WO2023029087A1 - Procédé d'imagerie par résonance magnétique rapide à faible champ, équipement terminal et support de stockage informatique - Google Patents

Procédé d'imagerie par résonance magnétique rapide à faible champ, équipement terminal et support de stockage informatique Download PDF

Info

Publication number
WO2023029087A1
WO2023029087A1 PCT/CN2021/118208 CN2021118208W WO2023029087A1 WO 2023029087 A1 WO2023029087 A1 WO 2023029087A1 CN 2021118208 W CN2021118208 W CN 2021118208W WO 2023029087 A1 WO2023029087 A1 WO 2023029087A1
Authority
WO
WIPO (PCT)
Prior art keywords
magnetic resonance
low
imaging method
data
field
Prior art date
Application number
PCT/CN2021/118208
Other languages
English (en)
Chinese (zh)
Inventor
梁栋
朱庆永
崔卓须
刘元元
郭一凡
朱燕杰
刘新
郑海荣
Original Assignee
中国科学院深圳先进技术研究院
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中国科学院深圳先进技术研究院 filed Critical 中国科学院深圳先进技术研究院
Publication of WO2023029087A1 publication Critical patent/WO2023029087A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R33/00—Arrangements or instruments for measuring magnetic variables
    • G01R33/20—Arrangements or instruments for measuring magnetic variables involving magnetic resonance
    • G01R33/44—Arrangements or instruments for measuring magnetic variables involving magnetic resonance using nuclear magnetic resonance [NMR]
    • G01R33/48—NMR imaging systems
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R33/00—Arrangements or instruments for measuring magnetic variables
    • G01R33/20—Arrangements or instruments for measuring magnetic variables involving magnetic resonance
    • G01R33/44—Arrangements or instruments for measuring magnetic variables involving magnetic resonance using nuclear magnetic resonance [NMR]
    • G01R33/48—NMR imaging systems
    • G01R33/54—Signal processing systems, e.g. using pulse sequences ; Generation or control of pulse sequences; Operator console
    • G01R33/56—Image enhancement or correction, e.g. subtraction or averaging techniques, e.g. improvement of signal-to-noise ratio and resolution
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R33/00—Arrangements or instruments for measuring magnetic variables
    • G01R33/20—Arrangements or instruments for measuring magnetic variables involving magnetic resonance
    • G01R33/44—Arrangements or instruments for measuring magnetic variables involving magnetic resonance using nuclear magnetic resonance [NMR]
    • G01R33/48—NMR imaging systems
    • G01R33/58—Calibration of imaging systems, e.g. using test probes, Phantoms; Calibration objects or fiducial markers such as active or passive RF coils surrounding an MR active material
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T11/00—2D [Two Dimensional] image generation
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
    • Y02A90/30—Assessment of water resources

Definitions

  • the present application relates to the technical field of nuclear magnetic resonance imaging, in particular to a low-field magnetic resonance fast imaging method, terminal equipment and computer storage media.
  • magnetic resonance imaging has no radiation damage, and has rich contrast information, higher soft tissue resolution and functional imaging capabilities.
  • the high-field magnetic resonance system with high image signal-to-noise ratio is complex.
  • my country's high-field magnetic resonance equipment has been heavily dependent on imports, which are expensive.
  • the operation and maintenance of low-field magnetic resonance are simple, and its open magnet structure can eliminate the claustrophobia of the object, and at the same time, it is convenient for interventional diagnosis and treatment under the supervision of the life support system.
  • the signal-to-noise ratio of low-field magnetic resonance signals is low, and the image quality is relatively poor.
  • multiple repeated acquisitions are required to obtain images with high signal-to-noise ratio. How to quickly obtain high-resolution magnetic resonance images has always been a core issue in the field of low-field magnetic resonance.
  • the main strategy for fast magnetic resonance imaging technology is to reduce the acquisition of data space (K space), and through mathematical modeling and optimization, the realization of accelerated imaging is transformed into a regularized solution to an ill-posed inverse problem, such as compressed sensing Methods 1-3.
  • K space data space
  • the K-space observation noise under low-field conditions is relatively large, and the low-field application based on the traditional compressive sensing sparse regularization method cannot meet the clinical requirements of the current low-field ecological layout in terms of imaging quality and acceleration magnification.
  • the present application provides a low-field magnetic resonance fast imaging method, a terminal device and a computer storage medium.
  • the application provides a low-field magnetic resonance fast imaging method, the low-field magnetic resonance fast imaging method comprising:
  • the convolution feature subspace projection regularization model is solved in parallel based on a preset algorithm to obtain high-quality magnetic resonance images.
  • said reading undersampled imaging data about said target image comprises
  • the under-sampled imaging data includes structural information of several directions and/or several orders of the target image.
  • x is the target image, is a set of convolution filter kernels, ⁇ is the preset weight parameter, z i is the i-th convolution feature.
  • the construction of the convolution feature subspace projection regularization model using the under-sampled imaging data includes:
  • the subspace projection regularization model of the convolution feature is constructed by using the undersampled imaging data, the basis function corresponding to the subspace and its transposition.
  • the convolution feature subspace projection regularization model is:
  • u i is the ith subspace projection coefficient.
  • v i is the i-th orthogonal subspace, ⁇ and ⁇ are the weight parameters, z i is the i-th convolution feature.
  • the preset algorithm is the first enhanced plug-and-play iterative algorithm, and the algorithm framework is:
  • x is the undersampled imaging data
  • D ⁇ is a generalized denoising operator
  • x noise is noise information
  • l is a multiplier term.
  • the preset algorithm is the second improved plug-and-play iterative algorithm, and the algorithm framework is:
  • x is the undersampled imaging data
  • D ⁇ is a generalized denoising operator
  • x noise is noise information
  • l is a multiplier term.
  • the present application also provides a terminal device, and the terminal device includes:
  • a reading module configured to read subsampled imaging data about the target image
  • the reconstruction module is used to solve the convolution feature subspace projection regularization model in parallel based on a preset algorithm to obtain high-quality magnetic resonance images.
  • the present application also provides another terminal device, where the terminal device includes a memory and a processor, wherein the memory is coupled to the processor;
  • the memory is used for storing program data
  • the processor is used for executing the program data to realize the above-mentioned low-field magnetic resonance fast imaging method.
  • the present application also provides a computer storage medium, the computer storage medium is used for storing program data, and when the program data is executed by a processor, it is used to realize the above-mentioned low-field magnetic resonance fast imaging method.
  • the terminal device reads the under-sampled imaging data about the target image; constructs a convolutional feature subspace projection regularization model based on low-field data characteristics; regularizes the convolutional feature subspace projection based on a preset algorithm
  • the optimized model is solved in parallel to obtain high-quality magnetic resonance images.
  • the low-field magnetic resonance fast imaging method of the present application uses the convolution feature subspace projection regularization model and the preset algorithm to accelerate the acquisition of comparable high-field magnetic resonance images, which is conducive to improving the efficiency of clinical application.
  • Fig. 1 is a schematic flow chart of an embodiment of the low-field magnetic resonance fast imaging method provided by the present application
  • Fig. 2 is a schematic framework diagram of an embodiment of the low-field magnetic resonance fast imaging method provided by the present application
  • Fig. 3 is a schematic diagram of comparison between high-field data and data of cardiac images provided by the present application.
  • FIG. 4 is a schematic structural diagram of an embodiment of a terminal device provided by the present application.
  • FIG. 5 is a schematic structural diagram of another embodiment of a terminal device provided by the present application.
  • Fig. 6 is a schematic structural diagram of an embodiment of a computer storage medium provided by the present application.
  • the main disadvantages of the traditional compressed sensing sparse regularization method for low-field fast MRI include: a strong degree of observation noise has an impact on the reliability of the measurement in the sparse domain of the image, that is, it cannot reliably express sparse image features; The solution accuracy of the problem and the efficiency of the algorithm implementation cannot meet the actual clinical needs.
  • the present application proposes a low-field magnetic resonance fast imaging method, specifically a convolution feature subspace projection regularized imaging method aimed at improving the quality of low-field magnetic resonance imaging.
  • FIG. 1 is a schematic flowchart of an embodiment of a low-field magnetic resonance fast imaging method provided by the present application.
  • the low-field magnetic resonance fast imaging method of the present application is applied to a terminal device, wherein the terminal device of the present application may be a server, or may be a system in which the server and the terminal device cooperate with each other.
  • the terminal device of the present application may be a server, or may be a system in which the server and the terminal device cooperate with each other.
  • various parts included in the terminal device such as various units, subunits, modules, and submodules, may all be set in the server, or may be set in the server and the terminal device separately.
  • the above server may be hardware or software.
  • the server When the server is hardware, it can be implemented as a distributed server cluster composed of multiple servers, or as a single server.
  • the server When the server is software, it can be implemented as multiple software or software modules, such as software or software modules used to provide a distributed server, or as a single software or software module, which is not specifically limited here.
  • the magnetic resonance imaging method in the embodiment of the present application may be implemented in a manner in which a processor invokes computer-readable instructions stored in a memory.
  • the low-field magnetic resonance fast imaging method of the embodiment of the present application specifically includes the following steps:
  • Step S11 Read the under-sampled imaging data about the target image.
  • the terminal device reads the original data file collected by the nuclear magnetic resonance equipment, and the original data is used as full-sampled K-space data, and the simulated under-sampling processing operation is performed to obtain randomly variable density under-sampled space data.
  • the MRI data acquisition process based on the K-space undersampling mechanism can be expressed discretely as:
  • ⁇ R + is a positive constant for balanced data fitting and regularization.
  • the terminal device performs convolution processing on the target image by using convolutional feature coding, so as to obtain under-sampled imaging data of the target image.
  • convolutional sparse coding is to convert an image equivalent to a set of linear filter kernels and corresponding sparse feature maps.
  • the specific model of convolutional sparse coding provided by the embodiment of the present disclosure is:
  • the benefit of using convolutional feature coding in the embodiments of the present disclosure is that different levels of structural information of the target image can be obtained, specifically including different directions and different orders.
  • Step S12 Construct a convolutional feature subspace projection regularization model based on the characteristics of the low-field data.
  • the signal-to-noise ratio of the image is low under low-field conditions, and the convolution feature zi is sensitive to noise and other interference factors, and the modeling is directly based on the feature zi (such as sparse metrics in the horizontal and vertical gradient domains) will limit the reliability of the overall regularization.
  • the embodiment of the present disclosure proposes a convolution feature subspace projection regularization model based on anti-noise interference.
  • the model is specifically:
  • u i is the ith subspace projection coefficient.
  • v i is the ith orthogonal subspace.
  • ⁇ and ⁇ are weight parameters.
  • z i is the i-th convolutional feature.
  • Step S13 Solving the convolution feature subspace projection regularization model in parallel based on a preset algorithm to obtain a high-quality magnetic resonance image.
  • the preset algorithm is specifically an improved plug-and-play iterative algorithm.
  • Plug-and-Play Prior (P3) iteration is rooted in the Proximal Gradient algorithm, taking the ADMM algorithm of primal-dual alternate optimization as an example.
  • the specific solution model including convolution feature subspace projection regularization is:
  • the embodiment of the present application proposes two improved plug-and-play iterative algorithms.
  • the denoising step uses an improved denoising mechanism to further improve the imaging quality and avoid smooth transitions in the reconstruction process:
  • the auxiliary variable w is introduced to solve a saddle-point problem as follows:
  • the standard ADMM algorithm iteration framework includes the quadratic programming subproblem about x, the proximal gradient projection of the auxiliary variable w, and the update of the multiplier term l.
  • the embodiments of the present disclosure directly replace the traditional plug-and-play iteration with a generalized denoising operator D ⁇ for the proximal gradient projection prox ⁇ (w; ⁇ ) (subproblem about w), such as non-local matching and filtering ( BM3D) 10.
  • D ⁇ for the proximal gradient projection prox ⁇ (w; ⁇ ) (subproblem about w), such as non-local matching and filtering ( BM3D) 10.
  • the algorithm steps of P3-ADMM are:
  • t represents the number of iterations of the denoising link.
  • the embodiment of the present application adopts the above-mentioned boosted plug-and-play prior (Boosting P3) iterative algorithm to further describe the details of the image and avoid excessive smoothing in the reconstruction process.
  • Boosting P3 boosted plug-and-play prior
  • FIG. 2 is a schematic framework diagram of an embodiment of a low-field magnetic resonance fast imaging method provided by the present application.
  • the feature subspace operators under different convolution kernels are separable, so different computing targets can be allocated to multiple computing units, so as to realize parallel iteration of the algorithm, which is beneficial to avoid lengthy waiting of the algorithm process.
  • the terminal device reads the under-sampled imaging data of the target image; constructs a convolutional feature subspace projection regularization model based on low-field data characteristics; and regularizes the convolutional feature subspace projection based on a preset algorithm.
  • the optimized model is solved in parallel to obtain high-quality magnetic resonance images.
  • FIG. 3 is a schematic diagram of a comparison between high-field data and low-field data of cardiac images provided by the present application.
  • the embodiment of the present application simulates a low-field scene by adding noise, taking heart data as an example to verify the effectiveness of the proposed magnetic resonance imaging method.
  • the left side of Figure 3 is a heart image of a certain frame (upper: high-field data; lower: low-field data), and the middle column of Figure 3 is the image convolution feature amplitude map in the horizontal gradient direction, which can be clearly observed. Compared with the sparsity of the original image, the sparsity of the noisy one is significantly reduced due to the influence of noise.
  • the right side of Figure 3 is the feature amplitude map based on subspace learning (low-rank space). It is not difficult to find that the sparsity of the subspace containing noisy features can still be guaranteed, and even shows a similar sparsity to the original image. .
  • the low-field magnetic resonance fast imaging method provided by this application strives for more reliable and robust imaging modeling against the background of strong noise on the one hand, and strives for higher accuracy and speed for problem solving efficiency faster.
  • the magnetic resonance imaging method provided by this application first constructs a convolution feature subspace regularization model, focusing on the effective expression of convolution features, which is used to avoid the sensitivity of the original space regularization to noise; secondly, an improved plug-and-play Using a priori iterative algorithm, and considering the advantages of parallel computing under the condition of partial operator separability, using operator separability, parallel computing can obtain more accurate problem solutions.
  • the writing order of each step does not mean a strict execution order and constitutes any limitation on the implementation process.
  • the specific execution order of each step should be based on its function and possible
  • the inner logic is OK.
  • FIG. 4 is a schematic structural diagram of an embodiment of the terminal device provided in the present application.
  • the terminal device 300 includes a reading module 31 , a building module 32 and a reconstruction module 33 .
  • the reading module 31 is used to read the under-sampled imaging data about the target image;
  • the construction module 32 is used to construct the convolution feature subspace projection regularization model based on the low-field data characteristics;
  • the algorithm is designed to solve the convolution feature subspace projection regularization model in parallel to obtain high-quality magnetic resonance images.
  • FIG. 5 is a schematic structural diagram of another embodiment of the terminal device provided in the present application.
  • the terminal device 400 in this embodiment of the present application includes a memory 41 and a processor 42, where the memory 41 and the processor 42 are coupled.
  • the memory 41 is used for storing program data
  • the processor 42 is used for executing the program data to realize the low-field magnetic resonance fast imaging method described in the above-mentioned embodiments.
  • the processor 42 may also be referred to as a CPU (Central Processing Unit, central processing unit).
  • the processor 42 may be an integrated circuit chip with signal processing capabilities.
  • the processor 42 can also be a general-purpose processor, a digital signal processor (DSP, Digital Signal Process), an application specific integrated circuit (ASIC, Application Specific Integrated Circuit), a field programmable gate array (FPGA, Field Programmable Gate Array) or other possible Program logic devices, discrete gate or transistor logic devices, discrete hardware components.
  • DSP digital signal processor
  • ASIC Application Specific Integrated Circuit
  • FPGA Field Programmable Gate Array
  • Program logic devices discrete gate or transistor logic devices, discrete hardware components.
  • the general purpose processor can be a microprocessor or the processor 42 can be any conventional processor or the like.
  • the present application also provides a computer storage medium.
  • the computer storage medium 300 is used to store program data 31.
  • the program data 31 is executed by the processor, it is used to realize the low-field magnetic Resonance Fast Imaging Method.
  • the present application also provides a computer program product, wherein the computer program product includes a computer program, and the computer program is operable to cause a computer to execute the magnetic resonance imaging method as described in the embodiment of the present application.
  • the computer program product may be a software installation package.
  • the low-field magnetic resonance fast imaging method described in the above-mentioned embodiments of the present application exists in the form of a software function unit and is sold or used as an independent product when implemented, and can be stored in a device, such as a computer-readable storage medium middle.
  • a device such as a computer-readable storage medium middle.
  • the technical solution of the present application is essentially or part of the contribution to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium , including several instructions to make a computer device (which may be a personal computer, server, or network device, etc.) or a processor (processor) execute all or part of the steps of the methods described in various embodiments of the present invention.
  • the aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disc, etc., which can store program codes. .

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • High Energy & Nuclear Physics (AREA)
  • Condensed Matter Physics & Semiconductors (AREA)
  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Radiology & Medical Imaging (AREA)
  • Signal Processing (AREA)
  • Theoretical Computer Science (AREA)
  • Magnetic Resonance Imaging Apparatus (AREA)

Abstract

La présente invention concerne un procédé d'imagerie par résonance magnétique rapide à faible champ, un dispositif terminal et un support de stockage informatique. Le procédé d'imagerie par résonance magnétique rapide à faible champ consiste à : lire des données d'imagerie sous-échantillonnées associées à une image cible ; construire un modèle de régularisation de projection sur un sous-espace de caractéristiques de convolution sur la base de caractéristiques de données de faible champ ; et résoudre le modèle de régularisation de projection sur un sous-espace de caractéristiques de convolution sur la base d'un algorithme prédéfini pour obtenir une image de résonance magnétique de haute qualité. De cette manière, le procédé d'imagerie par résonance magnétique rapide à faible champ de la présente demande utilise le modèle de régularisation de projection sur un sous-espace de caractéristiques de convolution et l'algorithme prédéfini pour obtenir rapidement l'image de résonance magnétique comparable à une image de résonance magnétique à champ puissant, et permet d'améliorer l'efficacité d'applications cliniques.
PCT/CN2021/118208 2021-09-03 2021-09-14 Procédé d'imagerie par résonance magnétique rapide à faible champ, équipement terminal et support de stockage informatique WO2023029087A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202111032442.0 2021-09-03
CN202111032442.0A CN115754857A (zh) 2021-09-03 2021-09-03 一种低场磁共振快速成像方法、终端设备及计算机存储介质

Publications (1)

Publication Number Publication Date
WO2023029087A1 true WO2023029087A1 (fr) 2023-03-09

Family

ID=85332451

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/118208 WO2023029087A1 (fr) 2021-09-03 2021-09-14 Procédé d'imagerie par résonance magnétique rapide à faible champ, équipement terminal et support de stockage informatique

Country Status (2)

Country Link
CN (1) CN115754857A (fr)
WO (1) WO2023029087A1 (fr)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20170053402A1 (en) * 2014-04-30 2017-02-23 Samsung Electronics Co., Ltd. Magnetic resonance imaging device and method for generating magnetic resonance image
CN108717171A (zh) * 2018-05-24 2018-10-30 上海理工大学 一种压缩感知低场磁共振成像算法
CN109671129A (zh) * 2018-12-14 2019-04-23 深圳先进技术研究院 一种自适应参数学习的动态磁共振图像重建方法和装置
CN110490832A (zh) * 2019-08-23 2019-11-22 哈尔滨工业大学 一种基于正则化深度图像先验方法的磁共振图像重建方法
US20200355774A1 (en) * 2017-06-06 2020-11-12 Shenzhen Institutes Of Advanced Technology One-dimensional partial fourier parallel magnetic resonance imaging method based on deep convolutional network

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20170053402A1 (en) * 2014-04-30 2017-02-23 Samsung Electronics Co., Ltd. Magnetic resonance imaging device and method for generating magnetic resonance image
US20200355774A1 (en) * 2017-06-06 2020-11-12 Shenzhen Institutes Of Advanced Technology One-dimensional partial fourier parallel magnetic resonance imaging method based on deep convolutional network
CN108717171A (zh) * 2018-05-24 2018-10-30 上海理工大学 一种压缩感知低场磁共振成像算法
CN109671129A (zh) * 2018-12-14 2019-04-23 深圳先进技术研究院 一种自适应参数学习的动态磁共振图像重建方法和装置
CN110490832A (zh) * 2019-08-23 2019-11-22 哈尔滨工业大学 一种基于正则化深度图像先验方法的磁共振图像重建方法

Also Published As

Publication number Publication date
CN115754857A (zh) 2023-03-07

Similar Documents

Publication Publication Date Title
Hauptmann et al. Real‐time cardiovascular MR with spatio‐temporal artifact suppression using deep learning–proof of concept in congenital heart disease
US11324418B2 (en) Multi-coil magnetic resonance imaging using deep learning
Koonjoo et al. Boosting the signal-to-noise of low-field MRI with deep learning image reconstruction
Liu et al. Projected iterative soft-thresholding algorithm for tight frames in compressed sensing magnetic resonance imaging
CN104933683B (zh) 一种用于磁共振快速成像的非凸低秩重建方法
WO2018099321A1 (fr) Procédé de reconstruction d'une imagerie par résonance magnétique de norme nucléaire pondérée à base d'arbre clairsemé généralisé
US10671939B2 (en) System, method and computer-accessible medium for learning an optimized variational network for medical image reconstruction
CN113795764A (zh) 用于磁共振图像重建的深度学习技术
WO2020114329A1 (fr) Imagerie paramétrique par résonance magnétique rapide et dispositif
Malavé et al. Reconstruction of undersampled 3D non‐Cartesian image‐based navigators for coronary MRA using an unrolled deep learning model
JP2022521499A (ja) 予測モデルを用いたモデルパラメータの判定
Pal et al. A review and experimental evaluation of deep learning methods for MRI reconstruction
US20220075017A1 (en) Machine learning for simultaneously optimizing an under-sampling pattern and a corresponding reconstruction model in compressive sensing
KR102622283B1 (ko) 맥스웰 병렬 이미징
Kelkar et al. Prior image-constrained reconstruction using style-based generative models
Sedlar et al. Diffusion MRI fiber orientation distribution function estimation using voxel-wise spherical U-net
van der Heide et al. High‐resolution in vivo MR‐STAT using a matrix‐free and parallelized reconstruction algorithm
Yang et al. Generative Adversarial Networks (GAN) Powered Fast Magnetic Resonance Imaging--Mini Review, Comparison and Perspectives
Manimala et al. Sparse MR image reconstruction considering Rician noise models: A CNN approach
WO2022193379A1 (fr) Procédé et appareil de génération de modèle de reconstruction d'image, procédé et appareil de reconstruction d'image, dispositif et support
WO2016090686A1 (fr) Procédé et appareil de débruitage d'un tenseur de diffusion par résonance magnétique et produit programme informatique
Kleineisel et al. Real‐time cardiac MRI using an undersampled spiral k‐space trajectory and a reconstruction based on a variational network
Sui et al. Simultaneous image reconstruction and lesion segmentation in accelerated MRI using multitasking learning
WO2024021796A1 (fr) Procédé et appareil de traitement d'image, dispositif électronique, support de stockage et produit de programme
WO2023029087A1 (fr) Procédé d'imagerie par résonance magnétique rapide à faible champ, équipement terminal et support de stockage informatique

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21955618

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE