Check patentability & draft patents in minutes with Patsnap Eureka AI!

A PET/MRI heterogeneous brain image information fusion method with improved neural network

A neural network and image information technology, applied in image enhancement, image analysis, graphics and image conversion, etc., can solve problems such as translation invariance, improve diagnostic efficiency and accuracy, improve medical image quality, and have wide application prospects Effect

Active Publication Date: 2021-05-18
XINXIANG MEDICAL UNIV
View PDF1 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Contourlet transform is a multi-resolution, localized and multi-directional sparse representation method, which makes up for the defects of wavelet transform, but it needs to be up and down sampling during the transformation process, and it does not have translation invariance.

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 PET/MRI heterogeneous brain image information fusion method with improved neural network
  • A PET/MRI heterogeneous brain image information fusion method with improved neural network
  • A PET/MRI heterogeneous brain image information fusion method with improved neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0077] Example 1: MRI-PET coronal image fusion

[0078] Specifically follow the steps below:

[0079]Step 1: Perform HIS transformation on the PET image to be fused to obtain the corresponding brightness, hue, saturation components and RGB three-channel components;

[0080] Step 2: Perform edge-based non-rigid registration on the PET image of the brightness component and the MRI image to achieve the alignment of the brain structure;

[0081] 1) Use templates in 8 directions to calculate the gradient image, and perform binarization and refinement on the gradient image to obtain brain contour pixels;

[0082] 1.1) The templates for 8 directions are as follows:

[0083]

[0084] 1.2) Binary processing is performed by automatically obtaining the threshold;

[0085] 1.3) Use the bwmorph function to realize the thinning algorithm, such as image 3 shown.

[0086] 2) The number of contour pixels is the same by equidistant point selection and interpolation method, the spatial ...

Embodiment 2

[0108] Example 2: MRI-PET cross-sectional image fusion

[0109] The specific implementation steps can refer to Example 1, and the parameter settings are exactly the same.

[0110] In order to verify the feasibility and effectiveness of the present invention, brain PET images and MRI images are used for heterogeneous information fusion. Table 1 below gives an objective evaluation of fusion results obtained by different fusion methods.

[0111] Table 1: Objective evaluation of fusion results in Example 1

[0112]

[0113]

[0114] By using standard deviation, entropy, sharpness, average gradient, Q abf To measure the quality of the fused image, the standard deviation reflects the degree of dispersion of the gray level, the larger the standard deviation, the richer the information; the entropy represents the amount of information in the image, the greater the entropy, the greater the amount of information contained; The ability to express tiny details, the higher the def...

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 PET / MRI different machine brain image information fusion method with improved neural network. The specific implementation steps are: 1) the PET image (index image) to be fused is subjected to HIS transformation and color transformation, and converted into IHS and RGB Channel information; 2) Use PET and MRI rigid registration based on brain contour pixels to align brain anatomical positions through translation, rotation, and scaling; 3) Use NSCT transformation to obtain frequency features of different scales and directions, and design based on local Neighborhood weighted average low-frequency information fusion rules and spatial frequency excitation PCNN high-frequency information fusion rules realize the information fusion of RGB three-channel PET images and MRI grayscale images. This method makes full use of the functional information provided by PET images and the anatomical and soft tissue information provided by MRI images to improve the efficiency and accuracy of doctors' diagnosis, and can replace some functions of PET / MRI fusion equipment.

Description

technical field [0001] The invention belongs to the technical field of medical image processing and application, and specifically relates to a PET / MRI different-machine brain image information fusion method with improved neural network, which can realize rigid registration of brain PET and MRI image features, alignment of anatomical positions, Moreover, the functional information and anatomical information are fused on one image to achieve a good fusion effect, and both visual and objective evaluation indicators meet the requirements of clinical diagnosis. Background technique [0002] CT, MRI, SPECT, PET and other imaging equipment can provide clinicians with different modalities of diagnostic information, and multimodal information fusion is the current development direction of imaging equipment. PET / CT fusion equipment has been used clinically, but there are still many technical problems to be solved for PET / MRI fusion. PET / MRI different-machine image information fusion ...

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 Patents(China)
IPC IPC(8): G06T3/00G06T5/00G06T5/50G06T7/33
CPCG06T5/50G06T7/33G06T2207/20221G06T2207/30016G06T2207/20084G06T2207/10088G06T2207/10104G06T3/14G06T5/90
Inventor 王昌任琼琼程雅青于毅赵宗亚秦鑫张文超
Owner XINXIANG MEDICAL UNIV
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

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

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com