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

Medical image registration method fusing multi-scale information

A magnetic resonance image and heart technology, applied in the field of medical image processing, can solve the problems of manual registration misjudgment, missed diagnosis, traditional registration methods can not meet the real-time medical image registration, etc., to achieve rapid image registration, registration The effect of fast speed and strong generalization ability

Inactive Publication Date: 2021-12-03
HENAN UNIVERSITY OF TECHNOLOGY
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] In order to solve the situation of misjudgment and missed diagnosis in manual registration in medical image registration, the traditional registration method cannot meet the real-time performance of medical image registration, and provides a method using convolutional neural network to automatically and accurately register 3D Cardiac Magnetic Resonance Imaging Methods

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 registration method fusing multi-scale information
  • Medical image registration method fusing multi-scale information

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0014] In order to verify the registration performance of the present invention in 3D cardiac magnetic resonance images, we selected the MM-WHS2020 public dataset for training, verification and testing.

[0015] Step 1: Preprocess the 40 cardiac magnetic resonance images, and import the SimpleITK library function under the Pytorch framework to realize the affine and normalization of the data.

[0016] Step 2: Train the improved U-shaped convolutional network in the Pycharm development software, use the Adam optimizer to optimize, and set the learning rate to 0.0001. The batch size is set to 2, every 100 training is an epoch, and the training is 1500epoch. The 40 images are randomly divided into 3:1:1 ratio for training, validation and testing. Train the network until the network converges and stop training.

[0017] Step 3: This experiment uses the untrained data in the MM-WHS2020 public data set for testing, and the experimental results are evaluated by the coincidence coef...

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

An existing 3D heart magnetic resonance image registration technology has defects in registration precision and speed, and is poor in coping capacity for large deformation in an image to be registered. The invention provides a heart magnetic resonance image registration method based on fusion of multi-scale information in order to realize efficient and rapid registration of heart magnetic resonance images. The multi-scale information is an advanced feature with image space information, and the multi-scale information is fused in the feature extraction process of heart images, so that the similarity between the images can be accurately measured. A structure for extracting multi-scale features is added in the U-shaped convolutional network, higher-level features in data are extracted, and the generalization ability of the model is enhanced by automatically learning the inherent internal relation between the same data set. According to the invention, accurate and rapid registration can be realized.

Description

technical field [0001] The invention belongs to the technical field of medical image processing, and in particular relates to a method for registration of 3D cardiac magnetic resonance images. Background technique [0002] Medical image registration refers to the process of matching two images in spatial position through a series of spatial transformations. It is mainly used in lesion detection, disease diagnosis, surgical planning, surgical navigation, and curative effect evaluation in clinical treatment. At present, in the process of clinical diagnosis, manual registration is generally performed by experienced experts based on existing clinical medical knowledge, personal experience and spatial imagination. The method of manual registration is not only time-consuming and laborious, but also the results of the registration vary from person to person and are highly subjective. When doctors work for a long time, they will also cause large errors due to fatigue. Therefore, u...

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): G06T7/00G06T7/11G06T7/33G06N3/04G06N3/08
CPCG06T7/0012G06T7/11G06T7/33G06N3/08G06T2207/10088G06T2207/20081G06T2207/20084G06T2207/20132G06T2207/30048G06N3/045
Inventor 白鑫昊崔晓娟杨铁军李磊樊超巩跃洪苗建雨
Owner HENAN UNIVERSITY OF TECHNOLOGY
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