Calibration-free reconstruction methods, systems, media, and electronic devices for dynamic magnetic resonance imaging.

Abstract

The present invention provides a method, system, medium, and electronic device for non-calibration reconstruction of dynamic magnetic resonance images, including: constructing a first implicit neural expression and a second implicit neural expression; inputting the three-dimensional spatiotemporal coordinates of the dynamic magnetic resonance image into the first implicit neural expression and the second implicit neural expression to obtain a reconstructed dynamic magnetic resonance image sequence and a coil sensitivity map sequence; calculating a total loss function based on the reconstructed dynamic magnetic resonance image sequence and the coil sensitivity map sequence; minimizing the total loss function using a gradient back propagation algorithm to obtain a trained first implicit neural expression and a second implicit neural expression; and reconstructing the dynamic magnetic resonance image based on the trained first implicit neural expression. The method, system, medium, and electronic device for non-calibration reconstruction of dynamic magnetic resonance images of the present invention realize high spatiotemporal resolution dynamic magnetic resonance imaging reconstruction based on implicit neural expressions, and do not require a training data set or additional coil sensitivity calibration.

Description

Technical Field

[0001] The present invention relates to the technical field of image calibration, and in particular to a method, system, medium and electronic equipment for non-calibration reconstruction of dynamic magnetic resonance images. Background Art

[0002] Dynamic MRI is a very important component of MRI technology. It can preserve the dynamic changes of human tissues while taking advantage of the good tissue contrast of MRI itself, and has a wide range of applications in moving tissues such as the heart and abdomen. However, due to the disadvantage of long imaging time of MRI itself, existing dynamic MRI technology often cannot achieve high temporal and spatial resolution at the same time, and requires highly downsampling the MRI data acquisition process and using corresponding reconstruction technology to restore images with high temporal and spatial resolution.

[0003] Although dynamic MRI reconstruction techniques based on compressed sensing and deep learning have been widely used in clinical and scientific research, these methods each have a series of defects. Compressed sensing-based technology often has unsatisfactory reconstruction effects when faced with high downsampling acceleration rates. Deep learning-based technology requires a large number of high-quality data sets for neural network training, and its stability and generalization have not yet been fully clinically verified. At the same time, the above methods all require coil sensitivity information as a priori, which often needs to be obtained from additional reference scans or fully sampled automatic calibration data, increasing the scanning time and complexity of the scanning process. The acquired sensitivity prior is difficult to estimate accurately under the high downsampling acceleration rates and moving imaging targets commonly seen in dynamic MRI, which in turn brings difficulties to the reconstruction of higher quality high temporal and spatial resolution images. Summary of the invention

[0004] In view of the above-mentioned problems, the purpose of the present invention is to provide a method, system, medium, and electronic device for calibration-free reconstruction of dynamic magnetic resonance images, which can realize high temporal and spatial resolution dynamic magnetic resonance imaging reconstruction based on implicit neural expression without the need for training data sets or additional coil sensitivity calibration.

[0005] In a first aspect, the present invention provides a method for calibration-free reconstruction of dynamic magnetic resonance images, the method comprising the following steps: constructing a first implicit neural expression and a second implicit neural expression; inputting the three-dimensional spatiotemporal coordinates of a dynamic magnetic resonance image into the first implicit neural expression to obtain a reconstructed dynamic magnetic resonance image sequence; inputting the three-dimensional spatiotemporal coordinates of the dynamic magnetic resonance image into the second implicit neural expression to obtain a coil sensitivity map sequence; calculating a total loss function of the first implicit neural expression and the second implicit neural expression based on the reconstructed dynamic magnetic resonance image sequence and the coil sensitivity map sequence; minimizing the total loss function using a gradient back-propagation algorithm to obtain trained first implicit neural expressions and second implicit neural expressions; and reconstructing a dynamic magnetic resonance image based on the trained first implicit neural expression.

[0006] In an implementation of the first aspect, the first implicit neural expression includes a first hash table and a first multi-layer perceptron connected to each other; the second implicit neural expression includes a second hash table and a second multi-layer perceptron connected to each other.

[0007] In an implementation of the first aspect, according to The reconstructed dynamic magnetic resonance image sequence is acquired, wherein represents the i-th reconstructed dynamic magnetic resonance image, represents the first implicit neural expression, (1,1,1) to (N,N,T) represent three-dimensional space-time coordinates, N represents the row / column size of the image matrix, and T represents the total number of dynamic magnetic resonance image frames.

[0008] In an implementation of the first aspect, according to Obtain the coil sensitivity map sequence, where represents the coil sensitivity diagram for the cth coil, represents the second implicit neural expression, (1,1,1) to (N,N,T) represent three-dimensional space-time coordinates, N represents the row / column size of the image matrix, and T represents the total number of dynamic magnetic resonance image frames.

[0009] In an implementation of the first aspect, calculating the total loss function of the first implicit neural expression and the second implicit neural expression based on the reconstructed dynamic magnetic resonance image sequence and the coil sensitivity map sequence comprises the following steps:

[0010] Calculate data consistency loss Total variational loss term and the regularization loss term

[0011] Calculate the total loss function where λ S With λ L is the proportional adjustment parameter.

[0012] In an implementation of the first aspect, according to Calculate the data consistency loss term, where C represents the number of coils, ⊙ represents the matrix element-by-element multiplication; F u represents the downsampled Fourier transform, represents the coil sensitivity diagram for the cth coil, represents the i-th reconstructed dynamic magnetic resonance image, y c represents the k-space data acquired from the c-th coil, and ∥…∥2 represents the 2-norm.

[0013] In an implementation of the first aspect, according to Calculate the total variational loss term according to Calculate the regularization loss term, where ∥…∥1 and ∥…∥ * Respectively represent the 1-norm and the nuclear norm, TV t (...) represents the total variation operator along the time dimension, represents the i-th reconstructed dynamic magnetic resonance image.

[0014] In a second aspect, the present invention provides a calibration-free reconstruction system for dynamic magnetic resonance images, the system comprising a construction module, a first acquisition module, a second acquisition module, a calculation module, a training module and a reconstruction module;

[0015] The building module is used to build a first implicit neural expression and a second implicit neural expression;

[0016] The first acquisition module is used to input the three-dimensional spatiotemporal coordinates of the dynamic magnetic resonance image into the first implicit neural expression to acquire and reconstruct a dynamic magnetic resonance image sequence;

[0017] The second acquisition module is used to input the three-dimensional spatiotemporal coordinates of the dynamic magnetic resonance image into the second implicit neural expression to acquire a coil sensitivity map sequence;

[0018] The calculation module is used to calculate the total loss function of the first implicit neural expression and the second implicit neural expression based on the reconstructed dynamic magnetic resonance image sequence and the coil sensitivity map sequence;

[0019] The training module is used to minimize the total loss function by using a gradient back propagation algorithm to obtain the trained first implicit neural expression and the second implicit neural expression;

[0020] The reconstruction module is used to realize the reconstruction of dynamic magnetic resonance images based on the trained first implicit neural expression.

[0021] In a third aspect, the present invention provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned method for reconstructing dynamic magnetic resonance images without calibration.

[0022] In a fourth aspect, the present invention provides an electronic device, comprising: a processor and a memory;

[0023] The memory is used to store computer programs;

[0024] The processor is used to execute the computer program stored in the memory, so that the electronic device executes the above-mentioned dynamic magnetic resonance image calibration-free reconstruction method.

[0025] As described above, the method, system, medium, and electronic device for dynamic magnetic resonance image calibration-free reconstruction of the present invention have the following beneficial effects:

[0026] (1) High spatiotemporal resolution dynamic magnetic resonance imaging reconstruction based on implicit neural expression;

[0027] (2) It uses an unsupervised learning strategy, does not require a large amount of additional training data sets, and has good generalization performance under different scanning locations and scanning acceleration rates;

[0028] (3) No need for coil sensitivity maps that change over time, effectively reducing the complexity of the scanning and reconstruction process;

[0029] (4) Effectively improve the dynamic magnetic resonance image reconstruction effect under higher acceleration rates. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 Shown is a flow chart of a method for reconstructing a dynamic magnetic resonance image without calibration in one embodiment of the present invention;

[0031] Figure 2 It is a schematic diagram showing a framework of a method for reconstructing dynamic magnetic resonance images without calibration according to an embodiment of the present invention;

[0032] Figure 3 It is a schematic diagram showing the comparison of reconstruction effects of the non-calibration reconstruction method for dynamic magnetic resonance images in one embodiment of the present invention and the prior art at 16 times acceleration;

[0033] Figure 4 Shown is a schematic structural diagram of a dynamic magnetic resonance image calibration-free reconstruction system in one embodiment of the present invention;

[0034] Figure 5 It is a schematic structural diagram of an electronic device of the present invention in one embodiment. DETAILED DESCRIPTION

[0035] The following describes the embodiments of the present invention by specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0036] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present invention, and thus the drawings only show components related to the present invention rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0037] The technical solutions in the embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0038] Such as 1 and Figure 2 As shown, in one embodiment, the method for reconstructing dynamic magnetic resonance images without calibration of the present invention comprises steps S1 to S6.

[0039] Step S1, constructing a first implicit neural expression and a second implicit neural expression.

[0040] Specifically, Implicit Neural Representation (INR) is a new unsupervised deep learning framework based on the continuity prior of fully connected neural networks. It effectively solves the stability and generalization problems of supervised learning by learning the mapping relationship between coordinates and corresponding signals and achieving high-quality solutions to multiple inverse problems with a single sample.

[0041] In the present invention, the first implicit neural expression includes a first connected hash table and a first multi-layer perceptron (MLP). The second implicit neural expression includes a second connected hash table and a second multi-layer perceptron. The hash table is a data structure that is directly accessed according to a key value. That is, it accesses records by mapping the key value to a position in the table to speed up the search. The multi-layer perceptron is a forward artificial neural network composed of multiple layers of neurons. Its basic structure includes an input layer, a hidden layer, and an output layer. Each layer is composed of multiple neurons, and there can be multiple hidden layers. Each neuron of the multi-layer perceptron is connected to all neurons in the previous layer, and information is transmitted and processed through weights and activation functions.

[0042] Step S2: input the three-dimensional spatiotemporal coordinates of the dynamic magnetic resonance image into the first implicit neural expression to obtain a reconstructed dynamic magnetic resonance image sequence.

[0043] Specifically, the three-dimensional spatiotemporal coordinates of the dynamic magnetic resonance image are input into the first implicit neural expression, according to The reconstructed dynamic magnetic resonance image sequence is acquired, wherein represents the first implicit neural expression, represents the i-th reconstructed dynamic magnetic resonance image, (1,1,1) to (N,N,T) represent the three-dimensional space-time coordinates, N represents the row / column size of the image matrix, T represents the total number of dynamic magnetic resonance image frames, θ i Representing parameters of the first implicit neural expression.

[0044] Step S3: input the three-dimensional spatiotemporal coordinates of the dynamic magnetic resonance image into the second implicit neural representation to obtain a coil sensitivity map sequence.

[0045] Specifically, the three-dimensional spatiotemporal coordinates of the dynamic magnetic resonance image are input into the second implicit neural expression, according to Obtain the coil sensitivity map sequence, where represents the second implicit neural expression, represents the coil sensitivity map of the cth coil (1≤c≤C) changing with time, (1,1,1) to (N,N,T) represent the three-dimensional space-time coordinates, N represents the row / column size of the image matrix, T represents the total number of dynamic magnetic resonance image frames, θ s Representing the parameters of the second implicit neural expression.

[0046] Step S4: calculating a total loss function of the first implicit neural expression and the second implicit neural expression based on the reconstructed dynamic magnetic resonance image sequence and the coil sensitivity map sequence.

[0047] Specifically, in the present invention, the total loss function includes the data consistency loss term Total variational loss term and the regularization loss term These three parts. So first calculate the data consistency loss term Total variational loss term and the regularization loss term Then calculate the total loss function where λ S With λ L is the total variational loss term and the regularization loss term The proportional adjustment parameter.

[0048] according to Calculate the data consistency loss term, where C represents the number of coils and ⊙ represents the element-by-element multiplication of the matrix; represents the downsampled Fourier transform, which is used to simulate the actual dynamic magnetic resonance acceleration acquisition process; M (M<N) represents the number of readout lines acquired per frame; represents the coil sensitivity diagram for the cth coil, represents the i-th reconstructed dynamic magnetic resonance image, represents the k-space data acquired from the c-th coil, and ∥…∥2 represents the 2-norm.

[0049] according to Calculate the total variational loss term according to Calculate the regularization loss term, where ∥…∥1 and ∥…∥ * Respectively represent the 1-norm and the nuclear norm, TV t (...) represents the total variation operator along the time dimension, represents the i-th reconstructed dynamic magnetic resonance image.

[0050] Therefore, the total loss function can be expressed as:

[0051]

[0052] Step S5: Minimize the total loss function using a gradient back-propagation algorithm to obtain the trained first implicit neural expression and the second implicit neural expression.

[0053] Specifically, the parameters of the first implicit neural expression and the second implicit neural expression are adjusted using the gradient back-propagation algorithm, and the total loss function is minimized by iteration until the first implicit neural expression and the second implicit neural expression converge, thereby obtaining the trained first implicit neural expression and the second implicit neural expression.

[0054] Step S6: reconstructing a dynamic magnetic resonance image based on the trained first implicit neural expression.

[0055] Specifically, the three-dimensional spatiotemporal coordinates of the dynamic magnetic resonance image to be reconstructed are input into the trained first implicit neural expression to obtain the corresponding reconstructed dynamic magnetic resonance image.

[0056] like Figure 3 As shown, by comparing the reconstructed dynamic magnetic resonance images obtained by using the full sampling gold standard algorithm, the K-space zero-filling algorithm and the dynamic magnetic resonance image calibration-free reconstruction algorithm of the present invention at 16 times acceleration, it can be clearly seen that the algorithm of the present invention has higher reliability and stability.

[0057] The protection scope of the dynamic magnetic resonance image calibration-free reconstruction method described in the embodiment of the present invention is not limited to the execution order of the steps listed in this embodiment. All solutions implemented by adding, reducing or replacing steps in the prior art based on the principles of the present invention are included in the protection scope of the present invention.

[0058] An embodiment of the present invention further provides a dynamic magnetic resonance image calibration-free reconstruction system, which can implement the dynamic magnetic resonance image calibration-free reconstruction method described in the present invention. However, the implementation device of the dynamic magnetic resonance image calibration-free reconstruction system described in the present invention includes but is not limited to the structure of the dynamic magnetic resonance image calibration-free reconstruction system listed in this embodiment. All structural deformations and replacements of the prior art made according to the principles of the present invention are included in the protection scope of the present invention.

[0059] like Figure 4 As shown, in one embodiment, the dynamic magnetic resonance image calibration-free reconstruction system of the present invention includes a construction module 41 , a first acquisition module 42 , a second acquisition module 43 , a calculation module 44 , a training module 45 and a reconstruction module 46 .

[0060] The construction module 41 is used to construct a first implicit neural expression and a second implicit neural expression.

[0061] The first acquisition module 42 is connected to the construction module 41, and is used to input the three-dimensional spatiotemporal coordinates of the dynamic magnetic resonance image into the first implicit neural expression to acquire and reconstruct a dynamic magnetic resonance image sequence.

[0062] The second acquisition module 43 is connected to the construction module 41 and is used to input the three-dimensional spatiotemporal coordinates of the dynamic magnetic resonance image into the second implicit neural expression to acquire a coil sensitivity map sequence.

[0063] The calculation module 44 is connected to the first acquisition module 42 and the second acquisition module 43, and is used to calculate the total loss function of the first implicit neural expression and the second implicit neural expression based on the reconstructed dynamic magnetic resonance image sequence and the coil sensitivity map sequence.

[0064] The training module 45 is connected to the calculation module 44 and is used to minimize the total loss function using a gradient back-propagation algorithm to obtain the trained first implicit neural expression and the second implicit neural expression.

[0065] The reconstruction module 46 is connected to the training module 45 and is used to realize the reconstruction of dynamic magnetic resonance images based on the trained first implicit neural expression.

[0066] Among them, the structures and principles of the construction module 41, the first acquisition module 42, the second acquisition module 43, the calculation module 44, the training module 45 and the reconstruction module 46 correspond one to one with the above-mentioned dynamic magnetic resonance image calibration-free reconstruction method, so they are not repeated here.

[0067] In the several embodiments provided by the present invention, it should be understood that the disclosed system, device or method can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of modules / units is only a logical function division, and there may be other division methods in actual implementation, such as multiple modules or units can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules or units, which can be electrical, mechanical or other forms.

[0068] The modules / units described as separate components may or may not be physically separated, and the components displayed as modules / units may or may not be physical modules, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules / units may be selected according to actual needs to achieve the purpose of the embodiments of the present invention. For example, the functional modules / units in the various embodiments of the present invention may be integrated into one processing module, or each module / unit may exist physically separately, or two or more modules / units may be integrated into one module / unit.

[0069] Those of ordinary skill in the art should further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0070] The embodiment of the present invention also provides a computer-readable storage medium. A person skilled in the art can understand that all or part of the steps in the method for reconstructing dynamic magnetic resonance images without calibration in the above embodiment can be completed by instructing a processor through a program, and the program can be stored in a computer-readable storage medium, and the storage medium is a non-transitory medium, such as a random access memory, a read-only memory, a flash memory, a hard disk, a solid-state hard disk, a magnetic tape, a floppy disk, an optical disc, and any combination thereof. The above storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or a data center that includes one or more available media. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a digital video disc (DVD)), or a semiconductor medium (e.g., a solid-state disk (SSD)), etc.

[0071] An embodiment of the present invention further provides an electronic device, which includes a processor and a memory.

[0072] The memory is used to store computer programs.

[0073] The memory includes: ROM, RAM, disk, USB flash drive, memory card or CD and other media that can store program codes.

[0074] The processor is connected to the memory and is used to execute the computer program stored in the memory so that the electronic device executes the above-mentioned dynamic magnetic resonance image calibration-free reconstruction method.

[0075] Preferably, the processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components.

[0076] like Figure 5As shown, the electronic device of the present invention is in the form of a general computing device. The components of the electronic device may include but are not limited to: one or more processors or processing units 51, a memory 52, and a bus 53 connecting different system components (including the memory 52 and the processing unit 51).

[0077] Bus 53 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor or a local bus using any of a variety of bus architectures. For example, these architectures include but are not limited to Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MAC) bus, Enhanced ISA bus, Video Electronics Standards Association (VESA) local bus and Peripheral Component Interconnect (PCI) bus.

[0078] Electronic devices typically include a variety of computer system readable media. These media can be any available media that can be accessed by the electronic device, including volatile and non-volatile media, removable and non-removable media.

[0079] The memory 52 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 521 and / or cache memory 522. The electronic device may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 523 may be used to read and write non-removable, non-volatile magnetic media ( Figure 5 not shown, usually called a "hard drive"). Although Figure 5 Not shown, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, a DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to the bus 53 via one or more data medium interfaces. The memory 52 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of the various embodiments of the present invention.

[0080] A program / utility 524 having a set (at least one) of program modules 5241 may be stored, for example, in the memory 52, such program modules 5241 including but not limited to an operating system, one or more application programs, other program modules, and program data, each of which or some combination thereof may include an implementation of a network environment. The program modules 5241 generally perform the functions and / or methods of the embodiments described herein.

[0081] The electronic device may also communicate with one or more external devices (e.g., keyboards, pointing devices, displays, etc.), one or more devices that enable a user to interact with the electronic device, and / or any device that enables the electronic device to communicate with one or more other computing devices (e.g., network cards, modems, etc.). Such communication may be performed via input / output (I / O) interface 54. Furthermore, the electronic device may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via network adapter 55. Figure 5 As shown, the network adapter 55 communicates with other modules of the electronic device via the bus 53. It should be understood that, although not shown in the figure, other hardware and / or software modules can be used in conjunction with the electronic device, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0082] The above embodiments are merely illustrative of the principles and effects of the present invention, and are not intended to limit the present invention. Anyone familiar with the art may modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by a person of ordinary skill in the art without departing from the spirit and technical concept disclosed by the present invention shall still be covered by the claims of the present invention.

Claims

1. A method for reconstructing dynamic magnetic resonance images without calibration, characterized in that: The method comprises the following steps: constructing a first implicit neural expression and a second implicit neural expression; Inputting the three-dimensional spatiotemporal coordinates of the dynamic magnetic resonance image into the first implicit neural expression to obtain a reconstructed dynamic magnetic resonance image sequence; inputting the three-dimensional spatiotemporal coordinates of the dynamic magnetic resonance image into the second implicit neural representation to obtain a coil sensitivity map sequence; calculating a total loss function of the first implicit neural expression and the second implicit neural expression based on the reconstructed dynamic magnetic resonance image sequence and the coil sensitivity map sequence; Minimizing the total loss function by using a gradient back propagation algorithm to obtain the trained first implicit neural expression and the second implicit neural expression; wherein the parameters of the first implicit neural expression and the second implicit neural expression are adjusted by using the gradient back propagation algorithm, and the total loss function is minimized by iterations until the first implicit neural expression and the second implicit neural expression converge to obtain the trained first implicit neural expression and the second implicit neural expression; Reconstructing a dynamic magnetic resonance image based on the trained first implicit neural expression; according to The reconstructed dynamic magnetic resonance image sequence is acquired, wherein represents the i-th reconstructed dynamic magnetic resonance image, represents the first implicit neural expression, (1,1,1) to (N,N,T) represent three-dimensional space-time coordinates, N represents the row / column size of the image matrix, and T represents the total number of dynamic magnetic resonance image frames; according to Obtain the coil sensitivity map sequence, where represents the coil sensitivity diagram for the cth coil, represents the second implicit neural expression, (1,1,1) to (N,N,T) represent three-dimensional space-time coordinates, N represents the row / column size of the image matrix, and T represents the total number of dynamic magnetic resonance image frames.

2. The method for dynamic magnetic resonance image calibration-free reconstruction according to claim 1, characterized in that: The first implicit neural expression includes a first hash table and a first multi-layer perceptron connected to each other; the second implicit neural expression includes a second hash table and a second multi-layer perceptron connected to each other.

3. The method for dynamic magnetic resonance image calibration-free reconstruction according to claim 1, characterized in that: Calculating the total loss function of the first implicit neural expression and the second implicit neural expression based on the reconstructed dynamic magnetic resonance image sequence and the coil sensitivity map sequence comprises the following steps: Calculate data consistency loss Total variational loss term and the regularization loss term Calculate the total loss function where λ S With λ L is the proportional adjustment parameter.

4. The method for dynamic magnetic resonance image calibration-free reconstruction according to claim 3, characterized in that: according to Calculate the data consistency loss term, where C represents the number of coils, ⊙ represents the matrix element-by-element multiplication; F u represents the downsampled Fourier transform, represents the coil sensitivity diagram for the cth coil, represents the i-th reconstructed dynamic magnetic resonance image, y c represents the k-space data acquired from the c-th coil, and ||…||2 represents the 2-norm.

5. The method for dynamic magnetic resonance image calibration-free reconstruction according to claim 3, characterized in that: according to Calculate the total variational loss term according to Calculate the regularization loss term, where ||…||1 and ||…|| * Respectively represent the 1-norm and the nuclear norm, TV t (...) represents the total variation operator along the time dimension, represents the i-th reconstructed dynamic magnetic resonance image.

6. A dynamic magnetic resonance image calibration-free reconstruction system, characterized in that: The system comprises a construction module, a first acquisition module, a second acquisition module, a calculation module, a training module and a reconstruction module; The building module is used to build a first implicit neural expression and a second implicit neural expression; The first acquisition module is used to input the three-dimensional spatiotemporal coordinates of the dynamic magnetic resonance image into the first implicit neural expression to acquire and reconstruct a dynamic magnetic resonance image sequence; The second acquisition module is used to input the three-dimensional spatiotemporal coordinates of the dynamic magnetic resonance image into the second implicit neural expression to acquire a coil sensitivity map sequence; The calculation module is used to calculate the total loss function of the first implicit neural expression and the second implicit neural expression based on the reconstructed dynamic magnetic resonance image sequence and the coil sensitivity map sequence; The training module is used to minimize the total loss function by using a gradient back propagation algorithm to obtain the trained first implicit neural expression and the second implicit neural expression; wherein the parameters of the first implicit neural expression and the second implicit neural expression are adjusted by using a gradient back propagation algorithm, and the total loss function is minimized by iterations until the first implicit neural expression and the second implicit neural expression converge to obtain the trained first implicit neural expression and the second implicit neural expression; The reconstruction module is used to realize the reconstruction of dynamic magnetic resonance images based on the trained first implicit neural expression; according to The reconstructed dynamic magnetic resonance image sequence is acquired, wherein represents the i-th reconstructed dynamic magnetic resonance image, represents the first implicit neural expression, (1,1,1) to (N,N,T) represent three-dimensional space-time coordinates, N represents the row / column size of the image matrix, and T represents the total number of dynamic magnetic resonance image frames; according to Obtain the coil sensitivity map sequence, where represents the coil sensitivity diagram for the cth coil, represents the second implicit neural expression, (1,1,1) to (N,N,T) represent three-dimensional space-time coordinates, N represents the row / column size of the image matrix, and T represents the total number of dynamic magnetic resonance image frames.

7. A storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method for dynamic magnetic resonance image calibration-free reconstruction according to any one of claims 1 to 5 is implemented.

8. An electronic device, characterized in that: include: Processor and memory; The memory is used to store computer programs; The processor is used to execute the computer program stored in the memory, so that the electronic device performs the dynamic magnetic resonance image calibration-free reconstruction method according to any one of claims 1 to 5.

Images