Training method and device for improving ventricular segmentation model, electronic equipment and medium

A technology of segmentation model and training method, applied in the field of computer vision, can solve the problems of wasting manpower and time, the error of model segmentation results is large, and the training samples cannot cover a variety of situations, so as to achieve the effect of improving accuracy.

Pending Publication Date: 2021-07-30
THE SECOND PEOPLES HOSPITAL OF SHENZHEN
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This method requires a large number of samples of hydrocephalus, which wastes a lot of manpower and time, and the deformation of the ventricles of hydrocephalus is diverse. The training samples may not cover a variety of situations, resulting in large errors in the segmentation results of the model.

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
  • Training method and device for improving ventricular segmentation model, electronic equipment and medium
  • Training method and device for improving ventricular segmentation model, electronic equipment and medium
  • Training method and device for improving ventricular segmentation model, electronic equipment and medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0028] In order to better understand the present application scheme, the technical solutions in the present application embodiment will be apparent from the drawings in the present application, and the described embodiments are clearly described herein. It is an embodiment of the present application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative labor, are the scope of the present application.

[0029] The specification and claims of the present application and the terms "first", "second", "second" or the like are used to distinguish different objects, not to describe a particular order. Moreover, the terms "including" and "have" and any variation, intended to cover the inclusion of the inclusion. For example, a series of steps or units comprising a series of steps or units are not limited to the listed steps or units, but alternatively include the steps or units not l...

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 training method and device for improving a ventricular segmentation model, electronic equipment and a medium. The method comprises the following steps: acquiring a normal brain sample image; acquiring a standard brain image and a ventricular label corresponding to the standard brain image, wherein the standard brain image is a human brain three-dimensional image in a preset standard space, and the ventricular label is used for marking the position of a ventricular area; calculating a transformation relation between the normal brain sample image and the standard brain image; converting the ventricular tag of the standard brain image into a space coordinate system corresponding to the normal brain sample image based on the transformation relation to obtain the ventricular tag of the normal brain sample image; performing three-dimensional image deformation processing on the plurality of normal brain sample images and the ventricular labels of the normal brain sample images to obtain a plurality of ventricular deformation sample images and ventricular labels of the ventricular deformation sample images; and training the network model by using a sample data set to obtain a ventricular segmentation model, the sample data set comprising a plurality of ventricular deformation sample images and ventricular tags of the plurality of ventricular deformation sample images.

Description

Technical field [0001] The present invention relates to the field of computer visual technology, in particular, to a training method, apparatus, electronic device, and medium for improving the split model of the ventricle. Background technique [0002] With the development of advanced medical imaging technology, many neuroscience studies have placed the difference between the anatomical structure in the brain, so as to seek the characteristics of the anatomical structural forms related to brain disease, to improve the reliability of brain disease diagnosis. And the effectiveness of the treatment plan. Medical image segmentation as an important branch in the field of image segmentation is the primary, key steps to achieve medical image analysis, and to complete medical image understanding. [0003] With the development of magnetic resonance imaging (MRI) technology, magnetic resonance (MR) images can provide high contrast and high resolution of three-dimensional (3D) medical image...

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): G06T7/00G06T7/11G06K9/62G06N3/04G06N3/08
CPCG06T7/0012G06T7/11G06N3/08G06T2207/10088G06T2207/20081G06T2207/20084G06T2207/30016G06N3/045G06F18/214
Inventor 林帆
Owner THE SECOND PEOPLES HOSPITAL OF SHENZHEN
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap