBIN UNIV OF COMMERCE
View PDF10 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 solve the problem of the recognition rate of pulmonary nodules caused by the loss of partial texture information of the image in the existing pulmonary nodule texture feature extraction method, and propose a pulmonary nodule based on the three-valued mode of the local direction of the body. Nodal texture feature extraction system and method

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 texture feature extraction system and method based on body local direction three-valued mode
  • Pulmonary nodule texture feature extraction system and method based on body local direction three-valued mode
  • Pulmonary nodule texture feature extraction system and method based on body local direction three-valued mode

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0036] A pulmonary nodule texture feature extraction system based on the ternary model of volume local direction in this embodiment, the system includes:

[0037] The step of extracting local patterns from adjacent slices of pulmonary nodules based on the VLBP pattern; the step of calculating the local adaptive threshold using the normal distribution function for the extracted local patterns; the calculated local ternary pattern is calculated as the center pixel The step of performing ternary probability statistics in each direction as the center; use the ternary probability obtained in each direction as a feature vector and KNN classification algorithm to identify pulmonary nodules, and use the accuracy rate, confusion matrix and receiver operating characteristic curve Evaluate the recognition result of the texture feature of the pulmonary nodule, and determine whether the recognition result is correct or not.

specific Embodiment approach 2

[0039] The difference from Embodiment 1 is that in this embodiment, a pulmonary nodule texture feature extraction system based on the three-valued model of volume local direction, the local pattern extraction of adjacent slices of pulmonary nodules based on the VLBP mode The steps are:

[0040] Arrange the selected lung nodule slices in order, the order is slice 1, 2,..., n; select each pixel as the center on each plane to form a 3×3×3 neighborhood pixel Matrix, that is, the matrix formed by the pixel center and its surrounding 26 neighboring pixels.

specific Embodiment approach 3

[0042] The difference from the second embodiment is that in this embodiment, a pulmonary nodule texture feature extraction system based on the three-valued model of the local direction of the volume, the local adaptive threshold is performed on the extracted local model using the normal distribution function. The calculation steps are:

[0043] The local adaptive threshold of the local pattern of the extracted lung nodule adjacent slices was calculated using the normal distribution function to obtain the local ternary pattern:

[0044] The calculation formula of local ternary mode is as follows:

[0045]

[0046] In the formula, μ represents the center pixel obtained by calculating the mean value based on 26 neighborhood pixel values ​​and the center pixel value; g p Represents the neighborhood pixels; σ represents the fixed threshold obtained by calculating the mean square error based on the 26 neighborhood pixel values ​​and the central pixel value, and k is the threshol...

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 texture feature extraction system and method based on a body local direction three-valued mode. The invention relates to a method for identifying pulmonary nodules from pulmonary CT images. An existing pulmonary nodule grammar feature extraction method has the problem of pulmonary nodule recognition rate caused by loss of partial texture information of animage. The pulmonary nodule texture feature extraction system and method based on a three-valued mode in a body local direction comprises: local mode extraction is carried out on adjacent slices of apulmonary nodule based on a VLBP mode; calculating a local adaptive threshold value of the extracted local mode by using a normal distribution function; performing three-valued probability statisticson the local three-valued mode in each direction by taking a central pixel as a center; and taking the three-valued probability in each direction obtained by statistics as a feature vector and a KNNclassification algorithm to identify the pulmonary nodules, and evaluating the pulmonary nodule texture feature identification result to obtain whether the identification result is correct or not. According to the method, the pulmonary nodule recognition accuracy is improved.

Description

technical field [0001] The present invention relates to a system and method for identifying pulmonary nodules from CT images of the lungs, in particular to a system and method for extracting pulmonary nodule texture features based on a ternary model of volume local directions. Background technique [0002] Cancer has become a major threat to human health. The article "2018 Global Cancer Statistics" published in the CA journal on September 12, 2018 counted the incidence and mortality of 36 cancers in 185 countries around the world, including the prevalence of lung cancer, female breast cancer, and colorectal cancer Ranked among the top three, and the cancer death rate in Asia is much higher than other regions. In recent years, the incidence of lung cancer in the population is increasing year by year due to a combination of factors such as severe urban pollution, a substantial increase in the number of smokers, a low detection rate of early-stage lung cancer, and a low cure r...

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): G06T7/00G06K9/46
CPCG06T7/0012G06T2207/30064G06T2207/10081G06T2207/20081G06V10/462G06V10/44
Inventor 赵志杰任聪金雪松王冉韩小为张立志孙华东范智鹏陈永超陶武超
Owner HARBIN UNIV OF COMMERCE
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