A magnetic resonance imaging method, system, apparatus, and storage medium

Abstract

Embodiments of the present specification provide a magnetic resonance imaging method, system, device and storage medium, the method comprising: acquiring a navigation signal and an imaging signal collected in a magnetic resonance scan of a target object; processing the navigation signal using a singular value decomposition algorithm to obtain L time basis functions corresponding to L singular values, L being a positive integer greater than 1; based on the imaging signal and the L time basis functions, determining L spatial basis functions corresponding to the L singular values by solving a spatial basis function optimization function; and generating a target magnetic resonance image of the target object based on the L time basis functions and the L spatial basis functions, wherein the spatial basis function optimization function comprises a first regularization term, and the first regularization term is related to a product of a spatial basis function corresponding to a first singular value of the L singular values and a time basis function.

Description

Technical Field

[0001] This specification relates to the field of medical imaging technology, and in particular to a magnetic resonance imaging method, system, device, and storage medium. Background Technology

[0002] Magnetic resonance imaging (MRI) is widely used in clinical practice due to its excellent soft tissue resolution, ability to perform multi-planar and multi-parameter imaging, and lack of radiation exposure to patients. In recent years, partially separable spatiotemporal imaging (SPEMI) has been proposed to accelerate imaging. It fully utilizes the low-rank characteristics of dynamic MRI images, transforming the reconstruction process into solving for the spatial and temporal bases separately. In traditional partially separable spatiotemporal imaging techniques, the spatial basis optimization functions typically lack regularization terms or only contain regularization terms that optimize the spatial basis functions, which can easily lead to a low signal-to-noise ratio in the reconstructed image.

[0003] Therefore, it is necessary to provide a magnetic resonance imaging method, system, apparatus, and storage medium to address the problem of poor imaging quality caused by regularization terms. Summary of the Invention

[0004] This specification provides one or more embodiments of a magnetic resonance imaging method, the method comprising: acquiring navigation signals and imaging signals collected in a magnetic resonance scan of a target object; processing the navigation signals using a singular value decomposition algorithm to obtain L time basis functions corresponding to L singular values, where L is a positive integer greater than 1; determining L spatial basis functions corresponding to the L singular values ​​by solving a spatial basis function optimization function based on the imaging signals and the L time basis functions; and generating a target magnetic resonance image of the target object based on the L time basis functions and the L spatial basis functions, wherein the spatial basis function optimization function includes a first regularization term, the first regularization term being related to the product of the spatial basis function and the time basis function corresponding to the first singular value among the L singular values.

[0005] In some embodiments, the first singular value is the largest of the M singular values ​​among the L singular values, where M is a positive integer less than L.

[0006] In some embodiments, the method for determining M includes: based on the imaging signal, determining L initial spatial basis functions corresponding to the L singular values ​​by solving a reference spatial basis function optimization function, wherein the reference spatial basis function includes a third regularization term, the third regularization term being related to the spatial basis functions corresponding to the L singular values; generating L initial spatial basis images based on the L initial spatial basis functions; and determining M by performing noise analysis on the L initial spatial basis images.

[0007] In some embodiments, the spatial basis function optimization function further includes a second regularization term, which is related to the spatial basis function corresponding to the second singular value among the L singular values.

[0008] In some embodiments, the second singular value is (LM) singular values ​​other than the first singular value among the L singular values.

[0009] In some embodiments, when the target object is the liver, L is 15 and M is 8.

[0010] In some embodiments, the first singular value and the second singular value overlap.

[0011] This specification provides one or more embodiments of a magnetic resonance imaging system, the system comprising: an acquisition module configured to acquire navigation signals and imaging signals acquired during magnetic resonance scanning of a target object; a processing module configured to process the navigation signals using a singular value decomposition algorithm to obtain L time basis functions corresponding to L singular values, where L is a positive integer greater than 1; an optimization module configured to determine L spatial basis functions corresponding to the L singular values ​​by solving a spatial basis function optimization function based on the imaging signals and the L time basis functions; and a generation module configured to generate a target magnetic resonance image of the target object based on the L time basis functions and the L spatial basis functions, wherein the spatial basis function optimization function includes a first regularization term, the first regularization term being related to the product of the spatial basis function and the time basis function corresponding to the first singular value among the L singular values.

[0012] This specification provides a magnetic resonance imaging apparatus according to one or more embodiments, the apparatus including at least one processor and at least one memory; the at least one memory is used to store computer instructions; the at least one processor is used to execute at least a portion of the computer instructions to implement a magnetic resonance imaging method according to any one of the above embodiments.

[0013] This specification provides one or more embodiments of a computer-readable storage medium that stores computer instructions. When a computer reads the computer instructions in the storage medium, the computer executes a magnetic resonance imaging method as described in any of the above embodiments.

[0014] According to the above scheme, by adding a regularization term that constrains the product of the time basis function and the space basis function, the signal-to-noise ratio and fidelity of the reconstructed target magnetic resonance image can be improved. Attached Figure Description

[0015] This specification will be further described by way of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting; in these embodiments, the same reference numerals denote the same structures, wherein:

[0016] Figure 1 These are schematic diagrams illustrating application scenarios of magnetic resonance imaging systems according to some embodiments of this specification;

[0017] Figure 2 These are exemplary block diagrams of a magnetic resonance imaging system according to some embodiments of this specification;

[0018] Figure 3 This is an exemplary flowchart of a magnetic resonance imaging method according to some embodiments of this specification;

[0019] Figure 4 This is a schematic diagram of an exemplary MRI pulse sequence according to some embodiments of this specification;

[0020] Figure 5 This is an exemplary flowchart illustrating the determination of M according to some embodiments of this specification;

[0021] Figure 6 This is a schematic diagram of the initial spatial basis image corresponding to different singular values ​​according to some embodiments of this specification;

[0022] Figure 7A This is a schematic diagram of an abdominal image obtained by constraining only the spatial basis functions, according to some embodiments of this specification.

[0023] Figure 7B This is a schematic diagram of an abdominal image obtained by simultaneously constraining the product of spatial basis functions and time basis functions, as shown in some embodiments of this specification. Detailed Implementation

[0024] To more clearly illustrate the technical solutions of the embodiments in this specification, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are merely some examples or embodiments of this specification. For those skilled in the art, these drawings can be applied to other similar scenarios without creative effort. Unless obvious from the context or otherwise specified, the same reference numerals in the drawings represent the same structures or operations.

[0025] It should be understood that the terms “system,” “device,” “unit,” and / or “module” used herein are one way to distinguish different components, elements, parts, sections, or assemblies at different levels. However, if other terms can achieve the same purpose, they may be replaced by other expressions.

[0026] As indicated in this specification and claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" do not specifically refer to the singular and may also include the plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of expressly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements.

[0027] Flowcharts are used in this specification to illustrate the operations performed by the system according to embodiments of this specification. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, the steps can be processed in reverse order or simultaneously. Furthermore, other operations can be added to these processes, or one or more steps can be removed from them.

[0028] Figure 1 These are schematic diagrams illustrating application scenarios of magnetic resonance imaging systems according to some embodiments of this specification. For example... Figure 1 As shown, the application scenario 100 of the magnetic resonance imaging system may include an MRI scanner 110, a processor 120, a network 130, a storage device 140, and a terminal 150.

[0029] MRI scanner 110 is configured to scan a target object (or a portion of a target object) to acquire image data, such as MRI signals (also known as MR signals) associated with the target object. For example, MRI scanner 110 can acquire MRI signals by applying a sequence of MRI pulses to the target object. The MRI pulse sequence can be any type of MRI pulse sequence, such as spin echo sequences, gradient echo sequences, diffusion sequences, inversion recovery sequences, etc. The target object can include biological objects (e.g., human bodies, animals, etc.), non-biological objects (e.g., phantoms), etc. In other embodiments, the target object may also include specific parts, organs, and / or tissues of a patient. For example, the target object can include the head, chest, legs, etc., or any combination thereof, without limitation.

[0030] Processor 120 can be used to process data and / or information obtained from MRI scanner 110, storage device 140, and / or terminal 150. For example, processor 120 can acquire navigation signals and imaging signals acquired in a magnetic resonance scan of a target object from MRI scanner 110. As another example, processor 120 can process the navigation signals using a singular value decomposition algorithm to obtain L time basis functions corresponding to L singular values. Yet another example, processor 120 can determine L spatial basis functions corresponding to L singular values ​​by solving a spatial basis function optimization function based on the imaging signal and L time basis functions; and generate a target magnetic resonance image of the target object based on the L time basis functions and L spatial basis functions.

[0031] In some embodiments, the processor 120 may send the processing result to the terminal 150. For example, the processor 120 may send the target magnetic resonance image to the terminal 150 and display it on one or more display devices in the terminal 150.

[0032] In some embodiments, processor 120 may be a single server or a group of servers. The server group may be centralized or distributed. In some embodiments, processor 120 may be local or remote. In some embodiments, processor 120 may access information and / or data stored thereon via network 130 or direct connection to MRI scanner 110, storage device 140, and / or terminal 150. In some embodiments, processor 120 may be integrated into MRI scanner 110. In some embodiments, processor 120 may be implemented on a cloud platform. By way of example only, the cloud platform may include private cloud, public cloud, hybrid cloud, community cloud, distributed cloud, internal cloud, multi-cloud, etc., or any combination thereof.

[0033] Network 130 may include any suitable network capable of facilitating information and / or data exchange within the magnetic resonance imaging system. In some embodiments, one or more components of the magnetic resonance imaging system (e.g., MRI scanner 110, processor 120, storage device 140, or terminal 150) may connect to and / or communicate with other components of the magnetic resonance imaging system via network 130. For example, processor 120 may acquire navigation and imaging signals from MRI scanner 110 via network 130. In some embodiments, network 130 may be any one or more of a wired or wireless network.

[0034] Storage device 140 may store data, instructions, and / or any other information. In some embodiments, storage device 140 may store data obtained from MRI scanner 110, processor 120, and / or terminal 150. In some embodiments, storage device 140 may store data and / or instructions used by processor 120 to perform or use the exemplary methods described herein.

[0035] In some embodiments, storage device 140 may include mass storage, removable storage, volatile read-write storage, read-only storage (ROM), or any combination thereof. In some embodiments, storage device 140 may be implemented on a cloud platform.

[0036] In some embodiments, storage device 140 may be connected to network 130 to communicate with one or more components of a magnetic resonance imaging system (e.g., MRI scanner 110, processor 120, terminal 150, etc.). One or more components of the magnetic resonance imaging system may access data or instructions stored in storage device 140 via network 130. In some embodiments, storage device 140 may be part of processor 120.

[0037] Terminal 150 can communicate and / or connect to MRI scanner 110, processor 120, and / or storage device 140. For example, terminal 150 can send one or more control commands to MRI scanner 110 via network 130 to control MRI scanner 110 to scan a target object according to the commands. In some embodiments, terminal 150 may include one or any combination of mobile device 150-1, tablet computer 150-2, laptop computer 150-3, or other devices with input and / or output functions. In some embodiments, terminal 150 can remotely operate MRI scanner 110. In some embodiments, terminal 150 can operate MRI scanner 110 via wireless connection. In some embodiments, terminal 150 can receive information and / or instructions input by a user and send the received information and / or instructions to MRI scanner 110 or processor 120 via network 130. In some embodiments, terminal 150 can receive data and / or information from processor 120. In some embodiments, terminal 150 may be part of processor 120. In some embodiments, terminal 150 may be integrated with processor 120 as an operating table for MRI scanner 110. In some embodiments, terminal 150 may be omitted.

[0038] It should be noted that the application scenario 100 of the magnetic resonance imaging system is provided for illustrative purposes only and is not intended to limit the scope of this specification. Those skilled in the art can make various modifications or variations based on the description in this specification. For example, the application scenario 100 of the magnetic resonance imaging system can be implemented on other devices to achieve similar or different functions. However, such variations and modifications will not depart from the scope of this specification.

[0039] Figure 2 This is an exemplary block diagram of a magnetic resonance imaging system according to some embodiments of this specification. In some embodiments, the magnetic resonance imaging system 200 may include an acquisition module 210, a processing module 220, an optimization module 230, and a generation module 240. In some embodiments, the magnetic resonance imaging system 200 may be composed of... Figure 1 The processor 120 shown is implemented.

[0040] The acquisition module 210 is configured to acquire navigation and imaging signals acquired during magnetic resonance scanning of the target object.

[0041] The processing module 220 is configured to process the navigation signal using the singular value decomposition algorithm to obtain L time basis functions corresponding to L singular values, where L is a positive integer greater than 1.

[0042] The optimization module 230 is configured to determine L spatial basis functions corresponding to L singular values ​​by solving a spatial basis function optimization function based on the imaging signal and L temporal basis functions. In some embodiments, the spatial basis function optimization function includes a first regularization term, which is related to the product of the spatial basis function and the temporal basis function corresponding to the first singular value among the L singular values. In some embodiments, the first singular values ​​are the M largest singular values ​​among the L singular values, where M is a positive integer less than L. In some embodiments, the spatial basis function optimization function further includes a second regularization term, which is related to the spatial basis function corresponding to the second singular value among the L singular values. In some embodiments, the second singular values ​​are the (LM) singular values ​​among the L singular values ​​excluding the first singular value. In some embodiments, the first and second singular values ​​overlap.

[0043] In some embodiments, the optimization module 230 is further configured to determine L initial spatial basis functions corresponding to L singular values ​​based on the imaging signal by solving the reference spatial basis function optimization function, wherein the reference spatial basis function includes a third regularization term, which is related to the spatial basis functions corresponding to the L singular values; generate L initial spatial basis images based on the L initial spatial basis functions; and determine M by performing noise analysis on the L initial spatial basis images.

[0044] The generation module 240 is configured to generate a target magnetic resonance image of the target object based on L time basis functions and L spatial basis functions.

[0045] For more information on the acquisition, processing, and optimization modules, please refer to [link / reference]. Figure 3 And its related description. For more information about the generation module, please refer to... Figure 3 and Figure 5 And its related descriptions.

[0046] It should be noted that the above description of the magnetic resonance imaging system 200 and its modules is for convenience only and should not be construed as limiting this specification to the scope of the illustrated embodiments. It is understood that those skilled in the art, after understanding the principles of the system, may arbitrarily combine the various modules or construct subsystems connected to other modules without departing from these principles. In some embodiments, Figure 2 The acquisition module 210, processing module 220, optimization module 230, and generation module 240 disclosed herein can be different modules within a single system, or a single module can implement the functions of two or more of the aforementioned modules. For example, the modules can share a single storage module, or each module can have its own separate storage module. Such variations are all within the scope of protection of this specification.

[0047] Figure 3This is an exemplary flowchart of a magnetic resonance imaging method according to some embodiments of this specification. In some embodiments, process 300 may be executed by processor 120 or magnetic resonance imaging system 200. For example, process 300 may be stored in storage device 140 in the form of a program or instructions, and process 300 may be implemented when processor 120 or magnetic resonance imaging system 200 executes the instructions. The operational schematic diagram of process 300 presented below is illustrative. In some embodiments, the process may be accomplished using one or more additional operations not described and / or one or more operations not discussed. Additionally, Figure 3 The order of operations shown in the diagram and described below in process 300 is not restrictive.

[0048] Step 310: Acquire navigation signals and imaging signals obtained during the magnetic resonance scan of the target object. In some embodiments, step 310 may be performed by the acquisition module 210.

[0049] For more information about the target object, please refer to [link / reference]. Figure 1 And its related descriptions.

[0050] Navigation signals, also known as auxiliary signals, comprise high temporal resolution data associated with at least one temporal dimension of the target object. The at least one temporal dimension can include any dimension reflecting characteristics or dynamic information of the target object changing over time. In some embodiments, the at least one temporal dimension of the target object can include a dimension associated with elapsed time. In some embodiments, the at least one temporal dimension of the scanned object can include one or more dimensions associated with other information, such as cardiac motion, respiratory motion, T1 relaxation, T2 relaxation, chemical exchange saturation transfer (CEST), contrast agent dynamics, T1 / P contrast, molecular diffusion, etc.

[0051] In some embodiments, the navigation signal may correspond to the same subset of the K-space (e.g., the K-space includes one or more K-space lines), which can be obtained by repeatedly sampling the subset of the K-space at a high sampling frequency. For example, the navigation signal may correspond to the same K-space line in the K-space, and the navigation signal can be obtained by repeatedly sampling the K-space line using a high sampling frequency. As used herein, a high sampling frequency refers to a sampling frequency higher than a threshold frequency. The threshold frequency may be a default value, manually determined by the user, or determined by the processor 120 through data analysis. For example, the threshold frequency may be determined based on at least one time-varying dimension to be analyzed. As an example only, if the time-varying dimension is related to the respiratory motion of the target object, and the respiratory cycle of the target object is close to 0.75 seconds, the sampling frequency may need to be a threshold frequency greater than 1 / 0.75 Hz in order to capture dynamic information related to the respiratory motion of the target object. As another example, the threshold frequency may be determined based on practical requirements (e.g., accuracy requirements), experience, data models, etc.

[0052] The imaging signal includes high spatial resolution data relating to at least one spatial variation dimension of the target object. Exemplary spatial variation dimensions include phase-encoded directions, frequency-encoded directions, etc., or any combination thereof. In some embodiments, the imaging signal can be acquired by sampling different K-space lines in K-space using a pseudo-random trajectory acquisition method.

[0053] In some embodiments, the processor can acquire navigation and imaging signals using any suitable sampling mode. For illustrative purposes, the acquisition methods for navigation and imaging signals are described below. In some embodiments, the processor can acquire navigation and imaging signals via radial sampling. The navigation signal may correspond to a radial line in K-space with a constant angle (e.g., 0°, 10°, 20°, 30°, 100°, 180°, etc.). By way of example only, the processor can repeatedly sample radial lines with constant angles in K-space at fixed time intervals to acquire the navigation signal. The imaging signal may correspond to at least two radial lines in K-space with different readout angles. In some embodiments, the processor can acquire the imaging signal by sampling at least two radial lines in K-space according to a golden angle radial sampling mechanism. By employing a golden angle radial sampling mechanism, multiple radial lines uniformly distributed in and covering K-space can be obtained in a relatively short time, improving scanning efficiency and reducing computational load and time. It should be understood that any other readout angle (e.g., a randomly set readout angle) can be used to sample the imaging signal according to actual needs (e.g., based on requirements regarding scan time and / or imaging quality).

[0054] In some embodiments, the processor can acquire navigation and imaging signals via Cartesian sampling. The navigation signal may correspond to the same Cartesian line in K-space. The imaging signal may correspond to different Cartesian lines in K-space. In some embodiments, the navigation signal may correspond to a Cartesian line passing through the center of K-space. In some embodiments, the imaging signal can be obtained via Cartesian sampling, while the signal can be obtained by repeatedly sampling a specific radial or helical line in K-space.

[0055] During a magnetic resonance imaging (MRI) scan of the target object, the processor can acquire navigation and imaging signals in any sampling order. In some embodiments, navigation and imaging signals can be acquired alternately during the MRI scan of the target object. For example, a first number of imaging signals can be acquired after or before each readout of a second number of navigation signals. The first and second numbers can be any positive integers, such as 1, 2, 3, 5, 10, etc. In some embodiments, the first and second numbers can be set according to actual needs, for example, the sampling frequency of the navigation signals needs to be greater than a threshold frequency and / or sufficient navigation signals need to be acquired for image reconstruction. In some embodiments, the ratio of the first to the second number can be related to the type of target object to be imaged. For example, for imaging a patient's heart, the ratio of the first to the second number can be equal to 1:1. As another example, for imaging organs other than the heart (e.g., the arm, knee), the ratio of the first to the second number can be equal to 10:1. In some embodiments, the navigation signal may not have phase encoding. The navigation signal can be used for time basis function calculation. The phase encoding of the imaging signal can conform to a certain pattern (e.g., a random Gaussian distribution).

[0056] In some embodiments, any type of MRI pulse sequence can be applied during a magnetic resonance scan to acquire navigation and imaging signals. This is merely an example. Figure 4 This is a schematic diagram of an exemplary MRI pulse sequence according to some embodiments of this specification. Figure 4 As shown, the processor can acquire four imaging signals after reading one navigation signal. In this case, the ratio of the number of imaging signals acquired in step 310 to the number of navigation signals can be 4:1. The navigation signals have no phase encoding and repeat periodically. The phase encoding of the imaging signals conforms to a random Gaussian distribution.

[0057] Step 320: Process the navigation signal using the singular value decomposition algorithm to obtain L time basis functions corresponding to L singular values, where L is a positive integer greater than 1. In some embodiments, step 320 can be executed by processing module 220.

[0058] Singular values ​​are the elements on the diagonal of a diagonal matrix obtained through singular value decomposition (SVD).

[0059] The time basis functions include one or more time basis functions related to natural time. In some embodiments, each time basis function may be related to the time variation dimension of the target object. In some embodiments, the time basis functions may include one or more cardiac time basis functions related to the cardiac motion of the target object, one or more respiratory time basis functions related to the respiratory motion of the target object, one or more T1 recovery time basis functions related to the T1 relaxation of the target object, etc., or any combination thereof. Time basis functions related to the time variation dimension can reflect dynamic information along the time variation dimension and include high temporal resolution information.

[0060] In some embodiments, the processor can fill the navigation signal into a K-space to obtain a K-space matrix, and then perform singular value decomposition on the K-space matrix to obtain L time basis functions corresponding to L singular values. As an example only, the K-space matrix can be represented as... In this context, an element in the K-space matrix represents the K-space data acquired by a specific coil channel at a given moment. For example, κ(k c ,t1) indicates that at time t1, the coil channel k c The collected K-space data. The processor can perform SVD on the K-space matrix to obtain the singular value matrix and determine the L time basis functions corresponding to the L singular values.

[0061] Step 330: Based on the imaging signal and L time basis functions, determine the L spatial basis functions corresponding to the L singular values ​​by solving the spatial basis function optimization function. In some embodiments, step 330 can be performed by the optimization module 230.

[0062] Spatial basis functions include high spatial resolution information along spatially varying dimensions. For example, spatial basis functions can reflect the relationship between pixel information in an image and spatial information in the physical domain. In some embodiments, spatial basis functions can be represented as basis images that include high spatial resolution information. Different spatial basis functions can be represented as basis images that include different levels of high spatial resolution information.

[0063] The spatial basis function optimization function refers to the optimization equation used to solve the spatial basis functions.

[0064] In traditional partially separable spatiotemporal imaging techniques, the processor can determine the L spatial basis functions corresponding to the L singular values ​​based on the imaging signal and L temporal basis functions, according to the spatial basis function optimization function shown in formula (1): in, This represents the optimal spatial basis functions (including the spatial basis functions corresponding to each singular value) determined by solving formula (1); d img This represents the K-space data obtained after the imaging signal is filled into the K-space; F represents the Fourier transform operator; S represents the coil sensitivity map; Ω represents the downsampling operator (which can be omitted in some cases); i represents the index of each singular value after sorting from largest to smallest; U i Φ represents the spatial basis function corresponding to the i-th singular value; i Let L represent the time basis function corresponding to the i-th singular value; L represents the number of singular values.

[0065] To improve the accuracy of spatial basis functions, a regularization term can be added to the spatial basis optimization function shown in formula (1). A regularization term refers to a restriction or constraint added to the solution during the solution-solving process of the spatial basis function optimization function, used to make the obtained solution more suitable for the requirements. For example, a regularization term may include the total variation operator, wavelet transform operator, etc. In existing technologies, spatial basis function optimization functions typically lack a regularization term or only contain a regularization term for optimizing the spatial basis functions, which can easily lead to a low signal-to-noise ratio in the reconstructed image.

[0066] To address the problems existing in the prior art, this specification proposes a new method for constructing regularization terms.

[0067] In some embodiments, the spatial basis optimization function includes a first regularization term, which is related to the product of the spatial basis function and the temporal basis function corresponding to the first singular value among L singular values. The product of the spatial basis function and the temporal basis function corresponding to the singular value is the image domain data corresponding to that singular value. That is, the first regularization term is used to constrain the image domain data corresponding to the first singular value. By setting the first regularization term to constrain the image domain data corresponding to the first singular value, the image fidelity and signal-to-noise ratio can be improved.

[0068] In some embodiments, the processor can determine any singular value from L singular values ​​as the first singular value. In some embodiments, the first singular value is the M largest singular values ​​among the L singular values, where M is a positive integer less than L. For example, the processor can sort the L singular values ​​in descending order and determine the first M singular values ​​as the first singular value. Here, M can be an empirical value, a pre-set value, etc.

[0069] In some embodiments, the spatial basis function optimization function further includes a second regularization term. The second regularization term relates to the spatial basis function corresponding to the second singular value among the L singular values. That is, the second regularization term is used to constrain the spatial basis function corresponding to the second singular value. For example, the processor can determine the L spatial basis functions corresponding to the L singular values ​​according to the spatial basis function optimization function shown in formula (2): Where R1 represents the first regularization term; λ1 represents the weight of the first regularization term; R2 represents the second regularization term; λ2 represents the weight of the second regularization term; Ω t Let R1 represent the time selection operator; the a-th to b-th singular values ​​are the first singular values; the c-th to d-th singular values ​​are the second singular values; a, b, c, and d are all positive integers greater than 0 and less than or equal to L, b is greater than or equal to a, and d is greater than or equal to c. In formula (2), the first regularization term R1 is related to the product of the time basis function and the space basis function corresponding to the a-th to b-th first singular values, and the second regularization term R2 is related to the space basis function corresponding to the c-th to d-th second singular values.

[0070] In some embodiments, the time selection operator Ω t It is possible to select some or all of the time points from a series of time points in the time direction. For example, the time selection operator Ω t Selecting a subset of time points from a series of time points along the time axis helps reduce the amount of computation and shorten the computation time.

[0071] In some embodiments, the processor can determine any singular value among the L singular values ​​as the second singular value. In some embodiments, the second singular value is the (LM) singular values ​​among the L singular values ​​excluding the first singular value. For example, the processor can sort the L singular values ​​in descending order, determine the first M singular values ​​as the first singular value, and determine the remaining (LM) singular values ​​as the second singular value. As an example only, the processor can determine the L spatial basis functions corresponding to the L singular values ​​according to the spatial basis function optimization function shown in formula (3): The first to the Mth singular values ​​are the first singular values; the (M+1)th to the Lth singular values ​​are the second singular values. In some embodiments, the processor can determine L initial spatial basis functions corresponding to L singular values ​​based on the imaging signal by solving a reference spatial basis function optimization function, and generate L initial spatial basis images; by performing noise analysis on the L initial spatial basis images, M is determined. More details regarding the foregoing embodiments can be found in [link to relevant documentation]. Figure 5 And its related descriptions.

[0072] In some embodiments of the present specification, by applying a regularization term purely in the image domain or a regularization term hybridized by spatial basis function and image domain, the image quality can be guaranteed while improving the fidelity of image details. By way of example only, Figure 7A is a schematic diagram of an abdominal image obtained by constraining only spatial basis functions according to some embodiments of the present specification. Figure 7B is a schematic diagram of an abdominal image obtained by constraining spatial basis functions and the product of spatial basis functions and temporal basis functions simultaneously according to some embodiments of the present specification. As shown in Figure 7A , in the abdominal image reconstructed by the processor according to the spatial basis function optimization function that only constrains the spatial basis functions, vascular artifacts appear in the framed part. As shown in Figure 7B , in the abdominal image reconstructed by the processor according to the spatial basis function optimization function shown in the aforementioned formula (3) (which constrains the spatial basis functions and the product of the spatial basis functions and the temporal basis functions at the same time), no vascular artifacts appear in the same part, which indicates that applying the regularization term hybridized by spatial basis function and image domain can effectively remove image artifacts and improve image quality.

[0073] In some embodiments, there is an overlap between the first singular value and the second singular value. That is to say, the first singular value and the second singular value may include partially identical singular values. By way of example only, when the spatial basis optimization function is the aforementioned formula (2), c<b<d, or a<d<b.

[0074] In some embodiments, when the target object is the liver, L is 15 and M is 8. That is, when the target object is the liver, the number of singular values is 15, and the first singular values are the first 8 largest singular values.

[0075] In step 340, a target magnetic resonance image of the target object is generated based on the L temporal basis functions and the L spatial basis functions. In some embodiments, step 340 may be performed by the generation module 240.

[0076] The target magnetic resonance image refers to an image related to the target object (for example, the chest).

[0077] In some embodiments, the target magnetic resonance image may be a dynamic image, which is used to reflect the change of the target object over time. In some embodiments, the target magnetic resonance image may also be a static image.

[0078] In some embodiments, the processor may add the L temporal basis functions to obtain a target temporal basis function; add the L spatial basis functions to obtain a target spatial basis function; then the processor multiplies the target temporal basis function and the target spatial basis function to obtain the target magnetic resonance image.

[0079] In some embodiments of this specification, by adding a first regularization term that constrains the product of the time basis function and the space basis function, and combining it with a second regularization term that constrains the space basis function, the signal-to-noise ratio and fidelity of the reconstructed target magnetic resonance image can be improved.

[0080] Figure 5 This is an exemplary flowchart illustrating the determination of M according to some embodiments of this specification. As described above, in some embodiments, the largest M singular values ​​among L singular values ​​are the first singular values. Flowchart 500 can be used to determine the size of M, i.e., the number of the first singular values.

[0081] Step 510: Based on the imaging signal, determine the L initial spatial basis functions corresponding to the L singular values ​​by solving the reference spatial basis function optimization function.

[0082] The reference spatial basis function optimization function refers to the optimization equation used to solve the initial spatial basis function. The reference spatial basis function optimization function does not contain a regularization term (i.e., the first regularization term) related to the image domain data. For example, the reference spatial basis function optimization function can be the spatial basis function optimization function used in traditional partially separable spatiotemporal imaging techniques, as shown in Equation (1). The spatial basis function obtained by solving Equation (1) can be used as the initial spatial basis function.

[0083] In some embodiments, the reference space basis function optimization function includes a third regularization term. The third regularization term relates to the space basis functions corresponding to the L singular values. That is, the third regularization term is used to constrain the space basis functions corresponding to the L singular values. For example, the processor can solve the reference space basis function optimization function shown in equation (4) and use the solved L space basis functions corresponding to the L singular values ​​as the L initial space basis functions corresponding to the L singular values: Where R3 represents the third regularization term; λ3 represents the weight of the third regularization term.

[0084] Step 520: Generate L initial spatial basis images based on L initial spatial basis functions.

[0085] In some embodiments, the processor may generate an initial spatial basis image based on the relationship between pixel information in the image and spatial information in the physical domain, as reflected by the initial spatial basis function.

[0086] Step 530: Determine M by performing noise analysis on L initial spatial basis images.

[0087] In some embodiments, the processor can determine whether the noise of the initial spatial basis image corresponding to each singular value is less than a noise threshold, thereby selecting an initial spatial basis image that is less than the noise threshold. The processor can determine the index of the singular value corresponding to the selected initial spatial basis image and set the maximum index as M. The noise threshold can be a default value, a pre-set value, an empirical value, etc., as an example only. Figure 6 These are schematic diagrams of the initial spatial basis images corresponding to different singular values, as shown in some embodiments of this specification, such as... Figure 6 As shown, when the singular value is greater than 8, the initial spatial basis image contains a large amount of noise. Therefore, the processor can determine the singular value index 8 as M. It should be understood that when noise analysis reveals that the noise of the initial spatial basis image corresponding to a singular value index exceeding M exceeds a threshold, it indicates that the accuracy of these singular values ​​and the corresponding spatial basis functions obtained based on the reference spatial basis function optimization function is low. It is necessary to recalculate the singular value using the image domain regularization term or the spatial basis function-image domain hybrid regularization term proposed in this specification, thereby improving the accuracy of the spatial basis function.

[0088] In some embodiments of this specification, by performing noise analysis on the initial spatial basis image, the value of M can be made more accurate, thereby making the subsequent reconstruction of the target magnetic resonance image more accurate and improving the quality of the obtained target magnetic resonance image.

[0089] It should be noted that the above descriptions of processes 300 and 500 are for illustrative purposes only and do not limit the scope of this specification. Those skilled in the art can make various modifications and changes to processes 300 and 500 under the guidance of this specification. However, these modifications and changes remain within the scope of this specification.

[0090] This specification provides a magnetic resonance imaging apparatus according to one or more embodiments, the apparatus including at least one processor and at least one memory; the at least one memory is used to store computer instructions; the at least one processor is used to execute at least a portion of the computer instructions to implement a magnetic resonance imaging method according to any one of the above embodiments.

[0091] This specification provides one or more embodiments of a computer-readable storage medium that stores computer instructions. When a computer reads the computer instructions in the storage medium, the computer executes a magnetic resonance imaging method as described in any of the above embodiments.

[0092] The basic concepts have been described above. Obviously, for those skilled in the art, the detailed disclosure above is merely illustrative and does not constitute a limitation of this specification. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements, and corrections to this specification. Such modifications, improvements, and corrections are suggested in this specification and therefore remain within the spirit and scope of the exemplary embodiments described herein.

[0093] Furthermore, this specification uses specific terms to describe embodiments thereof. For example, "an embodiment," "one embodiment," and / or "some embodiments" refer to a particular feature, structure, or characteristic associated with at least one embodiment of this specification. Therefore, it should be emphasized and noted that references to "an embodiment," "one embodiment," or "an alternative embodiment" in different locations throughout this specification do not necessarily refer to the same embodiment. Moreover, certain features, structures, or characteristics in one or more embodiments of this specification can be appropriately combined.

[0094] Furthermore, unless expressly stated in the claims, the order of processing elements and sequences, the use of numbers and letters, or other names described in this specification are not intended to limit the order of the processes and methods described herein. Although various examples have been discussed in the foregoing disclosure of some embodiments of the invention that are currently considered useful, it should be understood that such details are for illustrative purposes only, and the appended claims are not limited to the disclosed embodiments; rather, the claims are intended to cover all modifications and equivalent combinations that conform to the spirit and scope of the embodiments described herein. For example, while the system components described above can be implemented using hardware devices, they can also be implemented solely using software solutions, such as installing the described system on existing servers or mobile devices.

[0095] Similarly, it should be noted that, in order to simplify the description disclosed herein and thus aid in the understanding of one or more embodiments of the invention, the foregoing description of embodiments in this specification may sometimes combine multiple features into a single embodiment, drawing, or description thereof. However, this method of disclosure does not imply that the subject matter of this specification requires more features than those mentioned in the claims. In fact, the embodiments contain fewer features than all the features of a single embodiment disclosed above.

[0096] In some embodiments, numbers describing the quantity of components and attributes are used. It should be understood that such numbers used in the description of embodiments are modified in some examples with the terms "approximately," "approximately," or "generally." Unless otherwise stated, "approximately," "approximately," or "generally" indicates that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may be changed depending on the characteristics required by individual embodiments. In some embodiments, numerical parameters should take into account specified significant digits and employ a general method of digit reservation. Although the numerical ranges and parameters used to confirm their breadth of range in some embodiments of this specification are approximate values, in specific embodiments, such values ​​are set as precisely as feasible.

[0097] For each patent, patent application, patent application publication, and other material, such as articles, books, specifications, publications, and documents, referenced in this specification, the entire contents of which are incorporated herein by reference. This excludes historical application documents that are inconsistent with or conflict with the content of this specification, as well as documents that limit the broadest scope of the claims in this specification (currently or subsequently appended to this specification). It should be noted that in the event of any inconsistency or conflict between the descriptions, definitions, and / or terminology used in the supplementary materials to this specification and the content of this specification, the descriptions, definitions, and / or terminology used in this specification shall prevail.

[0098] Finally, it should be understood that the embodiments described in this specification are merely illustrative of the principles of the embodiments described herein. Other variations may also fall within the scope of this specification. Therefore, alternative configurations of the embodiments described herein are intended to be illustrative rather than limiting, and should be considered consistent with the teachings of this specification. Accordingly, the embodiments described herein are not limited to those explicitly introduced and described herein.

Claims

1. A magnetic resonance imaging method, characterized in that, The method includes: Acquire navigation and imaging signals from magnetic resonance scans of the target object; The navigation signal is processed using the singular value decomposition algorithm to obtain L time basis functions corresponding to L singular values, where L is a positive integer greater than 1; Based on the imaging signal and the L time basis functions, the L spatial basis functions corresponding to the L singular values ​​are determined by solving the spatial basis function optimization function; Based on the L time basis functions and the L spatial basis functions, a target magnetic resonance image of the target object is generated. The spatial basis function optimization function includes a first regularization term, which is related to the product of the spatial basis function and the time basis function corresponding to the first singular value among the L singular values.

2. The method according to claim 1, characterized in that, The first singular value is the largest M singular value among the L singular values, where M is a positive integer less than L.

3. The method according to claim 2, characterized in that, The spatial basis function optimization function further includes a second regularization term, which is related to the spatial basis function corresponding to the second singular value among the L singular values.

4. The method according to claim 3, characterized in that, The second singular value is (LM) singular values ​​other than the first singular value among the L singular values.

5. The method according to claim 4, characterized in that, When the target object is the liver, L is 15 and M is 8.

6. The method according to claim 3, characterized in that, The first singular value and the second singular value overlap.

7. The method according to claim 2, characterized in that, The method for determining M includes: Based on the imaging signal, by solving the reference space basis function optimization function, L initial space basis functions corresponding to the L singular values ​​are determined. The reference space basis function includes a third regularization term, which is related to the space basis functions corresponding to the L singular values. Based on the L initial spatial basis functions, generate L initial spatial basis images; The M is determined by performing noise analysis on the L initial spatial basis images.

8. A magnetic resonance imaging system, characterized in that, The system includes: The acquisition module is configured to acquire navigation and imaging signals obtained during magnetic resonance scanning of the target object; The processing module is configured to process the navigation signal using a singular value decomposition algorithm to obtain L time basis functions corresponding to L singular values, where L is a positive integer greater than 1; The optimization module is configured to determine the L spatial basis functions corresponding to the L singular values ​​by solving the spatial basis function optimization function based on the imaging signal and the L time basis functions; The generation module is configured to generate a target magnetic resonance image of the target object based on the L time basis functions and the L spatial basis functions, wherein the spatial basis function optimization function includes a first regularization term, which is related to the product of the spatial basis function and the time basis function corresponding to the first singular value among the L singular values.

9. A magnetic resonance imaging device, characterized in that, The device includes at least one processor and at least one memory; The at least one memory is used to store computer instructions; The at least one processor is configured to execute at least a portion of the computer instructions to implement a magnetic resonance imaging method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The storage medium stores computer instructions. When the computer reads the computer instructions in the storage medium, the computer executes a magnetic resonance imaging method as described in any one of claims 1 to 7.

Images