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

Non-rigid multimode medical image registration model establishment method and application thereof

A medical image and registration model technology, applied in image analysis, image enhancement, image data processing, etc., can solve problems such as difficult to achieve image registration, difficult to effectively deal with non-linear grayscale differences of multi-modal images, etc., to achieve fast Efficient medical image registration, clear and complete image representation results, and the effect of precise image registration

Active Publication Date: 2021-07-09
HUAZHONG UNIV OF SCI & TECH
View PDF10 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The above image registration method based on end-to-end deep learning is difficult to effectively deal with the nonlinear gray level difference of multimodal images, and it is difficult to achieve accurate image registration because of indiscriminately learning the deformation of smooth and edge regions. allow

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
  • Non-rigid multimode medical image registration model establishment method and application thereof
  • Non-rigid multimode medical image registration model establishment method and application thereof
  • Non-rigid multimode medical image registration model establishment method and application thereof

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0061] A method for establishing a registration model of a non-rigid multimodal medical image, such as figure 1 shown, including the following steps:

[0062] (1) Obtain a medical dataset in which each sample includes a paired reference image and a floating image and an actual registered image between these two images;

[0063] For the convenience of expression, the reference image and the floating image in each sample in the medical image dataset are denoted as I r and I f , in the subsequent training process, the actual registration image is the label image, which is recorded as I l ;

[0064] (2) Establish the first generative confrontation network GAN_dr, in which the generator G_dr takes the structural representation map of the reference image and the floating image as input to generate the structural representation map of the deformation recovery, and the discriminator D_dr uses the deformation recovery generated by the generator G_dr The structural representation ma...

Embodiment 2

[0126] A registration method for non-rigid multimodal medical images such as Figure 4 shown, including:

[0127] For the floating image to be registered floating image I f and the reference image I r , after calculating the structural characterization graphs IMIND_f and IMIND_r respectively, input to the registration model established by the registration model establishment method of the non-rigid multimodal medical image provided by the above-mentioned embodiment 1, so as to output the registration image I by the registration model g .

[0128] The non-rigid multimodal medical image registration method provided in this embodiment first calculates the structural representation diagram of the reference image and the floating image to be registered, which can convert the multimodal image into a single-modal image, reducing the damage caused by multimodality. impact, the registration model established based on this embodiment can achieve fast and accurate image registration, ...

Embodiment 3

[0130] A computer-readable storage medium, including a stored computer program; when the computer program is executed by a processor, the device where the computer-readable storage medium is located is controlled to execute the method for establishing a registration model of a non-rigid multimodal medical image provided in Embodiment 1 above, And / or the method for registering non-rigid multimodal medical images provided in Embodiment 2 above.

[0131] The registration effect of the present invention will be further explained in conjunction with the comparative experimental results below. During the experiment, five existing registration methods were selected as comparative examples of the above-mentioned embodiment 2, and the comparative examples are as follows:

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 non-rigid multimode medical image registration model establishment method and application thereof, and belongs to the field of medical image registration, and the method comprises the steps: establishing a generative adversarial network GAN_dr, a generator G_dr being used for generating a structural characterization graph for deformation recovery, and a discriminator D_dr being used for judging whether the structural characterization graph generated by the G_dr has effectively recovered deformation; calculating structural representation graphs of the reference image, the floating image and the actual registration image in each sample in the medical data set, and training the GAN_dr by using a calculation result; establishing a generative adversarial network GAN_ie, wherein a generator Gie takes the structural representation graph as input and is used for estimating a registration image, and a discriminator D_ie is used for judging whether the estimated registration image is consistent with an actual registration image or not; using the trained G_dr to generate a structural representation diagram of deformation recovery corresponding to each sample in the medical data set, and training the GAN_ie; and connecting the trained G_ie to the G_dr to obtain a registration model. According to the invention, rapid and accurate matching of medical images can be realized.

Description

technical field [0001] The invention belongs to the field of medical image registration, and more particularly relates to a method for establishing a registration model of a non-rigid multi-mode medical image and an application thereof. Background technique [0002] Due to the different principles of various imaging technologies, each imaging method has its own advantages in reflecting human body information. Functional imaging, such as functional magnetic resonance imaging (fMRI), focuses on reflecting human metabolic information, while anatomical imaging, such as T1-weighted MRI, can more clearly reflect the anatomical structure of the human body. Furthermore, even for anatomical imaging such as T1-weighted MRI and T2-weighted MRI, there are differences in the information they provide. Therefore, the fusion of different information from multimodal images can better assist the diagnosis and treatment of human diseases. Multimodal medical image registration is the basis of...

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): G06T7/33G06T7/13G06T5/50G06K9/62
CPCG06T7/13G06T7/344G06T5/50G06F18/214G06T7/33G06T2207/10088G06T2207/20081G06T2207/20084G06T2207/30016G16H30/20G16H30/40G16H50/20G06T7/337G06V10/454G06V10/761G06T5/20
Inventor 张旭明朱星星
Owner HUAZHONG UNIV OF SCI & TECH
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