Unlock instant, AI-driven research and patent intelligence for your innovation.

Multi-modal medical image registration method and device based on MRF model, platform and medium

A medical image and multi-modal technology, applied in the field of medical image processing, can solve problems such as automatic selection of marker points or large image segmentation errors, affecting registration accuracy, and high computational complexity

Inactive Publication Date: 2019-05-10
SOUTH CHINA UNIV OF TECH
View PDF4 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Among them, compared with other image registration methods, the image registration method based on mutual information has high computational complexity and takes a long time.
The registration technology based on feature points and image segmentation first needs to determine the marker points or segment the image. However, due to the low quality of the ultrasound image, the error of automatic selection of marker points or image segmentation is relatively large. Manual selection of marker points or manual image segmentation The real-time performance is not high
For the image registration method based on feature description, its operation based on image blocks will lose part of the local information, which will affect the registration accuracy, and this type of method is suitable for small deformations. As the degree of deformation increases, the registration error increases

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
  • Multi-modal medical image registration method and device based on MRF model, platform and medium
  • Multi-modal medical image registration method and device based on MRF model, platform and medium
  • Multi-modal medical image registration method and device based on MRF model, platform and medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0121] like figure 1 As shown, this embodiment provides a multimodal medical image registration method based on the MRF model, and the specific steps of the method include:

[0122] Input two images to be registered, which are respectively recorded as a fixed image and a floating image;

[0123] Construct the energy function of the Markov random field model;

[0124] Input the energy function into the improved optimization model for optimization calculation to obtain the final transformation parameters;

[0125] Display the final registration result.

[0126] like figure 2 Shown is the specific flow chart of the energy function steps for constructing the Markov random field model, including:

[0127] S201. Obtain an input image to be registered;

[0128] S202. Process the floating image, and normalize the fixed image and the processed floating image;

[0129] S203. Mark the corresponding special positions of the fixed map and the floating map in the form of a weight map...

Embodiment 2

[0187] like Figure 4 As shown, the present embodiment is a multimodal medical image registration device based on the MRF model, including:

[0188] An image generation module is used to input two images to be registered, which are respectively recorded as a fixed image and a floating image;

[0189] The energy function building block is used to construct the energy function of the Markov random field model;

[0190] An optimization calculation module is used to input the energy function into the improved optimization model for optimization calculation to obtain the final transformation parameters;

[0191] The registration result display module is used to display the final registration result.

[0192] Further, in the energy function building block, including:

[0193] The image conversion unit is used to perform arbitrary translation, flip, rotation and scale transformation in the x and y directions of the floating image;

[0194] An image normalization unit is used to n...

Embodiment 3

[0210] This embodiment provides a platform and a storage medium.

[0211] A platform includes a processor and a memory for storing a program executable by the processor. When the processor executes the program stored in the memory, the medical image registration method of Embodiment 1 above is implemented.

[0212] A storage medium, which is a computer-readable storage medium, stores a computer program, and when the computer program is executed by a processor, the medical image registration method in Embodiment 1 above is realized.

[0213] The storage medium in this embodiment may be a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM, Read-Only-Memory), a random-access memory (RAM, Random-Access-Memory), a U disk, a mobile hard disk, and the like.

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 multi-modal medical image registration method based on an MRF model, and the method comprises the specific steps: inputting two images to be registered, and respectively recording the two images as a fixed image and a floating image; Constructing an energy function of the Markov random field model; inputting the energy function into an improved optimization model to carryout optimization calculation to obtain a final transformation parameter; and displaying the finally obtained registration result. Modal transformation is introduced, a new Markov energy function is constructed according to the two image matrixes subjected to the modal transformation and the original registered image, and meanwhile, an improved gradient descent algorithm is introduced to optimizethe energy function, so that a registration result is obtained. And finally, different medical images are subjected to registration experiments to verify that the method has good effectiveness and anti-noise performance.

Description

technical field [0001] The present invention relates to the field of medical image processing, in particular to an MRF model-based multimodal medical image registration method, device, platform and medium. Background technique [0002] Image registration is the process of summarizing image information and aligning images, while multimodal image registration is a technique for aligning multimodal images with spatial geometric relationships. At present, most of the clinical needs can be met by existing algorithms, but considering that the content of individual images varies greatly and contains a lot of information, there are various registration algorithms today. Certain limiting factors exist, and registration can generally only be satisfied under specific circumstances. Considering that image registration has a wide range of applications in clinical medicine, and each different application requires a different deformation field, there is no universal algorithm to solve the...

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): G06T7/33
Inventor 陆以勤刘晓骏覃健诚
Owner SOUTH CHINA UNIV OF TECH