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

Auxiliary detecting method of mammary gland catheter preinvasive carcinoma

A technology for auxiliary detection and mammary ducts, applied in mammography, diagnostic recording/measurement, medical science, etc., can solve problems such as low accuracy, accuracy less than 73.2%, and great difficulty, and achieve high accuracy results

Inactive Publication Date: 2017-08-11
成都知识视觉科技有限公司
View PDF0 Cites 13 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

According to literature records, at present, the accuracy of manual inspection of IDC by pathologists is less than 73.2%, and the accuracy of using CNN to identify IDC is about 84%.
However, since IDC and DCIS have a high similarity in cell morphology, it is difficult to distinguish only from the cell morphology using the CNN neural network, and the accuracy is not high

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

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0027] Embodiment 1: an auxiliary detection method for ductal carcinoma in situ of the breast, comprising the following steps:

[0028] A. Pathologists manually label the digital slices of breast cancer to obtain images of DCIS and myoepithelial regions;

[0029] B. Read in the digital slice file image, cut the image into small pieces, and obtain whether the small piece image contains DCIS or myoepithelial region images by querying the information in the file marked by the pathologist, thereby obtaining three types of sample sets, namely : DCIS, myoepithelial, other tissues;

[0030] C. Start the neural network, start training, and establish a recognition model;

[0031] D. Use the identification model to identify the digital slices, find out the DCIS and myoepithelial regions, and record the corresponding probability;

[0032] E. Check the connectivity of the detected DCIS area and the myoepithelial area respectively, merge the areas connected together, and mark them as the...

Embodiment 2

[0035] Embodiment 2: an auxiliary detection method for breast ductal carcinoma in situ, comprising the following steps:

[0036] A. Pathologists manually label the digital slices of breast cancer to obtain images of DCIS and myoepithelial regions;

[0037] B. Read in the digitized slice file. The image is scaled according to the level (n), converted to the HSV color space, and the Hue space is selected to use the maximum inter-class variance method for threshold calculation, and the foreground and background areas are extracted, and then the image is corroded and processed. The expansion operation removes the small-area interference in the image and fills the small holes in the large area. Finally, the image under the level (n) scaling ratio is mapped back to the image under the specified ratio; by querying the pathologist’s annotation file information to obtain whether the image of the small foreground area contains DCIS or myoepithelial region images, thus obtaining three ty...

Embodiment 3

[0043] Embodiment 3: an auxiliary detection method for breast ductal carcinoma in situ, comprising the following steps:

[0044] 1) Pathologists manually label the digital slices of breast cancer to obtain images of DCIS and myoepithelial regions;

[0045] 2) Read in the image under the level(n) scaling ratio of the WSI file (for example: n=5, that is, downsampling 5 times, which is 1 / 32 of the original image size), convert it to HSV color space, and select it in Hue In the space, the maximum inter-class variance method is used to calculate the threshold value, extract the foreground area and the background area, and then perform image erosion and expansion operations to remove the small area interference in the image and fill the small holes in the large area. Finally, the level (n) The image under the zoom ratio is mapped back to the image under the specified ratio, such as: the original image under level0;

[0046]3) Cut the image of the foreground area into small patches ...

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 auxiliary detecting method of mammary gland catheter preinvasive carcinoma. The method comprises the steps that a breast cancer digital section is manually annotated, and DCIS and myoepithelium area images are obtained; digital section file images are read in and cut into fragments, whether DCIS or the myoepithelium area images are included in fragment images or not is acquired by inquiring information in pathologist annotated files, and therefore three kinds of sample sets are obtained; a neural network is started, and a recognition model is set up; the digital section is recognized, DCIS and a myoepithelium area are found, and a corresponding possibility probability is recorded; the probability that each area is considered as DCIS is calculated. According to the method for recognizing DCIS through CNN, DCIS cancer cells and myoepithelium tissue are detected at the same time, and high-accuracy DCIS automatic recognition is achieved.

Description

technical field [0001] The invention relates to an auxiliary detection method for breast ductal carcinoma in situ. Background technique [0002] The female mammary gland is composed of skin, fibrous tissue, mammary glands and fat. Breast cancer is a malignant tumor that occurs in the mammary gland epithelial tissue. 99% of breast cancer occurs in women, and only 1% in men. Breast cancer is the number one common malignant tumor in women. [0003] The global incidence of breast cancer has been on the rise since the late 1970s. 1 in 8 women in the U.S. will develop breast cancer in her lifetime. China is not a country with a high incidence of breast cancer, but it should not be optimistic. In recent years, the growth rate of the incidence of breast cancer in my country is 1 to 2 percentage points higher than that of countries with a high incidence. [0004] The mammary gland is not an important organ to maintain human life activities, and breast cancer in situ is not fatal; ...

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): A61B5/00
CPCA61B5/4312A61B5/7235A61B5/7264
Inventor 蒲洋向飞
Owner 成都知识视觉科技有限公司
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