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

Unsupervised self-adaptive mammary gland lesion segmentation method

An adaptive model and image segmentation technology, applied in the field of image processing, can solve the problem of poor generalization ability of the segmentation model

Active Publication Date: 2020-05-19
SHENZHEN INST OF ADVANCED TECH CHINESE ACAD OF SCI
View PDF3 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] In order to solve the shortcomings of the poor generalization ability of the segmentation model between domains mentioned in the above background technology, the present invention proposes an unsupervised domain adaptive image segmentation 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
  • Unsupervised self-adaptive mammary gland lesion segmentation method
  • Unsupervised self-adaptive mammary gland lesion segmentation method
  • Unsupervised self-adaptive mammary gland lesion segmentation method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0086] Such as figure 1 As shown, the present invention proposes a method for adaptive image conversion between different fields. Based on this method, even if there is a difference between the new data set and the labeled data set, there is no need to label the images in the new data set, but through image conversion Adaptive learning is performed on two datasets. As a result, the unlabeled dataset retains its high-level semantic information through adaptive transformation, and its shallow representations such as image style, texture, and brightness are converted into the features of the labeled dataset, so that the trained dataset in the labeled dataset can be directly The network model is directly applied to the new dataset.

[0087] In the adaptive image conversion method according to one embodiment of the present invention, comprise the following steps:

[0088] First, a dataset containing annotations is required as the source domain image, which can be regarded as an a...

Embodiment 2

[0105] Such as figure 2 As shown, another embodiment of the present invention is an image segmentation method for unsupervised field adaptation, which can be specifically applied to computer-aided recognition of MRI images of breast lesions. It is mainly divided into: establishing an image domain discrimination network (S401), performing image reconstruction learning of target domain images (S402), optimizing based on the reconstruction network to obtain a conversion network (S403, S404), performing image processing on labeled source domain images. Segmentation network training (S405), converting the image of the target domain through the conversion network (S406), and then performing image segmentation on the converted target image by using the image segmentation network (S407).

[0106] In the above steps, the steps in S401, S402, S403, and S404 are the same as the corresponding steps in the first embodiment, and will not be repeated here.

[0107] Segmenting an annotated ...

Embodiment 3

[0116] see Figure 4 , the present embodiment provides a breast cancer screening device based on adaptive image segmentation, including:

[0117] An acquisition unit, configured to acquire source domain images in a source image set, where the images in the source domain image set contain marked feature regions, the source domain image set is a marked breast MRI image, and the feature regions are Lumps or areas of cancerous tissue to be marked;

[0118] It is also used to acquire a target domain image in the target domain image set, the target domain image is an unmarked breast MRI image, and the target domain image may contain an image part corresponding to a tumor or a cancerous tissue area;

[0119] An image domain discrimination unit, configured to take the source domain image and the target domain image acquired by the acquisition unit as input, and establish an image domain discrimination network Ncl for discriminating the domain to which the image belongs through a trai...

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 provides a self-adaptive image segmentation method in the unsupervised field. The method includes: converting the target domain image to reserve semantic information of the target domainimage, and reconstructing shallow information of the image by taking the source domain image as a feature; and then performing image discrimination on the converted and reconstructed target domain image by utilizing a segmentation model established by the source domain image, thereby realizing migration between data sets of different domains of the model on the premise of no new data annotation.

Description

technical field [0001] The invention relates to the field of image processing, in particular to an unsupervised-based adaptive breast lesion segmentation method. Background technique [0002] Breast cancer is the cancer with the highest incidence rate in women. Early diagnosis and treatment can effectively improve the long-term survival rate of breast cancer patients. As a multi-parameter, multi-contrast imaging technique, Magnetic Resonance Imaging (MRI) can reflect tissue T1, T2 and proton density and other characteristics, and has the advantages of high resolution and sensitivity. One of the important tools for screening. Breast MRI technology has been increasingly used in clinical practice, especially in the early screening of breast cancer. [0003] In breast cancer MRI screening, computer-assisted image analysis is a development trend and a core technical issue in this field. Early medical image segmentation initially used edge detection, texture features, morpholog...

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): G06T7/11G06T3/00
CPCG06T7/11G06T2207/10088G06T2207/20081G06T2207/20084G06T2207/30068G06T2207/30096G06T3/04Y02T10/40
Inventor 李程王珊珊肖韬辉郑海荣刘新梁栋
Owner SHENZHEN INST OF ADVANCED TECH CHINESE ACAD OF SCI
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