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

Medical image segmentation network based on dual interleaving

A network and dual technology, applied in the field of computer vision, can solve the problems of segmentation and loss of information, and achieve the effects of assisting medical diagnosis, improving accuracy, and improving segmentation performance

Active Publication Date: 2020-04-10
UNIV OF ELECTRONIC SCI & TECH OF CHINA
View PDF10 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The technical problem to be solved by the present invention is: to provide an idea of ​​using criss-cross to expand the network depth at the same time, and to solve the problem of information loss in traditional network segmentation

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
  • Medical image segmentation network based on dual interleaving
  • Medical image segmentation network based on dual interleaving
  • Medical image segmentation network based on dual interleaving

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0016] The specific implementation of the present invention is divided into two parts: the training of the algorithm model and the use of the algorithm model. The specific implementation manners of the present invention will be described in further detail below according to the drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0017] The medical image network model architecture based on criss-cross is as follows: image 3 shown. Each training sample contains an original brain MRI picture and a corresponding labeled label, and the present invention uses a z-score normalization operation to balance the data set. This network structure is divided into two basic blocks with the same internal details. The up and down sampling adopts an asymmetrical style, and the down sampling uses densely connected blocks to enhance the ability to extract features during the down samplin...

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 network based on dual interleaving. The network comprises the following steps: processing a picture data set; by using a dual interlaced network structure, enabling a first basic block of a DCNN to obtain the characteristics of a medical image and fuse the characteristics with the characteristics of an original image to serve as the image input of a second basic block of the DCNN; in the upsampling and downsampling process, enabling each basic block to adopt a non-symmetric structure, employing a dense connection block in the downsampling process, so as to reduce the feature loss of down-sampling; employing a criss-criss mode in each basic block to serve as hidden features of different levels, thereby improving multi-level semantic feature fusion in the network, and improving the feature information extraction and retention capacity of the network. And higher accuracy is achieved in a medical image segmentation task. The problem that small objects are lost andboundary information is blurred and positioned due to the fact that Bottom-up is used is solved.

Description

technical field [0001] The invention relates to the field of computer vision, in particular to a medical image segmentation and implementation method based on a convolutional neural network and a stacked cascade network. Background technique [0002] In recent years, a large number of research methods for machine vision based on deep learning have been proposed. Compared with the previous manual segmentation of pictures, the implementation efficiency and system performance have been greatly improved. In terms of machine vision, convolutional neural networks are usually used including: Gan, DenseNet, Resnet, etc. to extract high-dimensional features of pictures, and use this feature as a high-dimensional representation of pictures. In terms of medical image segmentation, Unet proposed in 2015 is used as a milestone segmentation network, which can better capture the complex texture features of medical images. Medical image segmentation is a key issue that determines whether m...

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): G06T7/00G06T7/11G06T7/136
CPCG06T7/0012G06T7/11G06T7/136G06T2207/10088G06T2207/30016G06T2207/20081
Inventor 田文洪吴智兴陈伏娟
Owner UNIV OF ELECTRONIC SCI & TECH OF CHINA
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