Cardiac movie imaging method and magnetic resonance imaging equipment

Abstract

The invention discloses a heart film imaging method and magnetic resonance imaging equipment, and relates to the technical field of magnetic resonance imaging. According to the method, retrospective gating is adopted to obtain under-sampled K-t data of the heart in each cardiac cycle, and the actual phase number and the ideal phase number of each cardiac cycle may be different. In order to solve the problem that the actual phase number and the ideal phase number may be different, the method obtains an interpolation relation of each cardiac cycle, obtains a sampling operator of under-sampling K-t data based on the interpolation relation, constructs an image reconstruction model based on the sampling operator and the under-sampling K-t data, and solves the image reconstruction model so as to obtain an image reconstruction result. And obtaining a heart image. According to the method provided by the invention, the retrospective gating technology and the K-t compressed sensing algorithm are innovatively combined, so that more efficient data acquisition is realized, and the acceleration multiple is improved.

Description

Technical Field

[0001] The present application relates to the technical field of magnetic resonance imaging, and in particular to a cardiac movie imaging method and a magnetic resonance imaging device. Background Art

[0002] Cardiac cine imaging is a crucial tool in medical imaging for assessing cardiac function and structure. By dynamically capturing multiple frames of images throughout the cardiac cycle, cardiac cine imaging visually demonstrates the heart's contraction and relaxation processes, assisting physicians in diagnosis. Currently, retrospectively gated cardiac cine imaging, with its ability to accurately capture motion information throughout the entire cardiac cycle, has become a mainstream method in clinical practice.

[0003] However, retrospective gated cine imaging requires long, continuous acquisition of data from multiple cardiac cycles, resulting in a long scan time for a single scan, which reduces examination efficiency. To shorten the scan time, acceleration technology is needed to accelerate imaging. In related technologies, acceleration technology for retrospective cine imaging is typically a method that only accelerates K-space. However, the maximum acceleration factor of this acceleration method is generally around 3 times, which is low and results in poor acceleration effect. Summary of the Invention

[0004] This application provides a cardiac cine imaging method and magnetic resonance imaging (MRI) device, which can solve the problem of poor acceleration effect in related technologies. The technical solution is as follows:

[0005] In one aspect, a method for cardiac cine imaging is provided, the method comprising:

[0006] Using retrospective gating, undersampled Kt data of the heart in each cardiac cycle is obtained, wherein the actual number of phases in each cardiac cycle is the same as or different from the ideal number of phases;

[0007] For each of the cardiac cycles, an interpolation relationship of the cardiac cycle is obtained, where the interpolation relationship is used to interpolate first image data of the heart in the cardiac cycle into second image data, where the number of phases of the undersampled Kt data corresponding to the first image data is the ideal number of phases, and the number of phases of the undersampled Kt data corresponding to the second image data is the actual number of phases of the cardiac cycle;

[0008] Based on the interpolation relationship, a sampling operator of the under-sampled Kt data is acquired, where the sampling operator is used to describe a mapping relationship between the under-sampled Kt data and the magnetic resonance signal of the heart;

[0009] An image reconstruction model is constructed based on the sampling operator and the under-sampled Kt data, and the image reconstruction model is solved to obtain a cardiac image.

[0010] Optionally, obtaining the interpolation relationship of the cardiac cycle includes:

[0011] For each first sampling point in the actual phase number of the cardiac cycle, determining a position of the first sampling point in a sampling interval where the second sampling points in the ideal phase number of the cardiac cycle are located;

[0012] determining, based on a position of the first sampling point in the sampling interval, a weight of each of the second sampling points when mapping the first sampling point to the position;

[0013] An interpolation relationship of the cardiac cycle is obtained based on the weights of each of the second sampling points.

[0014] Optionally, determining, based on the position of the first sampling point in the sampling interval, the weight of each second sampling point when mapping the first sampling point to the position includes:

[0015] determining, based on the position of the first sampling point in the sampling interval, by interpolation, a weight of each of the second sampling points when mapping the first sampling point to the position;

[0016] The interpolation method is linear interpolation, polynomial interpolation or spline interpolation.

[0017] Optionally, obtaining a sampling operator for the undersampled Kt data based on the interpolation relationship includes:

[0018] Acquiring a sampling trajectory of the cardiac cycle and a sensitivity matrix of the channel coil;

[0019] A sampling operator for the under-sampled Kt data is obtained based on the sampling trajectory, the sensitivity matrix, and the interpolation relationship.

[0020] In each cardiac cycle, the number of the multiple K lines collected in any two phases is the same, and the positions of the multiple K lines in any two phases in the K space are not completely the same.

[0021] Optionally, constructing an image reconstruction model based on the sampling operator and the undersampled Kt data includes:

[0022] Obtaining solution constraints for the image reconstruction model;

[0023] The image reconstruction model is constructed based on the solution constraint conditions, the sampling operator and the under-sampling Kt data.

[0024] Optionally, the solution constraint conditions include: at least one of: total variation constraint, wavelet transform constraint, time-dimensional Fourier transform constraint and nuclear norm constraint.

[0025] Optionally, constructing an image reconstruction model based on the sampling operator and the undersampled Kt data includes:

[0026] The image reconstruction model is solved using the FISTA algorithm to obtain the heart image.

[0027] On the other hand, a magnetic resonance imaging device is provided, which 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 cardiac movie imaging method described in the above aspect is implemented.

[0028] On the other hand, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the cardiac movie imaging method as described in the above aspects is implemented.

[0029] On the other hand, a computer program product is provided, which includes a computer program or computer instructions, and when the computer program or the computer instructions are executed by a processor, the cardiac movie imaging method as described in the above aspects is implemented.

[0030] The beneficial effects of the technical solution provided by this application include at least:

[0031] The present application provides a cardiac movie imaging method and MRI equipment, which can use retrospective gating to obtain undersampled Kt data of the heart in each cardiac cycle, and the actual number of phases in the cardiac cycle is the same as or different from the ideal number of phases. The method can also obtain the interpolation relationship of each cardiac cycle, and derive the sampling operator of the undersampled Kt data based on the interpolation relationship, and then construct an image reconstruction model based on the sampling operator and the undersampled Kt data, and solve the image reconstruction model to obtain a cardiac image. The interpolation relationship is used to interpolate the first image data corresponding to the ideal number of phases in the cardiac cycle into the second image data corresponding to the actual number of phases. It can be seen that the method provided in the embodiment of the present application can innovatively combine retrospective gating technology with the Kt compressed sensing algorithm to achieve more efficient data acquisition, thereby effectively improving the acceleration factor. Moreover, in retrospective gating, the actual phase data of each cardiac cycle may be different from the ideal phase data due to heart rate fluctuations. Therefore, the embodiment of the present application innovatively introduces an interpolation relationship in the sampling operator, thereby effectively bridging the difference between the uniform ideal phase distribution and the non-uniform actual phase distribution, realizing the combination of retrospective gating and Kt compressed sensing method, so that the compressed sensing algorithm can be applied to the accelerated scenario of retrospective cardiac movie imaging, and then image reconstruction based on the undersampled Kt data collected by retrospective gating can be realized.

[0032] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 This is a flow chart of a cardiac cine imaging method provided in an embodiment of the present application;

[0034] Figure 2 is a flow chart of another cardiac cine imaging method provided in an embodiment of the present application;

[0035] Figure 3 This is a schematic diagram of a sampling trajectory within a cardiac cycle provided by an embodiment of the present application;

[0036] Figure 4 is a schematic diagram of another sampling trajectory within a cardiac cycle provided by an embodiment of the present application;

[0037] Figure 5 FIG. 1 is a schematic diagram of a heart image reconstructed when acceleration is performed only in k-space and the acceleration factor is 3 in the related art;

[0038] Figure 61 is a schematic diagram of a reconstructed cardiac image obtained by using the method provided in an embodiment of the present application at an acceleration factor of 8;

[0039] Figure 7 It is a schematic diagram of a heart image obtained based on full acquisition data;

[0040] Figure 8 This is a structural diagram of an MRI device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0041] The following describes in detail embodiments of the present application, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0042] The present application provides a cardiac movie imaging method, which is applied to MRI equipment. Figure 1 , the method comprising:

[0043] Step 101: Use retrospective gating to obtain undersampled Kt data of the heart in each cardiac cycle.

[0044] In an embodiment of the present application, the MRI device can use retrospective gating technology to spatially encode the magnetic resonance signals of the heart through a gradient magnetic field in the time dimension, and then undersample the encoded magnetic resonance signals, and fill the collected data into the K space according to the cardiac cycle sequence, thereby obtaining undersampled Kt data for each cardiac cycle.

[0045] The duration of the cardiac cycle may vary due to changes in heart rate, resulting in the actual number of cardiac phases being the same as or different from the preset ideal number of phases. The actual number of phases is the total number of time periods obtained by evenly dividing the cardiac cycle into multiple time periods. Each time period can reflect a state of the heartbeat, that is, each time period is a phase (also called a time phase) of the heart.

[0046] Step 102: For each cardiac cycle, obtain the interpolation relationship of the cardiac cycle.

[0047] The interpolation relationship is used to interpolate the first image data of the heart in the cardiac cycle into the second image data. The number of phases of the undersampled Kt data corresponding to the first image data is the ideal number of phases, and the number of phases of the undersampled Kt data corresponding to the second image data is the actual number of phases.

[0048] In retrospective gating technology, heart rate fluctuations may cause the actual number of phases in each cardiac cycle to be inconsistent. This non-uniform distribution of phases makes it difficult for traditional compressed sensing algorithms to be directly applied to the accelerated scenario of retrospective cardiac cine imaging. Based on this, for each cardiac cycle, the MRI device can obtain the interpolation relationship of the cardiac cycle to obtain a sampling operator, so that when solving the cardiac image, the number of acquired phases is ensured to be the same as the ideal number of phases. The number of acquired phases refers to the number of phases of the theoretical sampling data obtained by sampling the cardiac image to be solved based on the sampling operator.

[0049] In this way, the difference between the ideal uniform phase distribution and the actual non-uniform phase distribution can be effectively bridged, thereby realizing the combination of retrospective gating technology and Kt compressed sensing method, and effectively improving the acceleration effect of retrospective cardiac cine imaging.

[0050] Step 103: Based on the interpolation relationship, obtain the sampling operator of the undersampled Kt data.

[0051] For each cardiac cycle, the MRI device can obtain a sampling trajectory for sampling the heart during that cardiac cycle. The MRI device can then obtain a sampling operator for the undersampled Kt data for that cardiac cycle based on the sampling trajectory, the sensitivity matrix of the channel coil, and the interpolation relationship between the cardiac cycles. The sampling operator can be used to describe the mapping relationship between the undersampled Kt data and the cardiac magnetic resonance signal.

[0052] Since the actual number of phases of the cardiac cycle in the embodiment of the present application may be different from the ideal number of phases, and since the sampling operator can be determined based on the interpolation relationship, it can be ensured that the number of phases of the theoretical sampling data obtained based on the sampling operator matches the actual number of phases of the undersampled Kt data actually collected, thereby solving the cardiac image.

[0053] Step 104: construct an image reconstruction model based on the sampling operator and the under-sampled Kt data, and solve the image reconstruction model to obtain a cardiac image.

[0054] In an embodiment of the present application, an MRI device can construct an image reconstruction model based on a sampling operator, undersampled Kt data, and solved constraints. The MRI device can then solve the image reconstruction model to obtain a cardiac image. The sampling operator and undersampled Kt data constitute the fidelity terms of the image reconstruction model, and the solved constraints constitute the constraint terms of the image reconstruction model.

[0055] Compressed sensing technology is based on the theory of signal sparsity. Through nonlinear reconstruction algorithms, it can accurately restore the original image from undersampled data that is far below the requirements of the Nyquist sampling theorem. In the field of medical imaging, cardiac cine imaging aims to capture the periodic motion of the heart. Its core is to assume that each cardiac cycle is equally divided into a fixed number of time points (i.e., phases). Taking the common 20 phases as an example, they completely cover the entire physiological process of the heart from end-diastole to end-systole and then back to end-diastole. However, in the actual acquisition process, the human heart rate is in dynamic change due to the influence of individual physiological state and external factors. This leads to differences in the number of phases actually obtained in each acquisition cycle. The inconsistency in the number of phases will destroy the data consistency constraint, making compressed sensing technology unable to be directly applied to cardiac cine imaging.

[0056] In summary, the embodiment of the present application provides a cardiac movie imaging method, which can use retrospective gating technology to obtain undersampled Kt data of the heart in each cardiac cycle, and the actual number of phases of the cardiac cycle is the same as or different from the ideal number of phases. The method can also obtain the interpolation relationship of each cardiac cycle, and derive the sampling operator of the undersampled Kt data based on the interpolation relationship, and then construct an image reconstruction model based on the sampling operator and the undersampled Kt data, and solve the image reconstruction model to obtain a cardiac image. The interpolation relationship is used to interpolate the first image data corresponding to the ideal number of phases in the cardiac cycle into the second image data corresponding to the actual number of phases. It can be seen that the method provided by the embodiment of the present application can innovatively combine retrospective gating technology with the Kt compressed sensing algorithm to achieve more efficient data acquisition, thereby effectively improving the acceleration factor. Moreover, in retrospective gating, the actual phase data of each cardiac cycle may be different from the ideal phase data due to heart rate fluctuations. Therefore, the embodiment of the present application innovatively introduces an interpolation relationship in the sampling operator, thereby effectively bridging the difference between the uniform ideal phase distribution and the non-uniform actual phase distribution, realizing the combination of retrospective gating and Kt compressed sensing method, so that the compressed sensing algorithm can be applied to the accelerated scenario of retrospective cardiac movie imaging, and then image reconstruction based on the undersampled Kt data collected by retrospective gating can be realized.

[0057] Figure 2 This is another cardiac movie imaging method provided in the embodiment of the present application, which can be applied to MRI equipment. Figure 2 , the method may include:

[0058] Step 201: Use retrospective gating to obtain undersampled Kt data of the heart in each cardiac cycle.

[0059] In an embodiment of the present application, the MRI device may employ retrospective gating technology to spatially encode the cardiac magnetic resonance signals in the time dimension using a gradient magnetic field, then undersample the encoded magnetic resonance signals and fill the acquired data into K-space according to the cardiac cycle time sequence, thereby obtaining undersampled Kt data for each cardiac cycle. The undersampled Kt data for each cardiac cycle may include undersampled Kt sub-data acquired by each channel coil within that cardiac cycle.

[0060] During an actual scan, the heart rate typically fluctuates, causing the duration of the cardiac cycle to vary. Consequently, the actual number of cardiac phases may or may not be the same as the preset ideal number of phases. The actual number of phases refers to the total number of time periods obtained by evenly dividing the cardiac cycle into multiple time periods. Each time period can reflect a state of the heartbeat, meaning each time period is a phase of the heart. The cardiac cycle is the time interval between two adjacent R waves in an electrocardiogram.

[0061] In an embodiment of the present application, the MRI device can obtain a heartbeat signal while processing the magnetic resonance signal of the heart. The MRI device can then distribute the undersampled Kt data to each cardiac cycle based on the heartbeat signal to obtain undersampled Kt data within each cardiac cycle. The heartbeat signal can be determined based on an electrocardiogram (ECG), which can be collected by an electrocardiogram (ECG) machine and sent to the MRI device.

[0062] For example, Figure 3 A schematic diagram showing a sampling trajectory within a cardiac cycle (ie, RR interval) is shown. Figure 3 The white horizontal line in the middle indicates sampling. Figure 4 Schematic diagram of another sampling trajectory within a cardiac cycle provided by an embodiment of the present application. Figure 4 It can be seen that the actual number of phases in this cardiac cycle is 32. Furthermore, along the phase encoding direction, four K-lines can be collected in each phase. That is, the MRI device performs undersampling in the phase encoding direction, and the number of K-lines collected in any two phases in the undersampling data is the same.

[0063] Figure 4 Along the phase encoding direction, the number in each phase can represent the position of the K-line. Therefore, it can be concluded that the positions of the K-lines collected in any two phases in K space are not exactly the same.

[0064] Step 202: For each cardiac cycle, obtain the interpolation relationship of the cardiac cycle.

[0065] This interpolation relationship is used to interpolate the first image data of the heart during the cardiac cycle into the second image data. The number of phases of the undersampled Kt data corresponding to the first image data is the ideal number of phases, and the number of phases of the undersampled Kt data corresponding to the second image data is the actual number of phases. Thus, this interpolation relationship is used for interpolation in the image domain.

[0066] In retrospective gating technology, heart rate fluctuations may cause the actual number of phases in each cardiac cycle to be inconsistent. This non-uniform distribution of phases makes it difficult to directly apply traditional compressed sensing algorithms to the accelerated scenario of retrospective cardiac cine imaging. Based on this, for each cardiac cycle, the MRI device can obtain the interpolation relationship of the cardiac cycle to obtain a sampling operator, so that when solving the cardiac image, the number of acquired phases is guaranteed to be the same as the ideal number of phases. The number of acquired phases refers to the number of phases of the theoretical sampling data obtained by sampling the cardiac image to be solved based on the sampling operator.

[0067] In this way, the difference between the ideal uniform phase distribution and the actual non-uniform phase distribution can be effectively bridged, thereby realizing the combination of retrospective gating technology and Kt compressed sensing method, and effectively improving the acceleration effect of retrospective cardiac cine imaging.

[0068] It is understandable that, since the sampling trajectories of the channel coils are the same, the interpolation relationship corresponding to any two channel coils is the same in each cardiac cycle. Therefore, there is no need to consider the channel coils when determining the interpolation relationship in each cardiac cycle.

[0069] In the embodiment of the present application, for each cardiac cycle, the process of the MRI device acquiring the interpolation relationship of the cardiac cycle may include:

[0070] Step S1: for each first sampling point in the actual number of first sampling points of the cardiac cycle, determine the position of the first sampling point in the sampling interval where the ideal number of second sampling points is located.

[0071] In an embodiment of the present application, the MRI device can first directly map the first first sampling point among the actual number of first sampling points in the phase to the first second sampling point among the ideal number of second sampling points in the phase; and directly map the last first sampling point among the actual number of first sampling points in the phase to the last sampling point among the ideal number of second sampling points in the phase. The MRI device can then determine the position of each first sampling point within the sampling interval where the multiple second sampling points are located.

[0072] The first sampling point is one of the actual number of phases included in the cardiac cycle. The second sampling point is one of the ideal number of phases included in the cardiac cycle under ideal conditions. The sampling interval is an interval in which the first sampling point is the first second sampling point and the last sampling point is the last second sampling point, and the length is the interval of the cardiac cycle. The position of the first first sampling point is the position of the first second sampling point among the ideal number of second sampling points. The position of the last first sampling point is the position of the last second sampling point among the ideal number of second sampling points.

[0073] For each first sampling point except the first first sampling point and the last first sampling point, the position of the first sampling point may be the position of a second sampling point except the first second sampling point and the last second sampling point, or the position of the first sampling point may be between two adjacent second sampling points.

[0074] Step S2: Based on the position of the first sampling point in the sampling interval, determine the weights of each second sampling point when mapping the first sampling point to the position.

[0075] In an embodiment of the present application, the MRI device can determine the weights of each second sampling point when mapping the first sampling point to the position of the first sampling point in the sampling interval by interpolation. Optionally, the interpolation method can be linear interpolation, polynomial interpolation, or spline interpolation.

[0076] Taking the interpolation method as a linear interpolation method as an example, the process of determining the weight of each second sampling point by the MRI device is exemplified as follows:

[0077] In the case where the position of the first sampling point is the position of a second sampling point, the MRI device may determine that the weight of the second sampling point is 1, and the weight of each second sampling point except the second sampling point is 0.

[0078] If the location of the first sampling point is not the location of any second sampling point, the MRI device can determine two second sampling points adjacent to the location of the first sampling point. The MRI device can then determine that the weight of any second sampling point other than these two second sampling points can be 0. For each of these two second sampling points, the MRI device can also calculate the distance from the second sampling point to the location and determine the weight of the second sampling point based on the distance. The weight of the second sampling point is negatively correlated with the distance from the second sampling point to the location. In other words, the closer the distance, the greater the weight of the second sampling point.

[0079] Step S3: Obtain an interpolation relationship of the cardiac cycle based on the weights of the second sampling points.

[0080] Optionally, the interpolation relationship may be represented by an interpolation matrix.

[0081] The MRI device can organize the weights of each second sampling point when mapping each first sampling point to a corresponding position in the sampling interval into a matrix format, thereby obtaining the interpolation matrix. Each row of the interpolation matrix corresponds to a first sampling point, and each column corresponds to the weight of a second sampling point when mapping the first sampling point to the corresponding position.

[0082] It is understandable that if the actual number of phases in each cardiac cycle is the same as the ideal number of phases, the interpolation matrix determined by the MRI device may be a unit matrix.

[0083] Step 203: Acquire the sampling trajectory of each cardiac cycle.

[0084] When determining the sampling operator, the sampling trajectory during the sampling process is typically considered. Therefore, the MRI device can also obtain a sampling trajectory for each cardiac cycle. This sampling trajectory refers to the path along which the MRI device acquires data points in K-space, and the acquisition trajectory is determined by the actual undersampled Kt data collected. In other words, after obtaining the undersampled Kt data, the MRI device can then obtain a sampling trajectory based on this undersampled Kt data.

[0085] Optionally, the sampling trajectory can generally be represented by a matrix.

[0086] Step 204: Obtain the sensitivity matrix of the channel coil.

[0087] When determining the sampling operator, the sensitivity of the channel coils is often also considered. Therefore, the MRI device can also obtain a sensitivity matrix for each channel coil. Each channel coil's sensitivity matrix characterizes the sensitivity of the channel coil, which determines its ability to receive signals from different locations.

[0088] Step 205: Based on the sampling trajectory, the sensitivity matrix and the interpolation relationship, a sampling operator for the undersampled Kt data of the cardiac cycle is obtained.

[0089] For each cardiac cycle, the MRI device can obtain a sampling operator for the undersampled Kt data of that cardiac cycle based on the sampling trajectory of that cardiac cycle, the sensitivity matrix of the channel coils, and the interpolation relationship between the undersampled Kt data of that cardiac cycle. The sampling operator is used to describe the mapping relationship between the undersampled Kt data and the magnetic resonance signals of the heart during the MRI device's acquisition process. The sampling operator may include: the subsampling operator used by each channel coil to sample the heart during each cardiac cycle.

[0090] The subsampling operator used by each channel coil during a cardiac cycle describes the mapping relationship between the undersampled Kt sub-data acquired by the channel coil during that cardiac cycle and the cardiac magnetic resonance signals acquired by the channel coil. Each subsampling operator can be represented by a matrix. Each subsampling operator can be determined based on the sampling trajectory, the sensitivity matrix of the channel coil, and the interpolation relationship of the cardiac cycle.

[0091] It's understood that MRI imaging is based on data points in k-space, while the sensitivity matrix of the channel coil is constructed based on the image space coordinate system. k-space and image space can generally be linked via Fourier transforms. Therefore, the subsampling operator determined by the MRI device also includes a Fourier transform operator, implemented using a fast discrete Fourier transform.

[0092] For example, assume that the interpolation relationship is represented by an interpolation matrix, M is is the matrix representing the sampling trajectory in the is-th cardiac cycle, S ic is the sensitivity matrix of the icth channel coil, F represents the Fourier transform operator, G is is the interpolation matrix of the is-th cardiac cycle, then the subsampling operator A used by the ic-th channel coil in the is-th cardiac cycle is is,ic It can be expressed as: A is,ic =M is FS ic G is .

[0093] Step 206: Obtain solution constraints for the image reconstruction model.

[0094] The solution constraints can include total variation (TV) constraints, wavelet transform constraints, time-dimensional Fourier transform (TFT) constraints, and nuclear norm constraints. These solution constraints define the hard conditions that must be met by the solution to the image reconstruction model. This allows the most suitable solution to be selected from an infinite number of solutions.

[0095] Optionally, only a portion of the tissue in the reconstructed cardiac image moves, while the remaining tissue is essentially stationary. Based on this, the cardiac image can be decomposed into a background component and a dynamic component. The dynamic component is the motion component of the cardiac image. In this case, the solution constraint may include: a first solution constraint corresponding to the background component, and a second solution constraint corresponding to the dynamic component.

[0096] Optionally, the first solution constraint condition may be a nuclear norm constraint, and the second solution constraint condition may be a total variation constraint.

[0097] Step 207: construct an image reconstruction model based on the sampling operator, the undersampled Kt data, and the solution constraint conditions, and solve the image reconstruction model to obtain a cardiac image.

[0098] After the MRI device obtains the undersampled Kt data, the sampling operator, and the constraint conditions, it can construct an image reconstruction model based on the undersampled Kt data, the sampling operator, and the constraint conditions. The MRI device can then solve the image reconstruction model to obtain a cardiac image. The image reconstruction model is essentially a constrained optimization problem. The sampling operator and undersampled Kt data constitute the fidelity terms of the image reconstruction model, and the constraint conditions constitute the constraint terms of the image reconstruction model. That is, the total variation constraints, wavelet transform constraints, time-dimensional Fourier transform constraints, and nuclear norm constraints mentioned above can constitute the constraint terms of the image reconstruction model.

[0099] In the embodiment of the present application, assuming that the constraints to be solved include: total variation constraint and nuclear norm constraint, the constructed image reconstruction model can be: Among them, d is,ic is the undersampled Kt sub-data collected by the icth channel coil in the isth cardiac cycle. ic is greater than or equal to 1 and less than or equal to the total number of channel coils, chanls, is greater than or equal to 1 and less than or equal to the total number of complete cardiac cycles required to scan the entire K space (shotNum). X is the cardiac image to be solved, which is represented in the form of a vector. TV(X) represents the total variation constraint on the cardiac image, ||X|| * Indicates that the nuclear norm constraint is applied to the cardiac image. λ1 and λ2 are both regularization parameters, and λ1 can usually be 0.0025, and λ2 can usually be 0.001.

[0100] When the cardiac image is decomposed into background component and dynamic component, the image reconstruction model can satisfy: Where L is the background component of the cardiac image to be solved, and S is the dynamic component of the cardiac image. Both the background component and the dynamic component are represented in the form of vectors. TV(S) represents the total variation constraint on the dynamic component S, ||L|| * Indicates the nuclear norm constraint on the background component L.

[0101] Decomposing a cardiac image into a background component and a dynamic component for analysis, the resulting dynamic component S, after sparse transformation, is theoretically sparser than the result obtained by directly performing a sparse transformation on the cardiac image, thus offering higher acceleration capabilities. Furthermore, decomposing X into the background component L and the dynamic component S simplifies the image reconstruction model, enabling the use of a faster-converging algorithm for reconstruction, resulting in more accurate results.

[0102] Optionally, the MRI device may use a fast iterative shrinkage-thresholding algorithm (FISTA) to solve the image reconstruction model to obtain a cardiac image.

[0103] It is understood that, if the cardiac image is not decomposed, the MRI device can obtain the optimal cardiac image X after solving the image reconstruction model. If the cardiac image is decomposed into background and dynamic components, the MRI device can obtain the optimal background and dynamic components after solving the image reconstruction model. The MRI device can then obtain the cardiac image based on the optimal background and dynamic components.

[0104] The following describes a process for solving an image reconstruction model using the FISTA algorithm to obtain a cardiac image by decomposing the cardiac image to be solved into a background component and a dynamic component. This process may include:

[0105] Step S1, initialization parameters.

[0106] At the beginning of the algorithm iteration, the MRI device can initialize various variables in the algorithm iteration process. The variables may include: background component L, dynamic component S, extrapolation point y, iteration count parameter k and momentum parameter t k Among them, the initial background component L0 can satisfy: H represents the conjugate transpose. The initial dynamic component S0 is 0, the initial extrapolation point y0 is 0, the initial iteration count parameter k is 0, and the initial momentum parameter t0 is 1.

[0107] Step S2: Execute the iterative process until the iteration ends and output the optimal background component and dynamic component.

[0108] The iterative process includes: calculating the gradient z k ; Based on the gradient z k A shrinkage-thresholding operation is performed to update the background component and the dynamic component; a momentum parameter is updated; and an extrapolation point is calculated based on the updated momentum parameter.

[0109] Among them, the gradient z k Can satisfy:

[0110]

[0111] In formula (1), L c is the Lipschitz constant, f(L k ,S k ) means:

[0112]

[0113] Updated background component L k+1 Can satisfy:

[0114]

[0115] In formula (2), singular value represents a singular value.

[0116] Optionally, a fast gradient projection algorithm (FGP) can be used to calculate the TV denoising process. Based on this, the updated dynamic component S k+1 Can satisfy:

[0117]

[0118] Updated momentum parameter t k+1 Can satisfy:

[0119]

[0120] At this time, the extrapolated point y k+1 Can satisfy:

[0121]

[0122] The following describes the process of directly solving the cardiac image using the FISTA algorithm to solve the image reconstruction model and obtain the cardiac image. This process may include:

[0123] Step B1: Initialize parameters.

[0124] The specific implementation of step B1 can refer to the relevant implementation process of step A1, and the embodiment of this application will not be repeated here. The initial heart image X0 can meet the following requirements:

[0125] Step B2: Repeat the iterative process until the iteration ends and the optimal heart image is output.

[0126] The iterative process includes: calculating the gradient z k ; Based on the gradient z k Perform a contraction-thresholding operation to update the cardiac image; update the momentum parameters; and calculate the extrapolation points.

[0127] Among them, the gradient z k Can satisfy:

[0128]

[0129] Updated Heart Image X k+1Can satisfy:

[0130]

[0131] Updated momentum parameter t k+1 The above formula (4) can be satisfied. At this time, the updated extrapolated point y k+1 Can satisfy:

[0132]

[0133] Figure 5 This is a schematic diagram of a reconstructed cardiac image obtained when acceleration is performed only in k-space and the acceleration factor is 3 in the related art. Figure 6 This is a schematic diagram of a reconstructed heart image obtained by using the method provided in the example of this application with an acceleration factor of 8. Figure 5 and Figure 6 It can be seen that the heart image reconstructed using the method provided in the example of this application is clearer and has higher image quality.

[0134] Figure 7 This is a schematic diagram of a cardiac image reconstructed based on full acquisition data provided in an embodiment of the present application. Figure 6 and Figure 7 It can be seen that the cardiac image reconstructed using the method provided in the example of this application is basically the same as the cardiac image reconstructed based on the full-acquisition data, and the image quality is higher.

[0135] It can be seen that the method provided in the embodiment of the present application combines retrospective gated kt acceleration and compressed sensing technology, and further improves the acceleration multiple (such as up to about 15 times) while improving image quality and temporal resolution, thereby greatly shortening the reconstruction time, thereby controlling the reconstruction time within an acceptable range for clinical scanning.

[0136] It is understood that the order of the steps of the cardiac cine imaging method provided in the embodiment of the present application can be appropriately adjusted, and the steps can be increased or decreased accordingly. For example, step 204 can be performed before step 203, or simultaneously with step 203; or, step 203 can be deleted according to the circumstances; or, step 204 can also be deleted according to the circumstances. Any method that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be covered by the scope of protection of this application, and therefore will not be described in detail.

[0137] In summary, the embodiment of the present application provides a cardiac movie imaging method, which can use retrospective gating to obtain undersampled Kt data of the heart in each cardiac cycle, and the actual number of phases of the cardiac cycle is the same as or different from the ideal number of phases. The method can also obtain the interpolation relationship of each cardiac cycle, and derive the sampling operator of the undersampled Kt data based on the interpolation relationship, and then construct an image reconstruction model based on the sampling operator and the undersampled Kt data, and solve the image reconstruction model to obtain a cardiac image. The interpolation relationship is used to interpolate the first image data corresponding to the ideal number of phases in the cardiac cycle into the second image data corresponding to the actual number of phases. It can be seen that the method provided by the embodiment of the present application can innovatively combine the retrospective gating technology with the Kt compressed sensing algorithm to achieve more efficient data acquisition, thereby effectively improving the acceleration factor. Moreover, in retrospective gating, the actual phase data of each cardiac cycle may be different from the ideal phase data due to heart rate fluctuations. Therefore, the embodiment of the present application innovatively introduces an interpolation relationship in the sampling operator, thereby effectively bridging the difference between the uniform ideal phase distribution and the non-uniform actual phase distribution, realizing the combination of retrospective gating and Kt compressed sensing method, so that the compressed sensing algorithm can be applied to the accelerated scenario of retrospective cardiac movie imaging, and then image reconstruction based on the undersampled Kt data collected by retrospective gating can be realized.

[0138] The present application embodiment provides an MRI device, which can be used to perform the cardiac movie imaging method provided in the above method embodiment. Figure 8 The MRI device 300 includes a processor 301, which is configured to:

[0139] Retrospective gating is used to obtain undersampled Kt data of the heart in each cardiac cycle. The actual number of phases in each cardiac cycle is the same as or different from the ideal number of phases.

[0140] For each cardiac cycle, an interpolation relationship of the cardiac cycle is obtained, where the interpolation relationship is used to interpolate first image data of the heart in the cardiac cycle into second image data, where the number of phases of the undersampled Kt data corresponding to the first image data is an ideal number of phases, and the number of phases of the undersampled Kt data corresponding to the second image data is an actual number of phases of the cardiac cycle;

[0141] Based on the interpolation relationship, a sampling operator of the undersampled Kt data is obtained, where the sampling operator is used to describe a mapping relationship between the undersampled Kt data and the magnetic resonance signal of the heart;

[0142] An image reconstruction model is constructed based on the sampling operator and the undersampled Kt data, and the image reconstruction model is solved to obtain a cardiac image.

[0143] Optionally, the processor 301 may be configured to:

[0144] For each first sampling point in the actual number of first sampling points of the cardiac cycle, determining a position of the first sampling point in a sampling interval where the ideal number of second sampling points is located;

[0145] Determining, based on the position of the first sampling point in the sampling interval, a weight of each second sampling point when mapping the first sampling point to the position;

[0146] Based on the weights of the second sampling points, an interpolation relationship of the cardiac cycle is obtained.

[0147] Optionally, the processor 301 may be configured to:

[0148] Based on the position of the first sampling point in the sampling interval, the weights of the second sampling points when mapping the first sampling point to the position are determined by linear interpolation.

[0149] Optionally, the processor 301 may be configured to:

[0150] Obtaining the sampling trajectory of the cardiac cycle and the sensitivity matrix of the channel coil;

[0151] Based on the sampling trajectory, sensitivity matrix and interpolation relationship, the sampling operator of the undersampled Kt data is obtained.

[0152] Optionally, the processor 301 may be configured to:

[0153] Obtaining solution constraints for the image reconstruction model;

[0154] An image reconstruction model is constructed based on solving constraints, sampling operators and under-sampled Kt data.

[0155] Optionally, the constraint conditions to be solved include at least one of: a total variation constraint, a wavelet transform constraint, a time-dimensional Fourier transform constraint, and a nuclear norm constraint.

[0156] Optionally, the processor 301 may be configured to:

[0157] The FISTA algorithm is used to solve the image reconstruction model to obtain the heart image.

[0158] In summary, an embodiment of the present application provides an MRI device that can use retrospective gating to obtain undersampled Kt data of the heart in each cardiac cycle, where the actual number of phases in the cardiac cycle is the same as or different from the ideal number of phases. The MRI device can also obtain the interpolation relationship of each cardiac cycle, and derive the sampling operator of the undersampled Kt data based on the interpolation relationship, and then construct an image reconstruction model based on the sampling operator and the undersampled Kt data, and solve the image reconstruction model to obtain a cardiac image. The interpolation relationship is used to interpolate the first image data corresponding to the ideal number of phases in the cardiac cycle into the second image data corresponding to the actual number of phases. It can be seen that the MRI device provided by the embodiment of the present application can innovatively combine retrospective gating technology with the Kt compressed sensing algorithm to achieve more efficient data acquisition, thereby effectively improving the acceleration factor. Moreover, in retrospective gating, the actual phase data of each cardiac cycle may be different from the ideal phase data due to heart rate fluctuations. Therefore, by innovatively introducing an interpolation relationship in the sampling operator, MRI equipment can effectively bridge the gap between the uniform ideal phase distribution and the non-uniform actual phase distribution, realizing the combination of retrospective gating and Kt compressed sensing methods, so that the compressed sensing algorithm can be applied to the accelerated scenario of retrospective cardiac movie imaging, and then image reconstruction based on the undersampled Kt data collected by retrospective gating can be realized.

[0159] like Figure 8 As shown, MRI device 300 includes a processor 301 and a memory 303. Processor 301 and memory 303 are connected, for example, via a bus 302. Optionally, MRI device 300 may also include a transceiver 304. It should be noted that in actual applications, there is not limited to one transceiver 304, and the structure of MRI device 300 does not constitute a limitation on the embodiments of this application.

[0160] The processor 301 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It may implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. The processor 301 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like.

[0161] The bus 302 may include a path for transmitting information between the above components. The bus 302 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus 302 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 8 Only one thick line is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0162] The memory 303 is used to store a computer program corresponding to the cardiac cine imaging method of the above embodiment of the present application, and the computer program is controlled and executed by the processor 301. The processor 301 is used to execute the computer program stored in the memory 303 to implement the content shown in the above method embodiment.

[0163] The present application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the cardiac movie imaging method provided in the above method embodiment. For example, Figure 1 or Figure 2 The method shown.

[0164] The present application also provides a computer program product, which includes a computer program or computer instructions. When the computer program or computer instructions are executed by a processor, the cardiac movie imaging method provided in the above method embodiment is implemented. For example, Figure 1 or Figure 2 The method shown.

[0165] It should be noted that the logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic device), a portable computer disk cartridge (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and programmable read-only memory (EPROM or flash memory), fiber optic devices, and portable compact disc read-only memory (CDROM). Furthermore, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing it in another suitable manner if necessary, and then storing it in a computer memory.

[0166] It should be understood that various parts of the present application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used to implement: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0167] Throughout this specification, reference to terms such as "one embodiment," "some embodiments," "examples," "specific examples," or "some examples" means that a specific feature, structure, material, or characteristic described in conjunction with that embodiment or example is included in at least one embodiment or example of the present application. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0168] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of such features. Throughout the description of this application, "plurality" means at least two, for example, two, three, etc., unless otherwise specifically defined.

[0169] In this application, unless otherwise specified or limited, the terms "installed," "connected," "connect," "fixed," etc. should be understood in a broad sense. For example, they can refer to fixed connection, detachable connection, or integration; mechanical connection, electrical connection; direct connection, or indirect connection through an intermediate medium; internal communication between two elements, or interaction between two elements, unless otherwise specified. Those skilled in the art will understand the specific meanings of the above terms in this application based on specific circumstances.

[0170] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A cardiac movie imaging method, characterized in that: The method comprises: Using retrospective gating, undersampled Kt data of the heart in each cardiac cycle is obtained, wherein the actual number of phases in each cardiac cycle is the same as or different from the ideal number of phases; For each of the cardiac cycles, an interpolation relationship of the cardiac cycle is obtained, where the interpolation relationship is used to interpolate first image data of the heart in the cardiac cycle into second image data, where the number of phases of the undersampled Kt data corresponding to the first image data is the ideal number of phases, and the number of phases of the undersampled Kt data corresponding to the second image data is the actual number of phases of the cardiac cycle; Based on the interpolation relationship, a sampling operator of the under-sampled Kt data is acquired, where the sampling operator is used to describe a mapping relationship between the under-sampled Kt data and the magnetic resonance signal of the heart; An image reconstruction model is constructed based on the sampling operator and the under-sampled Kt data, and the image reconstruction model is solved to obtain a cardiac image.

2. The method according to claim 1, characterized in that Obtaining the interpolation relationship of the cardiac cycle includes: For each first sampling point in the actual phase number of the cardiac cycle, determining a position of the first sampling point in a sampling interval where the second sampling points in the ideal phase number of the cardiac cycle are located; determining, based on a position of the first sampling point in the sampling interval, a weight of each of the second sampling points when mapping the first sampling point to the position; An interpolation relationship of the cardiac cycle is obtained based on the weights of each of the second sampling points.

3. The method according to claim 2, characterized in that The determining, based on the position of the first sampling point in the sampling interval, the weight of each of the second sampling points when mapping the first sampling point to the position includes: determining, based on the position of the first sampling point in the sampling interval, by interpolation, a weight of each of the second sampling points when mapping the first sampling point to the position; The interpolation method is linear interpolation, polynomial interpolation or spline interpolation.

4. The method according to any one of claims 1 to 3, characterized in that The step of obtaining a sampling operator for the undersampled Kt data based on the interpolation relationship includes: Acquiring a sampling trajectory of the cardiac cycle and a sensitivity matrix of the channel coil; A sampling operator for the under-sampled Kt data is obtained based on the sampling trajectory, the sensitivity matrix, and the interpolation relationship.

5. The method according to claim 4, characterized in that In each cardiac cycle, the number of the multiple K lines collected in any two phases is the same, and the positions of the multiple K lines in any two phases in the K space are not completely the same.

6. The method according to any one of claims 1 to 3, characterized in that: Constructing an image reconstruction model based on the sampling operator and the under-sampled Kt data, comprising: Obtaining solution constraints for the image reconstruction model; The image reconstruction model is constructed based on the solution constraint conditions, the sampling operator and the under-sampling Kt data.

7. The method according to claim 6, characterized in that The solution constraint conditions include: at least one of: total variation constraint, wavelet transform constraint, time-dimensional Fourier transform constraint and nuclear norm constraint.

8. The method according to any one of claims 1 to 3, characterized in that: Solving the image reconstruction model to obtain a cardiac image includes: The image reconstruction model is solved using the FISTA algorithm to obtain the heart image.

9. A magnetic resonance imaging device, characterized in that The magnetic resonance imaging device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method according to any one of claims 1 to 8 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images