Electromagnetic interference suppression method and system for ultralow field magnetic resonance imaging

Abstract

The invention discloses an electromagnetic interference suppression method and system for ultra-low field magnetic resonance imaging, and belongs to the technical field of nuclear magnetic resonance, and the system comprises an electromagnetic interference signal collection module which is composed of a plurality of EMI coils and corresponding signal collectors and actively detects EMI signals from the environment; the electromagnetic interference suppression module is used for performing model training by using the data, performing electromagnetic interference suppression through a model, predicting and eliminating an EMI component based on the electromagnetic interference suppression model, and outputting k space data without EMI; and the image reconstruction module is used for receiving the EMI-free k-space data and carrying out averaging and image reconstruction on the data so as to generate a high-quality MRI image. According to the invention, EMI noise is effectively suppressed, image quality is improved, and mobile and portable development of MRI equipment is promoted.

Description

Background Art

[0002] Magnetic Resonance Imaging (MRI) is a versatile imaging modality that can provide non-invasive and non-radiative quantitative characterization of biological tissues. After decades of development, it has now become a routine procedure in clinical research and diagnosis, having a huge impact on modern healthcare.

[0003] Currently, the market is mainly dominated by high-field magnetic resonance imaging equipment with 1.5T and 3T. Although it has high imaging quality, its high cost and immobility seriously limit the further popularization of magnetic resonance equipment. With the development of hardware and computing, etc., the development of low-cost, movable, ultra-low-field magnetic resonance equipment has become possible. Through the research of Electromagnetic Interference (EMI) elimination technology, the imaging quality of low-field magnetic resonance without physical shielding has been greatly improved to meet the relevant diagnostic requirements.

[0004] Problems existing in the prior art: 1. Traditional magnetic resonance equipment needs to be used in conjunction with a shielding room to eliminate the influence of electromagnetic interference on imaging quality, but this also seriously limits the mobility of the equipment.

[0005] 2. Traditional methods such as threshold denoising and image filtering generally only target specific interferences. In real-world scenarios with complex electromagnetic environments, the removal effect of these methods is not good.

[0006] 3. The EMI induction coil is a noise receiving channel and needs to be reasonably designed and optimized for specific equipment.

[0007] Various external sources (including nearby electronic devices) generate electromagnetic interference signals, and radiation with a frequency close to the Larmor frequency contaminates the magnetic resonance imaging signal through electromagnetic induction in the magnetic resonance imaging coil. If not mitigated, the interference will significantly reduce the image quality and diagnostic utility. Traditional electromagnetic shielding rooms can protect magnetic resonance imaging equipment from external electromagnetic interference sources, but also severely limit the portability and mobility of the equipment. Summary of the Invention

[0008] The technical problem to be solved by the present invention is to provide an electromagnetic interference suppression method and system for ultra-low-field magnetic resonance imaging, aiming at the deficiencies in the above-mentioned prior art, to achieve dynamic suppression of electromagnetic interference of ultra-low-field magnetic resonance equipment in a non-shielded environment, optimize the acquisition of human coupling noise and the collaborative elimination of multi-source interference, and enhance the adaptability to complex electromagnetic environments.

[0009] The present invention adopts the following technical solutions: An electromagnetic interference suppression method for ultra-low-field magnetic resonance imaging, comprising the following steps: Detect EMI signals in the environment, calculate the Pearson correlation coefficient of the single coil signal, and optimize the number and relative positions of EMI coils; Synchronously collect coil signals within the MRI signal acquisition window and the EMI signal characterization window respectively; Construct a deep learning model, train the deep learning model based on the synchronously collected coil signals, establish the correspondence between the EMI signals collected by the RF receiving coil and the EMI induction coil signals, and obtain the electromagnetic interference suppression model; According to the signals detected during the acquisition process of each frequency encoding line signal, use the electromagnetic interference suppression model to predict the EMI components within the MRI signal acquisition window; subtract the predicted EMI components from the MRI signals to generate frequency encoding line signals without EMI; Repeat the above operations for all frequency encoding lines, perform averaging and image reconstruction on the K-space data with EMI noise removed, and generate MRI images.

[0010] Preferably, place multiple EMI induction coils around the ultra-low field magnetic resonance imaging device to actively detect EMI signals from the environment; analyze the electromagnetic interference signals collected by the RF receiving coil and the EMI signals collected by each EMI coil to determine the spatial correlation coefficient of each EMI induction coil; optimize and determine the required number of the best EMI induction coils according to the spatial correlation coefficient of each EMI induction coil; optimize and adjust the relative positions of each EMI induction coil with respect to the magnetic resonance device according to the spatial correlation coefficient of each EMI induction coil.

[0011] Preferably, use a simulation method to determine the optimal structure of the noise acquisition coil, and match the center frequency and bandwidth of the acquisition coil through the corresponding matching method.

[0012] Preferably, obtain the spatial correlation coefficient by calculating the Pearson correlation coefficient between the signals collected by a single EMI induction coil and the RF receiving coil; confirm the acting direction of the electromagnetic interference according to the spatial correlation coefficient of each EMI induction coil; optimize and adjust the number of induction coils in combination with the calculation speed, model generalization, and hardware limitations.

[0013] Preferably, synchronously collect coil signals within the MRI signal acquisition window and the EMI signal characterization window respectively, specifically: Use an imaging sequence to perform electromagnetic interference scanning, and multiple EMI induction coils and the RF receiving coil collect electromagnetic interference signals in the same time window.

[0014] Preferably, establish the correspondence between the EMI signals collected by the RF receiving coil and the EMI induction coil signals, specifically: Optimize the determination of relevant parameters of the EMI induction coil and collect signals in the EMI induction coil and the RF receiving coil ; Fit the electromagnetic interference signals in the RF receiving coil through multiple electromagnetic interference induction coils ; Process the collected signals as the training data of the deep learning model; The signals received by multiple EMI induction coils are used as the input of the deep learning model, and the signals received by the RF receiving coil are used as the target of the deep learning model. Train the deep learning model to obtain an electromagnetic interference suppression model.

[0015] Preferably, the estimated electromagnetic interference signal at the RF receiving coil is:

[0016] where, is the magnetic vector potential of coil i per unit current, is the magnetic field sensitivity of coil i, , are the effective coupling weights between the kth noise source and the ith coil respectively, is the current density of the kth noise source, is the magnetization of the kth noise source, is the number of EMI induction coils, m is the number of noise sources in space, is the magnetized noise source, is the space where the electromagnetic interference noise source is located.

[0017] Preferably, the signals collected in the EMI induction coil and the RF receiving coil are:

[0018] where, is the magnetic vector potential of coil i per unit current, is the magnetic field sensitivity of coil i , is the noise source current density, is the noise source magnetization, is the sample magnetization source, is the mutual inductance coefficient between coil i and coil j, is the current in coil j, is the Boltzmann constant, T is the temperature, is the signal bandwidth, is the resistance of coil i, is the space where the electromagnetic interference noise source is located, is the space where the sample magnetization source is located.

[0019] Preferably, the deep learning model is an EMIC-Net model, which includes an input layer, an encoder, a bottleneck layer, a decoder, and an output layer. The encoding layer consists of 4 convolutional blocks, each block containing two layers of one-dimensional convolution, batch normalization, and ReLU activation function. The bottleneck layer contains consecutive Transformer blocks. The decoder consists of 4 convolutional blocks, each block containing two layers of one-dimensional convolution and ReLU activation function. The output layer contains a 1×1 convolutional layer.

[0020] In a second aspect, an electromagnetic interference suppression system for ultra-low field magnetic resonance imaging provided by an embodiment of the present invention includes: An electromagnetic interference signal acquisition module that detects EMI signals in the environment, calculates the Pearson correlation coefficient of the signals of a single coil, and optimizes the number and relative positions of EMI coils; synchronously acquires coil signals within both the MRI signal acquisition window and the EMI signal characterization window; An electromagnetic interference suppression module that constructs a deep learning model, trains the deep learning model based on the synchronously acquired coil signals, establishes the correspondence between the EMI signals collected by the radio frequency receiving coil and the EMI induction coil signals, and obtains an electromagnetic interference suppression model; predicts the EMI components within the MRI signal acquisition window using the electromagnetic interference suppression model according to the signals detected during the acquisition process of each frequency-encoded line signal; subtracts the predicted EMI components from the MRI signals to generate frequency-encoded line signals without EMI; An image reconstruction module that repeats the above operations for all frequency-encoded lines, averages and reconstructs the images using the K-space data with EMI noise removed to generate MRI images.

[0021] In a third aspect, a computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the above-mentioned electromagnetic interference suppression method for ultra-low field magnetic resonance imaging are implemented.

[0022] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, including a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned electromagnetic interference suppression method for ultra-low field magnetic resonance imaging are implemented.

[0023] In a fifth aspect, a chip includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the above-mentioned electromagnetic interference suppression method for ultra-low field magnetic resonance imaging are implemented.

[0024] In a sixth aspect, an embodiment of the present invention provides an electronic device, including a computer program, which implements the steps of the above-mentioned electromagnetic interference suppression method for ultra-low field magnetic resonance imaging when executed by the electronic device.

[0025] Compared with the prior art, the present invention has at least the following beneficial effects: An electromagnetic interference suppression method for ultra-low field magnetic resonance imaging combines active detection, deep learning modeling, and signal post-processing techniques. By detecting ambient EMI signals and optimizing coil configurations, the accuracy of signal acquisition is improved; a synchronous acquisition window design is adopted to ensure the temporal consistency of MRI signals and EMI signals, avoiding phase errors; a deep learning model is used to establish the prediction relationship of EMI components, accurately separating noise from the original signal, and finally reconstructing high-quality images through k-space data. By dynamically predicting EMI components using a deep learning model, the high cost of traditional shielding methods is significantly reduced, which is applicable to ultra-low field MRI devices in open environments; there is no need to rely on a radiofrequency shielding room, simplifying the equipment installation requirements and improving patient comfort. Combining hardware configuration and software algorithms, end-to-end noise suppression is achieved.

[0026] Furthermore, by analyzing the spatial correlation coefficient between the signals of the EMI induction coil and the radiofrequency receiving coil, the number and position of the coils are dynamically optimized. This method is based on the propagation characteristics of electromagnetic fields and solves the problem of insufficient adaptability of traditional fixed coil layouts to complex interference sources. Dynamically adjusting the coils according to the direction of the interference source improves the suppression efficiency of EMI in specific frequency bands, especially applicable to multi-source interference environments. By correlation screening, the number of redundant coils is reduced, lowering the hardware cost and computational complexity. Combining simulation and actual data ensures the stability of the layout scheme in different magnetic field environments.

[0027] Furthermore, a simulation method is used to design the optimal structure of the noise acquisition coil, and the signal acquisition efficiency is improved through parameter matching. By electromagnetic simulation to simulate the coil sensitivity distribution, ensuring matching with the EMI signals in the target frequency band and reducing the cost of experimental trial and error. Dynamically adjusting the coil parameters to adapt to the interference characteristics at different field strengths and enhancing the generalization ability of the system.

[0028] Furthermore, the signal correlation between the EMI induction coil and the receiving coil is quantified by the Pearson correlation coefficient, and the induction coil configuration is optimized by combining interference direction analysis and hardware limitations. Data-driven decision-making: Screening key interference sources based on statistical methods to avoid subjective experience errors and improve the scientific nature of the configuration. Achieving a balance among computational speed, model generalization, and hardware cost, applicable to resource-constrained scenarios.

[0029] Furthermore, within a unified time window, the MRI signal and the EMI signal are synchronously acquired, and a non-linear mapping relationship between the two is established through a deep learning model. The phase deviation caused by asynchronous sampling is eliminated to ensure signal alignment and improve the model prediction accuracy. Taking the radio frequency receiving coil signal as the target output, the EMI suppression effect is directly optimized, and the traditional multi-stage filtering process is simplified.

[0030] Furthermore, the source of the EMI signal is accurately described by a physical model formula, combined with the mathematical expression of the noise source, to achieve quantitative separation of the interference components. Based on the electromagnetic field theory of Maxwell's equations, the physical meaning of the model is ensured to be clear and applicable to complex interference scenarios. The contributions of current sources and magnetization sources are distinguished to improve the targeted suppression of different types of interference.

[0031] Furthermore, the EMIC-Net model is adopted. The encoder extracts multi-scale time features, implicitly models the noise source distribution, and captures the time-frequency characteristics of the noise. The decoder gradually restores the time resolution through upsampling and deconvolution, and the skip connection transfers the fine-grained details of the encoder to the decoder to assist in the weighted synthesis of the target noise signal. The model automatically captures the local correlation of the EMI signal, which is superior to traditional frequency domain filtering methods. Through lightweight network design, the real-time processing requirements of MRI imaging are met.

[0032] It can be understood that the beneficial effects of the second to sixth aspects above can refer to the relevant descriptions in the first aspect above, and will not be elaborated here.

[0033] In summary, through the combination of hardware optimization and deep learning, the present invention solves the electromagnetic interference problem of ultra-low field MRI equipment in an open environment, significantly reduces the equipment cost and usage threshold, and at the same time ensures the imaging quality, having broad clinical application potential.

[0034] The technical solutions of the present invention will be further described in detail below with reference to the drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required to be used in the embodiments of the present application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application, and those of ordinary skill in the art can obtain other drawings based on these drawings without creative efforts.

[0036] Figure 1 is a flowchart of an electromagnetic interference suppression system and method for an ultra-low field magnetic resonance imaging of the present invention; Figure 2 is a schematic diagram of the placement positions of the ultra-low field magnetic resonance imaging equipment and the electromagnetic interference induction coil used in the present invention; Figure 3 Schematic diagram of the electromagnetic interference induction coil structure adopted by the present invention; Figure 4 is a comparison diagram of magnetic resonance signals for the water film experiment. Among them, (a) is the one-dimensional magnetic resonance signal diagram of the contaminated water film, and (b) is the one-dimensional magnetic resonance signal diagram after electromagnetic interference detection and suppression by several electromagnetic interference induction coils; Figure 5 Figure 5 is an imaging comparison diagram for the water film experiment. Among them, (a) is the k-space image of the contaminated water film; (b) is the image of the contaminated water film, (c) is the k-space image after electromagnetic interference detection and suppression by several electromagnetic interference induction coils; (d) is the imaging diagram after electromagnetic interference detection and suppression by several electromagnetic interference induction coils; Figure 6 is an imaging comparison diagram for the human experiment. Among them, (a) is the imaging image of the contaminated human brain; (b) is the k-space image after electromagnetic interference detection and suppression by several electromagnetic interference induction coils; (c) is the imaging diagram of the collected electromagnetic interference noise; Figure 7 Schematic diagram of the computer device provided by an embodiment of the present invention; Figure 8 Block diagram of an electronic device provided by the present invention according to an embodiment.

[0037] Among them, 60. Computer device; 61. Processor; 62. Memory; 63. Computer program; 600. Electronic device; 610. Processing unit; 620. Storage unit; 6201. Random access storage unit; 6202. Cache storage unit; 6203. Read-only storage unit; 6204. Program / utilities; 6205. Program module; 630. Bus; 640. Display unit; 650. Input / output interface; 660. Network adapter; 700. External device. Specific embodiments

[0038] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0039] In the description of the present invention, it should be understood that the terms "including" and "comprising" indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their combinations.

[0040] It should also be understood that the terms used in the specification of the present invention are merely for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include the plural forms.

[0041] It should be further understood that the term "and / or" used in the specification of the present invention and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations. For example, A and / or B may represent: the case where A exists alone, the case where A and B exist simultaneously, and the case where B exists alone. In addition, the character " / " in the present invention generally represents an "or" relationship between the preceding and following related objects.

[0042] It should be understood that although the terms first, second, third, etc. may be used in the embodiments of the present invention to describe preset ranges, etc., these preset ranges should not be limited to these terms. These terms are only used to distinguish the preset ranges from each other. For example, without departing from the scope of the embodiments of the present invention, the first preset range may also be referred to as the second preset range, and similarly, the second preset range may also be referred to as the first preset range.

[0043] Depending on the context, the word "if" as used herein may be interpreted as "when" or "while" or "in response to determining" or "in response to detecting". Similarly, depending on the context, the phrase "if determined" or "if detected (stated condition or event)" may be interpreted as "when determined" or "in response to determining" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)".

[0044] Structural schematic diagrams according to the disclosed embodiments of the present invention are shown in the drawings. These figures are not drawn to scale, where certain details are enlarged for the purpose of clear expression, and certain details may be omitted. The shapes of the various regions and layers shown in the figures and their relative sizes and positional relationships are merely exemplary, and may actually deviate due to manufacturing tolerances or technical limitations, and those skilled in the art can design regions / layers with different shapes, sizes, and relative positions according to actual needs.

[0045] The present invention provides a method for suppressing electromagnetic interference in ultra-low field magnetic resonance imaging. By calculating the Pearson correlation coefficient of single-coil signals, an evaluation mechanism for the contribution degree of EMI detection coils is established. Combining the directional analysis of noise sources, the number of coils and the spatial layout are dynamically adjusted. Compared with the traditional fixed-coil design, this method models the electromagnetic field coupling path, enabling the sensor configuration to be matched with the complex electromagnetic environment in real time, improving the noise reception efficiency and system adaptability; innovatively adopts a synchronous sampling mechanism for the MRI signal acquisition window and the EMI characterization window, and establishes a signal mapping relationship between the radio frequency receiving coil and the EMI induction coil through a convolutional neural network (CNN). The method of the present invention breaks through the limitations of traditional frequency-domain transfer function or low-rank decomposition algorithms, utilizes the strong characterization ability of deep learning for time-varying interference, can simultaneously suppress multi-source interferences from medical devices, radio signals, etc., and verifies the cross-field strength applicability in 0.055T and 1.5T systems; automatically constructs a calibration matrix based on the peripheral data of the K-space, combines wavelet decomposition and complex least squares method for interference fitting, and can achieve dynamic noise reduction without additional calibration scans; compared with the deep learning method that requires pre-scan training, this scheme uses the concept of correlation window. It shortens the algorithm time-consuming from hours to minutes, significantly improving the imaging timeliness; abandons the traditional Faraday cage passive shielding, and realizes high-quality imaging in a non-shielded environment through an active noise cancellation algorithm chain.

[0046] Embodiment 1 An electromagnetic interference suppression method for ultra-low field magnetic resonance imaging according to the present invention includes the following steps: S1. An electromagnetic interference signal acquisition module actively detects EMI signals from the environment by strategically placing a plurality of EMI induction coils around the ultra-low field magnetic resonance imaging device; S101. Place a plurality of EMI induction coils around the ultra-low field magnetic resonance imaging device; Please refer to Figure 3 , and the EMI induction coil is used to acquire EMI signals in the environment.

[0047] The EMI induction coil can be in various forms such as a solenoid coil, a Helmholtz coil, etc. Simulation methods such as finite element are used to determine the optimal structure of the noise acquisition coil, and its center frequency and bandwidth are matched through a suitable matching method.

[0048] S102. Analyze the electromagnetic interference signals collected by the radio frequency receiving coil and the EMI signals collected in each EMI coil, and determine the spatial correlation coefficient of each EMI induction coil; The spatial correlation coefficient is obtained by calculating the Pearson correlation coefficient between the signals collected by a single EMI induction coil and the radio frequency receiving coil. The higher the spatial correlation coefficient, the more similar the signals collected by the EMI induction coil and the radio frequency receiving coil are.

[0049] Analyze and confirm the main direction of electromagnetic interference according to the spatial correlation coefficients of each EMI induction coil.

[0050] Optimize and adjust the number of induction coils in combination with aspects such as calculation speed, model generalization, and hardware limitations.

[0051] S103. Optimize and determine the required number of the best EMI induction coils according to the spatial correlation coefficients of each EMI induction coil; S104. Optimize and adjust the relative positions of each EMI induction coil with respect to the magnetic resonance device according to the spatial correlation coefficients of each EMI induction coil.

[0052] Recalculate the spatial correlation coefficients between the signals received by a single EMI induction coil at different relative positions and the signals collected in the radio frequency receiving coil.

[0053] Optimize and adjust the relative positions of the EMI induction coil and the device according to the recalculated spatial correlation coefficients.

[0054] The specific content of the above deep learning method is as follows: The neural network model includes an input layer and an output layer; The deep learning model is the EMIC-Net model, which includes an input layer, an encoder, a bottleneck layer, a decoder, and an output layer. The encoding layer consists of 4 convolutional blocks, each block contains two layers of one-dimensional convolution, batch normalization, and ReLU activation function. The bottleneck layer contains consecutive Transformer blocks. The decoder consists of 4 convolutional blocks, each block contains two layers of one-dimensional convolution and ReLU activation function. The output layer contains a 1×1 convolutional layer; The deep learning method can also be other neural network models.

[0055] S2. The electromagnetic interference signal acquisition module uses corresponding pulse sequences to collect data, and the radio frequency receiving coil and the EMI induction coil synchronously collect data within the MRI signal acquisition window and the EMI signal characterization window respectively; Use an imaging sequence to perform electromagnetic interference scanning, and several EMI induction coils and radio frequency receiving coils collect electromagnetic interference signals in the same time window.

[0056] S3. The electromagnetic interference suppression module trains a neural network model based on the collected data to establish the corresponding relationship between the EMI signals collected by the radio frequency receiving coil and the signals of the EMI induction coil; S301. Optimize and determine the relevant parameters of the EMI induction coil, and collect the signals in the EMI induction coil and the radio frequency receiving coil; S302. Process the collected signals as the training data of the EMIC-Net model; S303. The signals received by several EMI induction coils are used as the input of the EMIC-Net model, and the signal received by the RF receiving coil is used as the target of the model to train the model.

[0057] The real-time suppression of dynamic electromagnetic noise is achieved by synchronously sampling the electromagnetic noise reference channel and the magnetic resonance signal channel. In a specific implementation, the magnetic resonance spectrometer synchronously controls the multi-group sensors of the electromagnetic noise reference channel and the RF receiving coil of the magnetic resonance signal channel to collect data. The signals obtained by each channel are first amplified with low noise and high gain by a preamplifier, and then enter a signal conditioning circuit to complete secondary amplification and high and low pass filtering; finally, a sampling card configured with a peripheral circuit completes digital conversion and synchronously outputs the magnetic resonance signal contaminated by electromagnetic interference and the corresponding multi-channel reference electromagnetic noise signals.

[0058] The signals collected in the EMI induction coil and the RF receiving coil are expressed as follows:

[0059] Among them, is the magnetic vector potential of the coil i for a unit current, is the magnetic field sensitivity of the coil i , is the noise source current density, is the noise source magnetization, is the sample magnetization source, is the coil i and the coil j the mutual inductance coefficient between them, is the current in coil j, is the Boltzmann constant, T is the temperature, is the signal bandwidth, is the resistance of coil i, is the space where the electromagnetic interference noise source is located, is the space where the sample magnetization source is located.

[0060] The electromagnetic interference signal in the RF receiving coil is fitted by several electromagnetic interference induction coils, which is expressed by the formula as follows:

[0061] Among them, represents the estimated electromagnetic interference signal at the RF receiving coil.

[0062] By using deep learning methods or other methods, learn the mapping relationship between the electromagnetic interference signals fitted by several electromagnetic interference induction coils in the RF receiving coil, that is, learn the mapping relationship for generating synthetic electromagnetic interference signals from electromagnetic interference induction coils.

[0063] S4. Use the trained EMIC-Net model and, based on the signals detected by the EMI induction coil during the signal acquisition process of each frequency-encoding line, predict the EMI components within the MRI signal acquisition window; All the frequency-encoding lines obtained from one scan need to be input into the EMIC-Net model in sequence, and the EMI components within the MRI signal acquisition window are predicted; Recombine all the obtained FE lines without EMI to form complete k-space data, perform image reconstruction, and obtain high-quality images with EMI removed.

[0064] S5. Subtract the predicted EMI from the signals of the MRI receiving coil to generate frequency-encoding line signals without EMI; S6. Repeat the above operations for all frequency-encoding lines based on the electromagnetic interference suppression model, input the k-space data with EMI noise removed into the image reconstruction module, and the image reconstruction module performs averaging and image reconstruction through the k-space data without EMI, thereby generating high-quality MRI images.

[0065] Those skilled in the art can understand that various aspects of the present invention can be implemented as a system, method, or program product. Therefore, various aspects of the present invention can be specifically implemented in the following forms, namely: a complete hardware implementation, a complete software implementation (including firmware, microcode, etc.), or an implementation combining hardware and software aspects, which can be collectively referred to as "circuit", "module", or "platform" here.

[0066] Embodiment 2 The present invention provides an electromagnetic interference suppression system for ultra-low field magnetic resonance imaging. This system can be used to implement the electromagnetic interference suppression method for ultra-low field magnetic resonance imaging. Specifically, the electromagnetic interference suppression system for ultra-low field magnetic resonance imaging includes an electromagnetic interference signal acquisition module, an electromagnetic interference suppression module, and an image reconstruction module.

[0067] Among them, the electromagnetic interference signal acquisition module detects the EMI signals in the environment, calculates the Pearson correlation coefficient of the single-coil signals, and optimizes the configuration of the number and relative positions of the EMI coils; synchronously acquires coil signals within both the MRI signal acquisition window and the EMI signal characterization window; The electromagnetic interference suppression module constructs a deep learning model, trains the deep learning model based on the synchronously acquired coil signals, establishes the correspondence between the EMI signals collected by the radio frequency receiving coil and the EMI induction coil signals, and obtains an electromagnetic interference suppression model; according to the signals detected during the signal acquisition process of each frequency-encoding line, uses the electromagnetic interference suppression model to predict the EMI components within the MRI signal acquisition window; subtracts the predicted EMI components from the MRI signals to generate frequency-encoding line signals without EMI; The image reconstruction module repeats the above operations for all frequency-encoding lines, averages and reconstructs images using the K-space data with EMI noise removed to generate an MRI image.

[0068] Embodiment 3 The present invention provides a terminal device, which includes a processor and a memory. The memory is used to store a computer program, and the computer program includes program instructions. The processor is used to execute the program instructions stored in the computer storage medium. The processor may be a Central Processing Unit (CPU), or may also be other general-purpose processors, Graphics Processing Units (GPU), Tensor Processing Units (TPU), Digital Signal Processors (DSP), Application Specific Integrated Circuits (ASIC), Field-Programmable Gate Arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, and is suitable for implementing one or more instructions. Specifically, it is suitable for loading and executing one or more instructions to implement the corresponding method flow or corresponding function. The processor described in the embodiments of the present invention can be used for the operation of the electromagnetic interference suppression method for ultra-low field magnetic resonance imaging, including: Detect EMI signals in the environment, calculate the Pearson correlation coefficient of the single-coil signal, and optimize the number and relative positions of EMI coils; synchronously collect coil signals within the MRI signal acquisition window and the EMI signal characterization window respectively; construct a deep learning model, train the deep learning model based on the synchronously collected coil signals, establish the correspondence between the EMI signals collected by the radio frequency receiving coil and the EMI induction coil signals, and obtain an electromagnetic interference suppression model; according to the signals detected during the signal acquisition process of each frequency-encoding line, use the electromagnetic interference suppression model to predict the EMI components within the MRI signal acquisition window; subtract the predicted EMI components from the MRI signals to generate frequency-encoding line signals without EMI; repeat the above operations for all frequency-encoding lines, average and reconstruct images using the k-space data with EMI noise removed to generate an MRI image.

[0069] Please refer to Figure 7, the terminal device is a computer device. The computer device 60 in this embodiment includes: a processor 61, a memory 62, and a computer program 63 stored in the memory 62 and executable on the processor 61. When the computer program 63 is executed by the processor 61, it implements the electromagnetic interference suppression method for ultra-low field magnetic resonance imaging in the embodiment. To avoid repetition, it will not be elaborated here one by one. Alternatively, when the computer program 63 is executed by the processor 61, it implements the functions of each model / unit in the electromagnetic interference suppression system for ultra-low field magnetic resonance imaging in the embodiment. To avoid repetition, it will not be elaborated here one by one.

[0070] The computer device 60 can be a computing device such as a desktop computer, a notebook, a palm computer, and a cloud server. The computer device 60 may include, but is not limited to, a processor 61 and a memory 62. Those skilled in the art can understand that Figure 7 This is only an example of the computer device 60 and does not constitute a limitation on the computer device 60. It may include more or fewer components than shown in the figure, or combine certain components, or different components. For example, the computer device may also include input / output devices, network access devices, buses, etc.

[0071] The so-called processor 61 may be a central processing unit (CPU), or other general-purpose processors, graphics processing units (GPUs), tensor processing units (TPUs), digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0072] The memory 62 may be an internal storage unit of the computer device 60, such as the hard disk or memory of the computer device 60. The memory 62 may also be an external storage device of the computer device 60, such as a plug-in hard disk equipped on the computer device 60, a smart media card (SMC), a secure digital (SD) card, a flash card, etc.

[0073] Further, the memory 62 may also include both the internal storage unit of the computer device 60 and external storage devices. The memory 62 is used to store computer programs as well as other programs and data required by the computer device. The memory 62 may also be used to temporarily store data that has been output or is to be output.

[0074] Please refer to Figure 8 , the terminal device is the electronic device 600, and the electronic device 600 is presented in the form of a general-purpose computing device. The components of the electronic device may include, but are not limited to: at least one processing unit 610, at least one storage unit 620, a bus 630 connecting different platform components (including the storage unit 620 and the processing unit 610), a display unit 640, etc.

[0075] Among them, the storage unit stores program codes, and the program codes can be executed by the processing unit 610, so that the processing unit 610 executes the steps according to various exemplary embodiments of the present invention described in the above method part of this specification. For example, the processing unit 610 can execute steps as shown in Figure 1 .

[0076] The storage unit 620 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit (RAM) 6201 and / or a cache storage unit 6202, and may further include a read-only storage unit (ROM) 6203.

[0077] The storage unit 620 may also include a program / utilities 6204 having a set (at least one) of program modules 6205. Such program modules 6205 include, but are not limited to: an operating system, one or more application programs, other program modules, and program data. The implementation of a network environment may be included in each or some combination of these examples.

[0078] The bus 630 may represent one or more of several types of bus structures, including a memory bus or a memory controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local bus using any of the multiple bus structures.

[0079] The electronic device 600 may also communicate with one or more external devices 700 (such as a keyboard, a pointing device, a Bluetooth device, etc.), and may also communicate with one or more devices that enable a user to interact with the electronic device 600, and / or communicate with any device (such as a router, a modem) that enables the electronic device 600 to communicate with one or more other computing devices. Such communication may be carried out through the input / output interface 650. Moreover, the electronic device 600 may also communicate with one or more networks (such as a local area network, a wide area network, and / or a public network, such as the Internet) through the network adapter 660. The network adapter 660 may communicate with other modules of the electronic device 600 through the bus 630. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in combination with the electronic device 600, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage platforms, etc.

[0080] Embodiment 4 The present invention also provides a storage medium, specifically a computer-readable storage medium, which is a memory device in a terminal device and is used to store programs and data. It can be understood that the computer-readable storage medium here may include both the built-in storage medium in the terminal device and, of course, the extended storage medium supported by the terminal device. It may be any tangible medium that contains or stores a program, and this program can be used by or in combination with an instruction execution system, apparatus, or device. The computer-readable storage medium provides a storage space that stores the operating system of the terminal. And, in this storage space, one or more instructions suitable for being loaded and executed by a processor are also stored, and these instructions may be one or more computer programs (including program codes). It should be noted that more specific examples of the computer-readable storage medium here include: an electrical connection having one or more wires, a portable disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory, an optical fiber, a portable compact disk read-only memory, an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0081] The computer-readable storage medium also includes a data signal propagated in a baseband or as part of a carrier wave, which carries the readable program code. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing. The readable storage medium can also be any readable medium other than the readable storage medium, and the readable medium can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the readable storage medium can be transmitted using any appropriate medium, including but not limited to wireless, wired, optical fiber cable, radio frequency, etc., or any suitable combination of the foregoing.

[0082] The program code for performing the operations of the present invention can be written in any combination of one or more programming languages, including object-oriented programming languages such as Java, C++, etc., and also including conventional procedural programming languages such as the "C" language or similar programming languages. The program code can be executed entirely on the user computing device, partially on the user device, executed as a stand-alone software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user computing device through any type of network, including a local area network or a wide area network, or can be connected to an external computing device (e.g., by using an Internet service provider to connect through the Internet).

[0083] One or more instructions stored in the computer-readable storage medium can be loaded and executed by a processor to implement the corresponding steps of the electromagnetic interference suppression method for ultra-low field magnetic resonance imaging in the above embodiments; the one or more instructions in the computer-readable storage medium are loaded and executed by the processor to perform the following steps: Detect EMI signals in the environment, calculate the Pearson correlation coefficient of the single-coil signals, and optimize the number and relative positions of the EMI coils; synchronously collect coil signals within the MRI signal acquisition window and the EMI signal characterization window respectively; construct a deep learning model, train the deep learning model based on the synchronously collected coil signals, establish the correspondence between the EMI signals collected by the radio frequency receiving coil and the EMI induction coil signals, and obtain an electromagnetic interference suppression model; according to the signals detected during the acquisition process of each frequency-encoding line signal, use the electromagnetic interference suppression model to predict the EMI components within the MRI signal acquisition window; subtract the predicted EMI components from the MRI signals to generate frequency-encoding line signals without EMI; repeat the above operations for all frequency-encoding lines, and perform averaging and image reconstruction using the K-space data with EMI noise removed to generate an MRI image.

[0084] In each of the embodiments provided in the present application, the database involved may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., without limitation. In each of the embodiments provided in the present application, the processor involved may be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., without limitation.

[0085] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. Generally, the components described and shown in the accompanying drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the present invention claimed, but merely represents selected embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0086] During the experiment and normal scanning process, the imaging sequence is normally used for scanning, and several electromagnetic interference induction coils and radio frequency receiving coils collect electromagnetic interference signals respectively within the same time window.

[0087] During the optimization and adjustment, multiple EMI induction coils are placed around the ultra-low field magnetic resonance device, and each coil collects signals simultaneously. The Pearson correlation coefficient between the signals collected by a single EMI induction coil and the signals collected by the radio frequency receiving coil is calculated respectively to determine the spatial correlation coefficient of each EMI induction coil. Then, based on the distribution of the coil spatial correlation coefficients, the main sources and directions of the electromagnetic interference noise around the device are analyzed and evaluated. Considering aspects such as calculation speed, model generalization, and hardware limitations, the optimal number of required EMI induction coils is determined through optimization.

[0088] Sample and are the signal sequences detected by the two coils respectively at the same time. Then, the Pearson correlation coefficient used to calculate the spatial correlation coefficient is as shown in the formula

[0089] where is the length of the signal sequence, are the sample data respectively , mean values.

[0090] The Pearson correlation coefficient has the following properties: 1. The Pearson correlation coefficient can calculate the correlation between two sample data with different dimensions.

[0091] 2. The absolute value of the Pearson correlation coefficient is not greater than 1. When the value is 1, it indicates that the two sample data are completely linearly positively correlated; when the value is -1, it indicates that the two sample data are completely linearly negatively correlated; when the value is 0, it indicates that the two sample data have no linear correlation at all; when the value is between -1 and 1, it indicates that the two sample data have a certain linear correlation, and the degree of correlation is related to the magnitude of the Pearson correlation coefficient.

[0092] 3. When the Pearson correlation coefficient is greater than 0, it indicates that the change trends of the two data samples are the same. When the Pearson correlation coefficient is less than 0, it indicates that the change trends of the two sample data are opposite.

[0093] Constructing a deep learning model: The above neural network model can be a multi-layer convolutional neural network or other neural network models. The multi-layer convolutional neural network includes an input layer and an output layer. Between the input and output layers, convolutional layers, pooling layers, and normalization layers are designed. After optimizing and determining the relevant parameters of the electromagnetic interference coil, signals in the EMI induction coil and the radio frequency receiving coil are collected, and the collected signals are appropriately processed. The signal received by the EMI induction coil is used as the input of the model, and the signal received by the radio frequency receiving coil is used as the target of the model to train the model.

[0094] Electromagnetic interference noise elimination method: Using several optimized EMI induction coils, electromagnetic interference signals are collected. Based on the trained deep learning model, the electromagnetic interference signals received by the radio frequency receiving coil during imaging are fitted to obtain a synthetic noise signal. When imaging, the signal collected by the radio frequency receiving coil is subtracted from the synthetic noise signal obtained by fitting the signal collected by the EMI induction coil to obtain an imaging signal after removing the electromagnetic interference noise.

[0095] Configure the EMI induction coil in the ultra-low field magnetic resonance imaging system as described above Figure 2 as shown. The results of data collection on the water film in an open environment (without electromagnetic shielding) are shown in Figure 4, Figure 5 as shown. Figure 4 is a comparison diagram of magnetic resonance signals for the water film experiment. (a) One-dimensional magnetic resonance signal diagram of the contaminated water film; (b) One-dimensional magnetic resonance signal diagram after EMI detection and suppression by several electromagnetic interference induction coils.

[0096] Figure 5For the imaging comparison diagram of the water film experiment, (a) k-space image of the contaminated water film; (b) image of the contaminated water film; (c) k-space image after EMI detection and suppression by several electromagnetic interference induction coils; (d) imaging diagram after EMI detection and suppression by several electromagnetic interference induction coils. It can be seen from the experimental data that the electromagnetic interference suppression system and method for ultra-low field magnetic resonance imaging proposed by the present invention has a good suppression effect on electromagnetic noise.

[0097] Figure 6 is the imaging comparison diagram of the human experiment, (a) the imaging image of the contaminated human brain; (b) the k-space image after EMI detection and suppression by several electromagnetic interference induction coils; (c) the imaging diagram of the collected electromagnetic interference noise.

[0098] The results of the human brain comparison images in the figure prove that the electromagnetic interference suppression system and method for ultra-low field magnetic resonance imaging proposed by the present invention has a good suppression effect on the electromagnetic noise received by the ultra-low field magnetic resonance imaging system in an open environment.

[0099] In summary, the electromagnetic interference suppression method and system for ultra-low field magnetic resonance imaging of the present invention have the following characteristics: 1. Abandon the traditional passive shielding electromagnetic noise suppression method and adopt the active noise reduction method. Using the proposed method, the surrounding interference noise signals are obtained through appropriate electromagnetic interference induction coils, the interference noise signals in the radio frequency receiving coil are fitted, and the signal obtained by fitting is subtracted from the total signal of the imaging coil to eliminate the interference noise, significantly improving the imaging quality of the image and promoting the further development of low-field magnetic resonance equipment towards true mobility and portability.

[0100] 2. Based on the deep learning method, there is no need to construct a complex EMI noise model, reducing the learning and application costs. At the same time, the trained model is used to remove EMI, improving the calculation efficiency and realizing the real-time performance of noise reduction.

[0101] 3. By analyzing and calculating the Pearson correlation coefficient between a single EMI induction coil and the radio frequency receiving coil, the number and position of the EMI induction coils are optimized and adjusted, improving the signal acquisition efficiency of the EMI induction coils.

[0102] 4. Using fewer EMI induction coils to collect signals reduces the learning cost of the deep learning neural network for training the mapping relationship between the interference noise obtained by the EMI induction coils and the interference noise of the imaging coil, reduces the requirements for hardware devices, and improves the generalization of the deep learning network model.

[0103] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the above division of each functional unit and module is used as an example. In actual applications, the above functions can be allocated to different functional units and modules according to needs, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiments can be integrated into a processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit. In addition, the specific names of each functional unit and module are only for the convenience of mutual distinction and do not limit the protection scope of the present application. The specific working processes of the units and modules in the above system can refer to the corresponding processes in the foregoing method embodiments and will not be described herein again.

[0104] In the above embodiments, the descriptions of the various embodiments have their own emphases. For the parts not detailed or recorded in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0105] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed in the present invention can be implemented by electronic hardware, or by a combination of computer software and electronic hardware. Whether these functions are executed in the form of hardware or software depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0106] In the embodiments provided by the present invention, it should be understood that the disclosed device / terminal and method can be implemented in other ways. For example, the device / terminal embodiments described above are only illustrative. For example, the division of the modules or units is only a logical function division, and there can be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of the devices or units can be in electrical, mechanical or other forms.

[0107] The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0108] In addition, in each embodiment of the present invention, the functional units may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit. The above integrated units may be implemented in the form of hardware or in the form of software functional units.

[0109] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it may be stored in a computer-readable storage medium. Based on such an understanding, to implement all or part of the processes in the above method embodiments of the present invention, it may also be completed by instructing relevant hardware through a computer program. The computer program may be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code may be in the form of source code, object code, executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc. It should be noted that the content included in the computer-readable medium may be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electrical carrier signals and telecommunication signals.

[0110] This application is described with reference to the flowcharts and / or block diagrams of methods, apparatuses, and computer program products according to embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, and the combination of processes and / or blocks in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate a device for implementing the specified function in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0111] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured product including an instruction device, and the instruction device implements the process in Figure 1One or more processes and / or blocks Figure 1 The functions specified in one or more blocks.

[0112] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one Figure 1 One or more processes and / or blocks Figure 1 or more processes and / or blocks.

[0113] The above content is only to illustrate the technical idea of the present invention and cannot be used to limit the protection scope of the present invention. Any modification made on the basis of the technical solution according to the technical idea proposed by the present invention falls within the protection scope of the claims of the present invention.

Claims

1. An electromagnetic interference suppression method for ultra-low field magnetic resonance imaging, characterized in that, Including the following steps: Detect EMI signals in the environment, calculate the Pearson correlation coefficient of the single coil signal, and optimize the number and relative positions of the EMI coils; Synchronously collect coil signals within the MRI signal acquisition window and the EMI signal characterization window respectively; Construct a deep learning model, train the deep learning model based on the synchronously collected coil signals, establish the correspondence between the EMI signals collected by the RF receiving coil and the EMI induction coil signals, and obtain an electromagnetic interference suppression model; According to the signals detected during the signal acquisition process of each frequency encoding line, use the electromagnetic interference suppression model to predict the EMI components within the MRI signal acquisition window; subtract the predicted EMI components from the MRI signals to generate frequency encoding line signals without EMI; Repeat the above operations for all frequency encoding lines, perform averaging and image reconstruction on the k-space data with EMI noise removed, and generate MRI images.

2. The electromagnetic interference suppression method for ultra-low field magnetic resonance imaging according to claim 1, wherein Place multiple EMI induction coils around the ultra-low field magnetic resonance imaging device to actively detect EMI signals from the environment; analyze the electromagnetic interference signals collected by the RF receiving coil and the EMI signals collected in each EMI coil to determine the spatial correlation coefficients of each EMI induction coil; optimize and determine the required number of the best EMI induction coils according to the spatial correlation coefficients of each EMI induction coil; optimize and adjust the relative positions of each EMI induction coil with respect to the magnetic resonance device according to the spatial correlation coefficients of each EMI induction coil.

3. The electromagnetic interference suppression method for ultra-low field magnetic resonance imaging according to claim 2, wherein Use a simulation method to determine the optimal structure of the noise acquisition coil, and match the center frequency and bandwidth of the acquisition coil through the corresponding matching method.

4. The electromagnetic interference suppression method for ultra-low field magnetic resonance imaging according to claim 2, wherein Obtain the spatial correlation coefficients by calculating the Pearson correlation coefficients between the signals collected by a single EMI induction coil and the RF receiving coil; confirm the acting direction of the electromagnetic interference according to the spatial correlation coefficients of each EMI induction coil; optimize and adjust the number of induction coils in combination with the calculation speed, model generalization ability, and hardware limitations.

5. The electromagnetic interference suppression method for ultra-low field magnetic resonance imaging according to claim 1, characterized in that Synchronously collect coil signals within the MRI signal acquisition window and the EMI signal characterization window respectively, specifically: Use an imaging sequence to perform electromagnetic interference scanning, and multiple EMI induction coils and the RF receiving coil collect electromagnetic interference signals in the same time window.

6. The electromagnetic interference suppression method for ultra-low field magnetic resonance imaging according to claim 1, characterized in that Establish the correspondence between the EMI signals collected by the RF receiving coil and the EMI induction coil signals, specifically: Optimize the determination of relevant parameters of the EMI induction coil, and collect signals in the EMI induction coil and the RF receiving coil ; Fit the electromagnetic interference signal in the RF receiving coil through multiple electromagnetic interference induction coils ; Process the collected signals as the training data of the deep learning model; The signals received by multiple EMI induction coils are used as the input of the deep learning model, and the signals received by the RF receiving coil are used as the target of the deep learning model to train the deep learning model and obtain an electromagnetic interference suppression model.

7. The electromagnetic interference suppression method for ultra-low field magnetic resonance imaging according to claim 6, characterized in that Estimated electromagnetic interference signal at the radio frequency receiving coil is as follows: Among them, is the magnetic vector potential of the coil i for unit current, is the magnetic field sensitivity of the coil i ; 、 are respectively the effective coupling weights between the k -th noise source and the i-th coil, is the current density of the k -th noise source, is the magnetization of the k -th noise source, is the number of EMI induction coils, m is the number of noise sources in space, is the magnetized noise source, is the space where the electromagnetic interference noise source is located.

8. The electromagnetic interference suppression method for ultra-low field magnetic resonance imaging according to claim 7, characterized in that Signals collected in the EMI induction coil and the RF receiving coil are as follows: Among them, is the coil i magnetic vector potential of unit current, is the coil i magnetic field sensitivity, is the noise source current density, is the noise source magnetization, is the sample magnetization source, is the coil i and the coil j mutual inductance coefficient between them, is the current in the coil j is the Boltzmann constant, T is the temperature, is the signal bandwidth, is the coil i resistance, is the space where the electromagnetic interference noise source is located, is the space where the sample magnetization source is located.​ 9. The electromagnetic interference suppression method for ultra-low field magnetic resonance imaging according to claim 1, characterized in that The deep learning model is EMIC-Net, which includes an input layer, an encoder, a bottleneck layer, a decoder, and an output layer. The encoding layer consists of 4 convolutional blocks, each block contains two layers of one-dimensional convolution, batch normalization, and ReLU activation functions. The bottleneck layer contains consecutive Transformer blocks. The decoder consists of 4 convolutional blocks, each block contains two layers of one-dimensional convolution and ReLU activation functions. The output layer contains a 1×1 convolutional layer.

10. An electromagnetic interference suppression system for ultra-low field magnetic resonance imaging, characterized in that, Including: The electromagnetic interference signal acquisition module detects EMI signals in the environment, calculates the Pearson correlation coefficient of the single coil signal, and optimizes the number and relative positions of EMI coils; synchronously acquires coil signals within the MRI signal acquisition window and the EMI signal characterization window respectively; The electromagnetic interference suppression module constructs a deep learning model, trains the deep learning model based on the synchronously acquired coil signals, establishes the correspondence between the EMI signals collected by the radio frequency receiving coil and the EMI induction coil signals, and obtains the electromagnetic interference suppression model; according to the signals detected during the signal acquisition process of each frequency encoding line, uses the electromagnetic interference suppression model to predict the EMI components within the MRI signal acquisition window; subtracts the predicted EMI components from the MRI signals to generate frequency encoding line signals without EMI; The image reconstruction module repeats the above operations for all frequency encoding lines, averages and reconstructs the K-space data with EMI noise removed to generate MRI images.

Images