Magnetic resonance fast shimming method, system, terminal and medium

Abstract

The present invention discloses a rapid shimming method, system, terminal, and medium for magnetic resonance imaging. This involves generating a field map template in standard space and converting a T1 structure diagram to standard space to obtain a field map template in image space. Based on the acquired field map, a predicted field map is obtained, located in the same image space as the acquired field map, for accurately describing the magnetic field distribution. The predicted field map is then de-wound in phase, and the shimming coil numerically optimizes the de-wound predicted field map to complete the shimming process. This invention enables the acquisition of a field map that accurately describes the magnetic field distribution in the imaging region and a rapid shimming process. It is particularly effective for achieving better shimming of magnetic field inhomogeneities caused by differences in magnetic susceptibility among different tissues. Furthermore, by simulating and predicting the field map, it reduces iteration and repetition in the shimming process, enabling the acquisition of a more accurate field map describing the magnetic field distribution within the imaging region in a shorter time. It can also provide better shimming for imaging regions of tissues such as the abdomen and brain.

Description

Technical Field

[0001] The present invention relates to the field of magnetic resonance, and in particular to a magnetic resonance rapid field shimming method, system, terminal and medium. Background Art

[0002] In high-field and ultra-high-field MRI, it is crucial to ensure the uniformity of the magnetic field. However, in cases such as abdominal scanning, due to the large FOV, multiple organs, complex tissue composition, and tissue movement and local magnetic field distortion caused by human breathing, how to ensure the high-field MRI B0 uniform field effect in such cases is the key to carrying out advanced applications such as abdominal imaging examinations and liver quantitative analysis.

[0003] In magnetic resonance systems, common shimming measures include active shimming and passive shimming. Passive shimming, also known as passive shimming, is generally achieved by attaching small magnetic sheets inside the magnet to achieve preliminary uniformity adjustment of the magnetic field. Active shimming, also known as active shimming, is an optimization based on passive shimming. It compensates for the non-uniformity of the original magnetic field by placing a magnetic field generated by an energized shimming coil, and makes fine corrections to the non-uniformity of the main magnetic field.

[0004] Active shimming is divided into first-order shimming and high-order shimming. Most high-field MRI machines above 1.5T used in clinical practice are equipped with second-order and higher shimming coils. By performing spherical harmonic expansion on the original magnetic field distribution and the magnetic field distribution generated by the coil, the DC current of each coil is calculated to generate the optimal magnetic field distribution after shimming. High-order shimming requires the hardware support of high-order shimming coils. High-order shimming coils can generate a high-order distributed local magnetic field in space, thereby offsetting the high-order components of the inhomogeneity of the main magnetic field.

[0005] The common active shimming steps can be described as follows:

[0006] Step 1. Use the double echo steady-state free precession (DESS) sequence to acquire the field map.

[0007] Step 2. If the TE difference between the two echoes is large, the phase may be wrapped and phase unwrapping is required.

[0008] Step 3. Subtract the phase diagrams of the two echoes to obtain the distribution difference of the magnetic field and calculate the first-order and high-order components of the field diagram.

[0009] Step 4. Use different shimming algorithms to adjust the current value of the shimming coil of the corresponding order, so that the orthogonal high-order shimming coils work together in space to offset the uneven high-order components of the main magnetic field.

[0010] For the traditional shimming process, it is impossible to have both a field map that accurately describes the magnetic field distribution in the imaging area and a fast shimming process. At the same time, during the shimming process, without holding your breath, the interference of movement causes the magnetic sensitivity of human tissue to have a greater impact on the magnetic field. During the data acquisition and algorithm iteration process, multiple excitations and algorithm iterations are required to achieve the best results, so it takes a long time. For this reason, when the main magnetic field itself is severely non-uniform, the shimming parameters may not converge.

[0011] In order to obtain the uneven distribution of the magnetic field in the imaging area in three directions, three-dimensional or multi-layer two-dimensional imaging is required. The acquisition of accurate field maps requires more echo acquisitions and algorithm iterations, which corresponds to longer acquisition times. In the case of longer acquisition times, image distortion and unstable effects caused by patient tissue movement are more likely to occur. However, field maps that more accurately describe the uneven distribution of the magnetic field often correspond to better, more refined, and more suitable for high-order decomposition of uniform field effects.

[0012] According to the existing technology, it can be found that the traditional shimming technology is difficult to quickly acquire field images in areas where motion artifacts are prone to exist, such as abdominal imaging, and therefore it is difficult to obtain a good shimming effect. Summary of the invention

[0013] In view of the above-mentioned shortcomings of the prior art, an object of the present invention is to provide a method, system, terminal and medium for rapid magnetic resonance shimming, so as to solve the above-mentioned problems of the prior art.

[0014] To achieve the above-mentioned purpose and other related purposes, the present invention provides a method for rapid shimming of magnetic resonance, the method comprising: generating a field map template in a standard space based on a plurality of collected field map samples and a T1 structural image corresponding to each field map sample; acquiring a field map obtained by adjusting the parameters of a sequence and a T1 structural map corresponding to the field map and in the same space as the field map; obtaining a field map template in an image space by converting the T1 structural map to the standard space; based on the field map template in the image space, obtaining a predicted field map for accurately describing the magnetic field distribution in the same image space as the field map according to the field map; performing phase unwrapping on the predicted field map, and performing numerical optimization of shimming coils on the unwrapped predicted field map to complete shimming.

[0015] In one embodiment of the present invention, the generation of a field image template in a standard space based on multiple collected field image samples and the T1 structural image corresponding to each field image sample includes: based on multiple collected field image samples and the T1 structural image corresponding to each field image sample, converting and aligning each field image sample to construct a training database in a standard space; and using the training database to train an initial network to generate a field image template.

[0016] In one embodiment of the present invention, based on collecting multiple field map samples and the T1 structural image corresponding to each field map sample, converting and aligning each field map sample to construct a training database in a standard space includes: converting each field map sample to the standard space by converting each T1 structural image to the standard space; weighted aligning the amplitude image in each converted field map sample with the corresponding converted T1 structural image; weighted averaging each field map sample in the standard space after the weighted alignment to obtain each optimized field map sample; and constructing the training database using each field map sample and the corresponding optimized field map sample.

[0017] In one embodiment of the present invention, obtaining the field image template in the image space by converting the T1 structure diagram to the standard space includes: converting the T1 structure diagram to the standard space to obtain an inverse transformation relationship corresponding to converting the standard space to the image space; based on the inverse transformation relationship, converting the field image template into a field image template in the image space.

[0018] In one embodiment of the present invention, the initial network adopts a residual network.

[0019] In an embodiment of the present invention, each field image sample corresponds to a plurality of age groups and genders.

[0020] In one embodiment of the present invention, the sequence is a 2D and 3D gradient echo sequence or a DESS sequence.

[0021] To achieve the above-mentioned purpose and other related purposes, the present invention provides a magnetic resonance rapid shimming system, the system comprising: a field image template generation module, used to generate a field image template in a standard space based on a plurality of collected field image samples and a T1 structural image corresponding to each field image sample; a data acquisition module, connected to the field image template generation module, used to acquire a field image acquired by adjusting the parameters of a sequence and a T1 structural image corresponding to the field image and in the same space as the field image; a template conversion module, connected to the data acquisition module and the field image template generation module, used to acquire a field image template in an image space by converting the T1 structural image into the standard space; a predicted field image acquisition module, connected to the template conversion module, used to acquire a predicted field image in the same image space as the field image based on the field image template in the image space and for accurately describing the magnetic field distribution; and a shimming module, connected to the predicted field image acquisition module, used to perform phase unwinding on the predicted field image and perform numerical optimization of shimming coils on the unwound predicted field image to complete shimming.

[0022] To achieve the above-mentioned purpose and other related purposes, the present invention provides a magnetic resonance rapid shimming terminal, comprising: one or more memories and one or more processors; the one or more memories are used to store computer programs; the one or more processors are connected to the memories and are used to run the computer programs to execute the magnetic resonance rapid shimming method.

[0023] To achieve the above-mentioned object and other related objects, the present invention provides a computer-readable storage medium storing a computer program, wherein the computer program executes the magnetic resonance rapid shimming method when executed by one or more processors.

[0024] As described above, the present invention is a method, system, terminal and medium for rapid shimming of magnetic resonance, which has the following beneficial effects: the present invention can generate a field map template in a standard space by collecting multiple field map samples and the T1 structural image corresponding to each field map sample; and obtain a field map obtained by adjusting the parameters of the sequence and a T1 structural map corresponding to the field map and in the same space as the field map; obtain a field map template in the image space by converting the T1 structural map to the standard space; based on the field map template in the image space, obtain a predicted field map for accurately describing the magnetic field distribution in the same image space as the field map according to the field map; perform phase unwrapping on the predicted field map, and perform numerical optimization of the shimming coil on the unwrapped predicted field map to complete shimming. The present invention can realize a field map that accurately describes the magnetic field distribution in the imaging area and a fast shimming process, and has a better shimming effect, especially for the magnetic field inhomogeneity caused by the difference in magnetic sensitivity of different tissues. By simulating and predicting the field map, the iteration and repetition process in the shimming process are reduced, and the field map in the imaging area that more accurately describes the magnetic field distribution can be obtained in a shorter time; it can also, to a certain extent, perform better shimming effects on the imaging area when imaging tissues such as the abdomen and brain. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a schematic flow chart of a method for rapid magnetic resonance shimming in one embodiment of the present invention.

[0026] Figure 2 Shown is a schematic diagram of generating a field map template in an embodiment of the present invention.

[0027] Figure 3 Shown is a schematic diagram of the structure of a residual network in one embodiment of the present invention.

[0028] Figure 4 Shown is a schematic diagram of the conversion of the T1 structure diagram in an embodiment of the present invention.

[0029] Figure 5 It is a schematic diagram showing the effect of the shim coil on shimming in one embodiment of the present invention.

[0030] Figure 6 It is a schematic flow chart of a numerical optimization method for shim coils in one embodiment of the present invention.

[0031] Figure 7 Shown is a schematic diagram of the process of generating a scene map template in an embodiment of the present invention.

[0032] Figure 8 It is a schematic diagram of the shimming process in one embodiment of the present invention.

[0033] Fig. 9 It is a schematic structural diagram of a magnetic resonance rapid shimming system in one embodiment of the present invention.

[0034] Fig.10 Shown is a schematic structural diagram of a magnetic resonance rapid shimming terminal in one embodiment of the present invention. 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 in the following description, reference is made to the accompanying drawings, which describe several embodiments of the present invention. It should be understood that other embodiments may also be used, and that mechanical composition, structural, electrical and operational changes may be made without departing from the spirit and scope of the present invention. The following detailed description should not be considered restrictive, and the scope of the embodiments of the present invention is limited only by the claims of the published patents. The terms used herein are only for describing specific embodiments and are not intended to limit the present invention. Spatially related terms, such as "upper", "lower", "left", "right", "below", "below", "lower", "above", "upper", etc., may be used in the text to facilitate the description of the relationship between an element or feature shown in the figure and another element or feature.

[0037] Throughout the specification, when a part is said to be "connected" to another part, this includes not only the case of "direct connection" but also the case of "indirect connection" by placing other elements therebetween. In addition, when a part is said to "include" a certain constituent element, unless otherwise stated, it does not exclude other constituent elements, but means that other constituent elements may be included.

[0038] The terms first, second and third mentioned herein are used to describe various parts, components, regions, layers and / or segments, but are not limited thereto. These terms are only used to distinguish a certain part, component, region, layer or segment from other parts, components, regions, layers or segments. Therefore, the first part, component, region, layer or segment described below may refer to the second part, component, region, layer or segment within the scope of the present invention.

[0039] Furthermore, as used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless there is an indication to the contrary in the context. It should be further understood that the terms "comprise", "include" indicate the presence of the described features, operations, elements, components, items, kinds, and / or groups, but do not exclude the presence, occurrence or addition of one or more other features, operations, elements, components, items, kinds, and / or groups. The terms "or" and "and / or" used herein are interpreted as inclusive, or mean any one or any combination. Therefore, "A, B or C" or "A, B and / or C" means "any of the following: A; B; C; A and B; A and C; B and C; A, B and C". Exceptions to this definition will only occur when the combination of elements, functions or operations is inherently mutually exclusive in some way.

[0040] The invention discloses a method for rapid shimming of magnetic resonance. A field image template in a standard space can be generated by collecting multiple field image samples and T1 structural images corresponding to each field image sample; a field image obtained by adjusting the parameters of a sequence and a T1 structural image corresponding to the field image and in the same space as the field image are obtained; a field image template in the image space is obtained by converting the T1 structural image to the standard space; based on the field image template in the image space, a predicted field image for accurately describing the magnetic field distribution in the same image space as the field image is obtained according to the field image; the predicted field image is phase unwound, and the predicted field image after unwound is numerically optimized for shimming coils to complete shimming. The invention can realize a field image that accurately describes the magnetic field distribution in an imaging area and a rapid shimming process, especially for magnetic field inhomogeneity caused by differences in magnetic sensitivity of different tissues, and through simulation and prediction of the field image, the iteration and repetition process in the shimming process are reduced, so that a field image in the imaging area that more accurately describes the magnetic field distribution can be obtained in a shorter time; and better shimming can be performed on the imaging area when imaging tissues such as the abdomen and the brain to a certain extent.

[0041] The following is a detailed description of the embodiments of the present invention with reference to the accompanying drawings so that those skilled in the art can easily implement the present invention. The present invention can be embodied in many different forms and is not limited to the embodiments described herein.

[0042] like Figure 1A schematic flow chart showing a method for rapid magnetic resonance shimming in an embodiment of the present invention.

[0043] The method comprises:

[0044] Step S11: Generate a field image template in a standard space based on the collected multiple field image samples and the T1 structural image corresponding to each field image sample.

[0045] In detail, multiple field map samples are collected and T1 structural images are collected at the same time; wherein, when the T1 structural image (amplitude image) changes, the corresponding field map sample also changes, and each field map sample includes: an amplitude image and a phase image. Preferably, each field map sample can correspond to multiple age groups and genders, so that the adaptability of the generated template can be adjusted according to the actual needs of the scan. It should be noted that the standard space can be set according to demand. Generally speaking, different tissues have different standard spaces. For example, the standard space related to brain MR images includes the MNI standard space designed by the University of Montreal, etc.

[0046] Preferably, if rapid high-order MRI shimming of the abdomen is performed, a gating module is used during the T1 structural image acquisition process, that is, the T1 structural image is acquired using the gating module.

[0047] Optionally, step S11 includes: based on the collected multiple field image samples and the T1 structural image corresponding to each field image sample, converting and registering each field image sample to construct a training database in a standard space; and training an initial network using the training database to generate a field image template.

[0048] Optionally, based on collecting multiple field image samples and T1 structural images corresponding to each field image sample, each field image sample is converted and registered to construct a training database in a standard space, such as Figure 2 As shown, including:

[0049] Each field image sample is converted to the standard space by converting each T1 structural image to the standard space; specifically, the collected field image samples are normalized to the standard space by converting the T1 structural image to the standard space.

[0050] The amplitude image in each converted field image sample is weighted aligned with its corresponding converted T1 structural image; specifically, the amplitude image of each converted field image sample and the converted T1 structural image are weighted aligned using image registration.

[0051] The field map samples in the standard space after weighted alignment are weighted averaged to obtain the optimized field map samples, thereby offsetting the differences in local field distribution caused by different local fields and tissue activities of the same subject.

[0052] The training database is constructed using each field image sample and the corresponding optimized field image sample, wherein each field image sample is used as the input of the network, each optimized field image sample is used as the output of the network, and the initial network is trained.

[0053] Due to the use of standard space templates and image registration, the obtained field image template has higher credibility. And the T1 image is used to obtain structural information to identify the shape and position structure of the tissue, which further improves the robustness of the shimming method.

[0054] Optionally, the initial network may be other neural networks such as SRCNN, FSRCNN, and DRCN.

[0055] Optionally, the initial network uses a residual network to speed up the shimming process; Figure 3 As shown in Figure 1, it is a schematic diagram of the structure of the residual network.

[0056] Step S12: Acquire a field image obtained by adjusting the parameters of the sequence and a T1 structure image corresponding to the field image and in the same space as the field image.

[0057] Specifically, the parameters of the sequence are adjusted, the field image is acquired by the sequence, and the corresponding T1 structure image is acquired at the same time; wherein the resolution of the field image acquired in this way is relatively low. And the field image and the T1 structure image are in the same image space.

[0058] Optionally, if rapid high-order MRI field shimming of the abdomen is performed, respiratory gating is used to adjust sequence parameters to obtain field images.

[0059] Alternatively, the sequence may be a DESS (Dual Echo Steady State Free Precession) sequence or a 2D and 3D dual echo gradient echo sequence.

[0060] Step S13: Obtaining a field map template in the image space by converting the T1 structure map into the standard space.

[0061] Optional, such as Figure 4 As shown, step S13 includes:

[0062] The T1 structure diagram is converted to the standard space to obtain an inverse transformation relationship corresponding to the conversion of the standard space to the image space; specifically, the T1 structure diagram in the image space is converted to the standard space, and the inverse transformation relationship is calculated based on the T1 structure diagram in the image space and the T1 structure diagram in the standard space, which relationship can convert the standard space to the image space.

[0063] Based on the inverse transformation relationship, the field map template is converted into a field map template in the image space. Specifically, based on the inverse transformation relationship, the field map template is converted back to the image space and input into the residual network for predicting the field map. It should be noted that the input of the field map template in the image space is a collected low-resolution field map, and the output is a predicted field map that accurately describes the magnetic field distribution in the imaging area; and the input field map and the predicted field map belong to the same image space.

[0064] Step S14: Based on the field map template in the image space, a predicted field map for accurately describing the magnetic field distribution in the same image space as the field map is obtained according to the field map.

[0065] Specifically, the field map is input into a field map template in the image space, and a predicted field map for accurately describing the magnetic field distribution in the same image space as the field map is output.

[0066] Step S15: performing phase unwrapping on the predicted field map, and performing numerical optimization of shimming coils on the unwrapped predicted field map to complete shimming.

[0067] Optionally, if the TE difference between the two echoes is large, the phase may be wrapped, and phase unwrapping is required. That is, if the echo time difference between the two echoes in the sequence is large, phase unwrapping is required. Preferably, the phase unwrapping is performed using algorithms such as the 3D optimal path method, the regional growing method, or the optimal path method.

[0068] Optionally, the shimming coil is numerically optimized for the unwound field map, and the shimming coil value is obtained after numerical optimization, and the current value of the shimming coil is applied to complete the shimming process. It should be noted that the shimming coil can adopt common first-order and second-order shimming coils. The effect of the shimming coil on shimming can be seen in Figure 5 .

[0069] Common methods for numerical optimization of shim coils, such as Figure 6 As shown, it includes: field map measurement, superconducting shim coil parameter input, optimization model establishment, parameter optimization and outputting the shim coil when all constraints are met.

[0070] In order to better illustrate the above-mentioned magnetic resonance rapid shimming method, the present invention provides the following specific embodiments.

[0071] Embodiment 1: A method for rapid high-order abdominal magnetic resonance shimming.

[0072] Since the tissue distribution in the abdomen of most humans is relatively similar, the B0 field distribution is also relatively similar. This scheme intends to perform rapid high-order MRI field shimming of the abdomen based on field map template prediction, using structural images and field maps, utilizing gating modules during structural image acquisition, and performing standard space template conversion through image registration. The residual learning method is used to predict and simulate the field map, reducing the iteration and repetition process in the field shimming process, and achieving ultra-fast B0 field shimming.

[0073] The whole scheme is divided into two parts: field map template generation and field uniformity process.

[0074] 1. Generation of field map template;

[0075] The generation of the scene map template is initially divided into four steps. The specific process is shown in Figure 7 :

[0076] (1) The data collection phase will confirm that the designed data can cover all age groups and genders.

[0077] (2) Use the gating module to acquire T1 structural images, and use the T1 structural images to normalize the acquired data to the standard space.

[0078] (3) Use image registration to weighted align the amplitude image of the field map and the T1 structural image.

[0079] (4) The field map in the standard space is weighted averaged to offset the differences in local field distribution caused by different local fields and tissue activities of the same subject, and is added to the database as a training set for the deep learning network (here a residual network is used) to obtain a field map template.

[0080] 2. Field shimming process;

[0081] The shimming process is divided into three parts: field map acquisition, phase unwinding and shimming coil parameter optimization, such as Figure 8 As shown:

[0082] (1) Use respiratory gating to adjust sequence parameters and use the sequence to obtain field images. The sequence can be a DESS (dual echo steady-state free precession) sequence or a 2D and 3D dual echo gradient echo sequence. Obtain the T1 structure image, convert it to the standard space, and then use the inverse transformation of the registration of the structure and standard space to convert the field image template back to the image space, input it into the residual network, and predict the field image.

[0083] (2) If the echo time difference between the two echoes in the sequence is large, phase unwrapping is required, and algorithms such as the region growing method or the optimal path method can be used.

[0084] (3) The shimming coil is numerically optimized for the unwrapped field diagram, and the shimming coil value is obtained after the numerical optimization, and the current value of the shimming coil is applied to complete the shimming process.

[0085] Compared with the traditional field shimming method, the problems solved by this embodiment are mainly the following: In the traditional field shimming process, it is impossible to have both a field map that accurately describes the magnetic field distribution in the imaging area and a fast field shimming process. By simulating and predicting the field map, the iteration and repetition process in the field shimming process are reduced, and a field map in the imaging area that more accurately describes the magnetic field distribution can be obtained in a shorter time. Magnetic resonance imaging is often used for tissue imaging in which there is organ movement, such as the abdomen. However, in the process of quantitative analysis of each organ in this type of tissue, the participation of the fat suppression module is required, and the magnetic susceptibility of the human body itself changes due to tissue movement caused by breathing, resulting in localized magnetic field distortion. Since different tissues have different magnetic sensitivities and the tissues are constantly moving, image distortion caused by movement is prone to occur, making it difficult to correct magnetic field inhomogeneity.

[0086] Based on the above problems, this embodiment has a better uniform effect on the magnetic field inhomogeneity caused by the difference in magnetic sensitivity of different tissues. By simulating and predicting the field map, the iteration and repetition process in the field uniformity process is reduced, and the field map in the imaging area that more accurately describes the magnetic field distribution can be obtained in a shorter time. For the imaging area when imaging abdominal tissue, the field uniformity is better.

[0087] Similar to the principle of the above embodiment, the present invention provides a magnetic resonance rapid shimming system.

[0088] The following provides specific embodiments in conjunction with the accompanying drawings:

[0089] like Fig. 9 A schematic structural diagram of a magnetic resonance rapid shimming system in an embodiment of the present invention is shown.

[0090] The system comprises:

[0091] A field image template generating module 91 is used to generate a field image template in a standard space based on a plurality of collected field image samples and a T1 structural image corresponding to each field image sample;

[0092] A data acquisition module 92, connected to the field image template generation module 91, is used to acquire a field image obtained by adjusting the parameters of the sequence and a T1 structure image corresponding to the field image and in the same space as the field image;

[0093] A template conversion module 93, connected to the data acquisition module 92 and the field image template generation module 91, is used to obtain the field image template in the image space by converting the T1 structure diagram into the standard space;

[0094] A predicted field map acquisition module 94, connected to the template conversion module 93, is used to obtain a predicted field map for accurately describing the magnetic field distribution in the same image space as the field map based on the field map template in the image space;

[0095] The shimming module 95 is connected to the predicted field map acquisition module 94 and is used to perform phase unwinding on the predicted field map and perform shimming coil numerical optimization on the unwound predicted field map to complete shimming.

[0096] Optionally, the field image template generation module 91 is used to transform and align each field image sample based on the collected multiple field image samples and the T1 structural image corresponding to each field image sample to construct a training database in the standard space; use the training database to train the initial network to generate a field image template

[0097] Optionally, based on collecting multiple field map samples and the T1 structural image corresponding to each field map sample, converting and aligning each field map sample to construct a training database in the standard space includes: converting each field map sample to the standard space by converting each T1 structural image to the standard space; weighted aligning the amplitude image in each converted field map sample with its corresponding converted T1 structural image; weighted averaging each field map sample in the standard space after the weighted alignment to obtain each optimized field map sample; and constructing the training database using each field map sample and the corresponding optimized field map sample.

[0098] Optionally, the template conversion module 93 is used to convert the T1 structure diagram to the standard space to obtain an inverse transformation relationship corresponding to converting the standard space to the image space; based on the inverse transformation relationship, convert the field image template into a field image template under the image space.

[0099] like Fig.10 A schematic structural diagram of a magnetic resonance rapid shimming terminal 100 in an embodiment of the present invention is shown.

[0100] The magnetic resonance fast shimming terminal 100 comprises: a memory 101 and a processor 102. The memory 101 is used to store computer programs; the processor 102 runs the computer programs to implement the following Figure 2 The magnetic resonance rapid shimming method.

[0101] Optionally, the number of the memory 101 can be one or more, the number of the processor 102 can be one or more, and Fig.10 Take one as an example.

[0102] Optionally, the processor 102 in the magnetic resonance rapid shimming terminal 100 will perform the following operations: Figure 1In the above steps, one or more instructions corresponding to the process of the application are loaded into the memory 101, and the processor 102 runs the application stored in the first memory 101, thereby achieving the following Figure 2 Various functions in the magnetic resonance fast shimming method.

[0103] Optionally, the memory 101 may include but is not limited to high-speed random access memory and non-volatile memory. For example, one or more disk storage devices, flash memory devices or other non-volatile solid-state storage devices; the processor 102 may include but is not limited to a central processing unit (CPU), a network processor (NP), etc.; it may 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.

[0104] Optionally, the processor 102 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 gate or transistor logic devices, discrete hardware components.

[0105] The present invention also provides a computer-readable storage medium storing a computer program, wherein the computer program is executed as follows: Figure 1The magnetic resonance fast shimming method shown. The computer-readable storage medium may include, but is not limited to, a floppy disk, an optical disk, a CD-ROM (Compact Disc Read Only Memory), a magneto-optical disk, a ROM (Read Only Memory), a RAM (Random Access Memory), an EPROM (Erasable Programmable Read Only Memory), an EEPROM (Electrically Erasable Programmable Read Only Memory), a magnetic card or an optical card, a flash memory, or other types of media / machine-readable media suitable for storing machine-executable instructions. The computer-readable storage medium may be a product that is not connected to a computer device, or may be a component that is connected to a computer device for use.

[0106] In summary, the magnetic resonance rapid shimming method, system, terminal and medium of the present invention can generate a field image template in a standard space through the collected multiple field image samples and the T1 structural image corresponding to each field image sample; and obtain the field image obtained by adjusting the parameters of the sequence and the T1 structural image corresponding to the field image and in the same space as the field image; obtain the field image template in the image space by converting the T1 structural image to the standard space; based on the field image template in the image space, obtain a predicted field image for accurately describing the magnetic field distribution in the same image space as the field image according to the field image; perform phase unwrapping on the predicted field image, and perform numerical optimization of the shimming coil on the unwrapped predicted field image to complete shimming. The present invention can realize a field map that accurately describes the magnetic field distribution in the imaging area and a fast shimming process, especially for the magnetic field inhomogeneity caused by the difference in magnetic sensitivity of different tissues, it has a better shimming effect, and through the simulation and prediction of the field map, it reduces the iteration and repetition process in the shimming process, and can obtain a field map in the imaging area that more accurately describes the magnetic field distribution in a shorter time; it can also perform better shimming effects on the imaging area when imaging tissues such as the abdomen and brain to a certain extent. Therefore, the present invention effectively overcomes various shortcomings in the prior art and has a high industrial utilization value.

[0107] 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 technology may modify or change the above embodiments without violating the spirit and scope of the present invention. Therefore, all equivalent modifications or changes made by a person of ordinary skill in the art without departing from the spirit and technical ideas disclosed by the present invention shall still be covered by the claims of the present invention.

Claims

1. A method for rapid shimming of magnetic resonance, characterized in that: The method comprises: Based on the collected multiple field image samples and the T1 structural image corresponding to each field image sample, a field image template in a standard space is generated; Acquire a field map obtained by adjusting the parameters of the sequence and a T1 structure map corresponding to the field map and in the same space as the field map; Obtaining a field map template in an image space by converting the T1 structure map into the standard space; Based on the field map template in the image space, obtaining a predicted field map for accurately describing the magnetic field distribution in the same image space as the field map according to the field map; Phase unwinding is performed on the predicted field map, and the unwound predicted field map is numerically optimized for shimming coils to complete shimming; The step of generating a field image template in a standard space based on the collected multiple field image samples and the T1 structural image corresponding to each field image sample comprises: based on the collected multiple field image samples and the T1 structural image corresponding to each field image sample, converting and registering each field image sample to construct a training database in a standard space; and training an initial network using the training database to generate a field image template; The method of obtaining the field image template in the image space by converting the T1 structure diagram to the standard space includes: converting the T1 structure diagram to the standard space to obtain an inverse transformation relationship corresponding to converting the standard space to the image space; based on the inverse transformation relationship, converting the field image template into a field image template in the image space.

2. The magnetic resonance rapid shimming method according to claim 1, characterized in that: The method of acquiring a plurality of field image samples and T1 structural images corresponding to each field image sample, converting and registering each field image sample to construct a training database in a standard space includes: Converting each field image sample to the standard space by converting each T1 structure image to the standard space; Perform weighted alignment on the magnitude image in each converted field image sample and its corresponding converted T1 structural image; Perform weighted averaging on each field image sample in the standard space after weighted alignment to obtain each optimized field image sample; The training database is constructed using each field image sample and the corresponding optimized field image sample.

3. The magnetic resonance rapid shimming method according to claim 1, characterized in that: The initial network adopts a residual network.

4. The magnetic resonance rapid shimming method according to claim 1, characterized in that: Each scene image sample corresponds to multiple age groups and genders.

5. The magnetic resonance rapid shimming method according to claim 1, characterized in that: The sequences are 2D and 3D gradient echo sequences or DESS sequences.

6. A magnetic resonance rapid shimming system, characterized in that: The system comprises: A field image template generation module, used to generate a field image template in a standard space based on a plurality of collected field image samples and a T1 structural image corresponding to each field image sample; A data acquisition module, connected to the field map template generation module, for acquiring a field map obtained by adjusting the parameters of the sequence and a T1 structure map corresponding to the field map and in the same space as the field map; A template conversion module, connected to the data acquisition module and the field image template generation module, for obtaining a field image template in an image space by converting the T1 structure diagram into the standard space; A predicted field map acquisition module, connected to the template conversion module, for obtaining a predicted field map for accurately describing magnetic field distribution in the same image space as the field map based on the field map template in the image space; A shimming module, connected to the predicted field map acquisition module, is used to perform phase unwinding on the predicted field map, and perform shimming coil numerical optimization on the unwound predicted field map to complete shimming; The step of generating a field image template in a standard space based on the collected multiple field image samples and the T1 structural image corresponding to each field image sample comprises: based on the collected multiple field image samples and the T1 structural image corresponding to each field image sample, converting and registering each field image sample to construct a training database in a standard space; and training an initial network using the training database to generate a field image template; The method of obtaining the field image template in the image space by converting the T1 structure diagram to the standard space includes: converting the T1 structure diagram to the standard space to obtain an inverse transformation relationship corresponding to converting the standard space to the image space; based on the inverse transformation relationship, converting the field image template into a field image template in the image space.

7. A magnetic resonance fast shimming terminal, characterized in that: include: one or more memories and one or more processors; The one or more memories are used to store computer programs; The one or more processors, connected to the memory, are configured to run the computer program to perform the method according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that: A computer program is stored, and when the computer program is executed by one or more processors, the method according to any one of claims 1 to 5 is executed.

Images