A lung CT and MRI image fusion algorithm

A CT image and image fusion technology, applied in the fields of medical image processing and computer vision, can solve problems such as being susceptible to noise interference and loss of image details

Pending Publication Date: 2019-05-17
FUDAN UNIV
View PDF0 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The algorithm of medical image fusion also mainly involves the traditional pixel-level image fusion method, including the small improvement of the pixel value selection scheme, that is, without any transformation domain processing on the image, and directly averages the corresponding pixels on each source image. After simple operations such as selection, it is fused into a new image; the image is decomposed into multiple layers in the wavelet domain, the image wavelet pyramid is constructed, and the wavelet coefficients of different layers are processed differently. This method has a great breakthrough in medical image fusion. In some scenes, better results than traditional image fusion algorithms are obtained, but this method also has certain limitations, sometimes resulting in the loss of image details, and being susceptible to noise interference, etc.

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
  • A lung CT and MRI image fusion algorithm
  • A lung CT and MRI image fusion algorithm
  • A lung CT and MRI image fusion algorithm

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0063] The present invention proposes an automatic image fusion algorithm of a lung CT image and an MRI sequence image, which is further described in detail in conjunction with the accompanying drawings and embodiments as follows:

[0064] Prepare initial data. Collect scan images of 50 lung cancer patients, each patient collects its CT scan images, MRI T1 or T2, and DWI sequence scan images, and asks three chief physicians in the radiology department of the hospital to select some layers with greater fusion significance, and Finally, opinions on the fusion effect are given at these levels.

[0065] (1) Initial dataset registration (image preprocessing):

[0066] (1-1) Based on the T1 or T2 sequence images, set the reference point, first register the lung CT images to the T1 or T2 sequence images through the normalized mutual information maximization criterion method, and check the registration on the 3D images. quasi effect.

[0067] (1-2) Take the sequence with a b value ...

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 belongs to the technical field of medical image processing and computer vision, and particularly relates to a lung CT and MRI image fusion algorithm. According to the method, an image isfused under an inverse problem model, a loss function is constructed under a variational framework, and an optimal solution is obtained by minimizing the loss function. According to the method, the research advantage of the image in a wavelet domain is utilized, the wavelet coefficient is put into a loss function, and non-convex regularization is carried out on the wavelet coefficient to increasethe sparsity of the wavelet coefficient, so that a better image recovery effect is obtained. The convex property of the whole loss function is kept by adjusting parameters, so that a global optimal solution is obtained through convex optimization, and finally, a fusion image is obtained through wavelet inverse transformation. According to the method, the advantages of CT imaging in the aspect oflung texture and the advantages of an MRI imaging sequence in the aspect of lung focus are combined, the information amount of the fusion image is enriched, a doctor can more clearly find the advantage information of the two imaging sensors in lung imaging on the fusion image, and more accurate judgment can be made in a shorter time.

Description

technical field [0001] The invention belongs to the technical field of medical image processing and computer vision, and in particular relates to a lung CT and MRI image fusion algorithm. Background technique [0002] The lung is an important organ that maintains the human respiratory system. Lung disease is one of the important diseases that threaten human life today. Common lung diseases include lung cancer and tuberculosis. Among them, lung cancer is the fastest growing morbidity and mortality rate. It is one of the most serious malignant tumors that threaten health and life. In the past 50 years, many countries have reported that the morbidity and mortality of lung cancer have increased significantly. The incidence and mortality of lung cancer in men rank first among all malignant tumors, and the incidence and mortality of lung cancer occupy the second place in women. For the detection of lung cancer and other diseases, lung CT scans are usually used. CT scans can cover...

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): G06T5/50G06T7/33
Inventor 郑忍成王鹤单飞杨舒一施裕新
Owner FUDAN UNIV
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