Nuclear magnetic resonance image reconstruction method and terminal

A nuclear magnetic resonance image and terminal technology, which is applied in the field of medical image processing, can solve problems such as image artifacts and affect image quality, and achieve the effect of stable and highly robust nuclear magnetic resonance image reconstruction.

Pending Publication Date: 2022-05-13
医诺智能科技(广州)有限公司
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, in the process of MRI, the way of image reconstruction has a great influence on the final imaging effect, and problems such as image artifacts often occur, which affect the quality of the image.

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
  • Nuclear magnetic resonance image reconstruction method and terminal
  • Nuclear magnetic resonance image reconstruction method and terminal
  • Nuclear magnetic resonance image reconstruction method and terminal

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0079] Please refer to figure 1 , Embodiment 1 of the present invention is:

[0080] A nuclear magnetic resonance image reconstruction method, comprising steps:

[0081] S1. Acquire a nuclear magnetic resonance image sample set, where each nuclear magnetic resonance image sample in the nuclear magnetic resonance image sample set includes a K-space image sample and a first reconstructed image;

[0082] Wherein, the K-space image represents a frequency domain image, such as an image in the Fourier domain;

[0083] S2. Pass the K-space image sample through a preset deep neural network to obtain a predicted image and a predicted sensitivity map kernel set; and obtain a second reconstructed image according to the predicted image and the predicted sensitivity map kernel set, specifically:

[0084] S21. Establish an overall optimization target according to the K-space image samples

[0085] Wherein, y represents the K-space image sample, A m Represents a linear operator compo...

Embodiment 2

[0103] A nuclear magnetic resonance image reconstruction method, which differs from Embodiment 1 in that S4 includes:

[0104] S41. Receive the nuclear magnetic resonance image to be processed, and perform preprocessing on the nuclear magnetic resonance image to be processed:

[0105] Pre-generate the sampling mask, the sampling mask is equivalent to a mask, and image data is processed within this range;

[0106]Obtain relevant parameters: MRI (Magnetic Resonance Imaging, nuclear magnetic resonance imaging) sample list index, number of MRI slices nums, number of MRI center slices center_slice_idx, downsampling factor, sampling mask, direction y; and predict sensitivity map maps;

[0107] Convert the data corresponding to the MRI image to be processed in the MRI sample list into a digitized tensor;

[0108] Separate slices and samples; map always starts counting from zero, and the initial value of the counting parameter count=0;

[0109] Load the MRI image to be processed;

...

Embodiment 3

[0134] Please refer to figure 2 , Embodiment three of the present invention is:

[0135] A nuclear magnetic resonance image reconstruction terminal 1, comprising a processor 2, a memory 3, and a computer program stored on the memory 3 and operable on the processor 2, and the embodiment is realized when the processor 2 executes the computer program Each step in one.

[0136] To sum up, this application provides a method and terminal for nuclear magnetic resonance image reconstruction. Through a deep learning method, the reconstruction of nuclear magnetic resonance images is realized on the basis of unfolding and alternating minimization, and the robustness of the reconstruction process is enhanced. Interleaving a learnable model with an optimization step, the entire system is end-to-end trained with a supervised loss. Using a calibration-free approach to structure in parallel nuclear magnetic resonance (MRI) models, sensitivity maps are spatially varied smoothly and low-rank...

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 provides a nuclear magnetic resonance image reconstruction method and a terminal, and the method comprises the steps: obtaining a nuclear magnetic resonance image sample set, and enabling each nuclear magnetic resonance image sample in the nuclear magnetic resonance image sample set to comprise a K space image sample and a first reconstruction image; enabling the K space image sample to pass through a preset deep neural network to obtain a prediction image and a prediction sensitivity mapping kernel set; obtaining a second reconstructed image according to the predicted image and the predicted sensitivity mapping kernel set; optimizing the preset deep neural network according to the difference between the first reconstructed image and the second reconstructed image to obtain a target deep neural network; reconstructing a nuclear magnetic resonance image to be processed according to the target deep neural network; the method can guarantee the stability of the image reconstruction effect under the change of different parameters, does not depend on a determined sensitivity mapping kernel set, and achieves the high-robustness nuclear magnetic resonance image reconstruction.

Description

technical field [0001] The invention relates to the field of medical image processing, in particular to a nuclear magnetic resonance image reconstruction method and a terminal. Background technique [0002] Using nuclear magnetic resonance imaging to generate tissue imaging of the human body to assist in the judgment of various diseases is a commonly used technical means, and because magnetic resonance imaging does not have the problem of radiation exposure, it is also safer for the human body. However, in the process of MRI, the way of image reconstruction has a great influence on the final imaging effect, and problems such as image artifacts often occur, which affect the quality of the image. Contents of the invention [0003] The technical problem to be solved by the present invention is to provide a nuclear magnetic resonance image reconstruction method and a terminal to realize nuclear magnetic resonance image reconstruction with higher robustness. [0004] In order ...

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): G06T11/00G06T7/00G06V10/774G06K9/62G06N3/04G06N3/08
CPCG06T11/003G06T7/97G06N3/04G06N3/08G06T2207/10088G06T2207/20081G06F18/214
Inventor 陈明黄雪芳谢宝文王恩培贾文廷李建东雷国胜龙腾
Owner 医诺智能科技(广州)有限公司
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

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap