Magnetic resonance image data generation method and device, storage medium, and terminal

Abstract

The present invention discloses a method and device for generating magnetic resonance image data, a storage medium, and a terminal, which relate to the technical field of data processing, and the main purpose is to solve the problem of poor accuracy in the existing generation of magnetic resonance image data. It includes: acquiring magnetic resonance scanning data of different levels, the magnetic resonance scanning data is obtained by collecting trajectories between different narrow bands at different levels; parsing the correlation coefficients between different narrow bands at different levels based on the magnetic resonance scanning data, and determining the movement distance between different levels through the correlation coefficient; performing motion correction on the magnetic resonance scanning data of the different levels according to the movement distance between the levels, and performing image reconstruction based on the corrected magnetic resonance scanning data to obtain magnetic resonance image data. It is mainly used for the generation of magnetic resonance image data.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method and device for generating magnetic resonance image data, a storage medium, and a terminal. Background Art

[0002] Magnetic resonance imaging (MRI) is one of the main imaging methods in modern medical imaging. It refers to the use of magnetic resonance phenomenon, radio frequency excitation to excite hydrogen protons in the human body, and the use of gradient fields for spatial position encoding, so that the receiving coil receives the signal with position information, and finally reconstructs the image information through Fourier transform. In this process, since the human body's respiratory movement will cause significant movement of the organs, resulting in image blur and artifacts, the PROPELLER sequence is usually used in combination with respiratory gating triggering to collect scanning data. Among them, the PROPELLER sequence is composed of multiple narrow bands passing through the same K space center, and the multiple narrow bands are evenly distributed in the entire K space.

[0003] The PROPELLER acquisition sequence usually collects all layer data of a narrow band with a specific rotation angle at the end of each exhalation. The narrow bands of multiple cycles cover the entire K space. During the image reconstruction process, the artifacts are reduced by configuring weights for each narrow band at different levels. However, configuring weights cannot completely reduce the appearance of artifacts, and radial artifacts or noise ratio loss will be introduced during the weight configuration process, resulting in unclear images and greatly reducing the accuracy of magnetic resonance image data generation. Summary of the invention

[0004] In view of this, the present invention provides a method and device for generating magnetic resonance image data, a storage medium, and a terminal, the main purpose of which is to solve the problem of poor accuracy in the existing generation of magnetic resonance image data.

[0005] According to one aspect of the present invention, a method for generating magnetic resonance image data is provided, comprising:

[0006] Acquire magnetic resonance scanning data of different levels, wherein the magnetic resonance scanning data is acquired by rotating narrow bands in different periods, and within the same period, narrow bands of the same rotation angle of the different levels are acquired, and the magnetic resonance scanning data of any level includes multiple narrow bands of different rotation angles;

[0007] Analyzing the correlation coefficients between different narrow bands at different levels based on the magnetic resonance scan data, and determining the movement distances between different levels through the correlation coefficients;

[0008] Motion correction is performed on the magnetic resonance scan data of the different levels according to the inter-level motion distance, and image reconstruction is performed based on the corrected magnetic resonance scan data to obtain magnetic resonance image data.

[0009] Further, analyzing the correlation coefficients between different narrow bands at different levels based on the magnetic resonance scan data, and determining the movement distances between different levels by the correlation coefficients includes:

[0010] Constructing a linear correlation coefficient between any two narrow bands at different levels, and generating a motion distance curve corresponding to the linear correlation coefficient;

[0011] A second-order fitting is performed on the movement distance curve to construct a second-order linear equation group for solving the inter-level movement distance, so as to solve the inter-level movement distance. The coefficients of each term in the second-order linear equation group are the coefficient matrix obtained based on the second-order fitting.

[0012] Furthermore, constructing a linear correlation coefficient between any two narrowbands at different levels includes:

[0013] Select any two narrow bands from a plurality of narrow bands with different rotation angles;

[0014] The angle between the arbitrary two narrow bands is obtained, and one of the two narrow bands is angularly rotated according to the angle, so as to construct a linear correlation coefficient between different levels based on the narrow band after the angular rotation.

[0015] Further, the performing motion correction on the magnetic resonance scanning data of the different levels according to the inter-level motion distance includes:

[0016] Calculating the inter-level motion distance to match the linear phases of different levels;

[0017] Multiplying the frequency domain matrix of the magnetic resonance scanning data by the linear phase according to different levels to obtain a corrected frequency domain matrix;

[0018] The corrected frequency domain matrix is ​​subjected to Fourier transformation to obtain corrected magnetic resonance scanning data.

[0019] Furthermore, before performing motion correction on the magnetic resonance scan data of the different levels according to the inter-level motion distance, the method further includes:

[0020] Acquiring magnetic resonance imaging parameters, and parsing hierarchical dimensions in the magnetic resonance imaging parameters;

[0021] A three-dimensional matrix of the magnetic resonance scanning data is constructed according to the hierarchical dimension, and a one-dimensional inverse Fourier transform is performed on the three-dimensional matrix according to the hierarchical dimension to obtain a frequency domain matrix, wherein the frequency domain matrix includes phase encoding and frequency encoding of different hierarchical dimensions.

[0022] Furthermore, before constructing the linear correlation coefficient between any two narrow bands at different levels and generating the movement distance curve corresponding to the linear correlation coefficient, the method further includes:

[0023] The inter-layer motion search range is configured, and magnetic resonance scanning data is obtained according to the inter-layer motion search range. The magnetic resonance imaging parameters include the number of narrow bands, the angle between any two narrow bands, the total number of layers, and the thickness between layers.

[0024] According to another aspect of the present invention, there is provided a magnetic resonance image data generating device, comprising:

[0025] An acquisition module, used for acquiring magnetic resonance scanning data of different levels, wherein the magnetic resonance scanning data is acquired by rotating a narrow band in different periods, and within the same period, narrow bands of the same rotation angle of the different levels are acquired, and the magnetic resonance scanning data of any level includes a plurality of narrow bands of different rotation angles;

[0026] A determination module, configured to analyze correlation coefficients between different narrow bands at different levels based on the magnetic resonance scan data, and determine movement distances between different levels through the correlation coefficients;

[0027] A correction module is used to perform motion correction on the magnetic resonance scanning data of the different levels according to the inter-level motion distance, and to perform image reconstruction based on the corrected magnetic resonance scanning data to obtain magnetic resonance image data.

[0028] Furthermore, the determining module includes:

[0029] A construction unit, used to construct a linear correlation coefficient between any two narrow bands at different levels, and generate a motion distance curve corresponding to the linear correlation coefficient;

[0030] A fitting unit is used to perform a second-order fitting on the movement distance curve, construct a second-order linear equation group for solving the inter-level movement distance, so as to solve the inter-level movement distance, and each coefficient in the second-order linear equation group is a coefficient matrix obtained based on the second-order fitting.

[0031] Furthermore, the construction unit is specifically used to select any two narrow bands from a plurality of narrow bands with different rotation angles; obtain the angle between the any two narrow bands, and perform angular rotation on one of the any two narrow bands according to the angle, so as to construct a linear correlation coefficient between different levels based on the narrow band after the angular rotation.

[0032] Furthermore, the correction module comprises:

[0033] A first calculation unit, used for calculating the linear phase of different levels of the inter-level motion distance matching;

[0034] A second calculation unit is used to multiply the frequency domain matrix of the magnetic resonance scanning data with the linear phase according to different levels to obtain a corrected frequency domain matrix;

[0035] The processing unit is used to perform Fourier transform on the corrected frequency domain matrix to obtain corrected magnetic resonance scanning data.

[0036] Furthermore, the device further comprises: a parsing module,

[0037] The parsing module is used to obtain magnetic resonance imaging parameters and parse the hierarchical dimensions in the magnetic resonance imaging parameters;

[0038] The construction module is also used to construct a three-dimensional matrix of the magnetic resonance scanning data according to the hierarchical dimension, and perform a one-dimensional inverse Fourier transform on the three-dimensional matrix according to the hierarchical dimension to obtain a frequency domain matrix, wherein the frequency domain matrix includes phase encoding and frequency encoding of different hierarchical dimensions.

[0039] Furthermore, the determining module further includes:

[0040] The configuration unit is specifically used to configure the inter-layer motion search range, and start the magnetic resonance scanning according to the inter-layer motion search range and the magnetic resonance imaging parameters to obtain magnetic resonance scanning data. The magnetic resonance imaging parameters include the number of narrow bands, the angle between any two narrow bands, the total number of layers, and the thickness between layers.

[0041] According to another aspect of the present invention, a storage medium is provided, wherein at least one executable instruction is stored in the storage medium, and the executable instruction enables a processor to execute operations corresponding to the above-mentioned method for generating magnetic resonance image data.

[0042] According to another aspect of the present invention, there is provided a terminal, comprising: a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus;

[0043] The memory is used to store at least one executable instruction, and the executable instruction enables the processor to execute operations corresponding to the above-mentioned magnetic resonance image data generating method.

[0044] By means of the above technical solution, the technical solution provided by the embodiment of the present invention has at least the following advantages:

[0045] The present invention provides a method and device for generating magnetic resonance image data, a storage medium, and a terminal. Compared with the prior art, the embodiments of the present invention acquire magnetic resonance scanning data of different levels, wherein the magnetic resonance scanning data is acquired by rotating a narrow band in different periods, and within the same period, narrow bands of the same rotation angle of the different levels are acquired, and the magnetic resonance scanning data of any level includes multiple narrow bands of different rotation angles; the correlation coefficients of different narrow bands at different levels are analyzed based on the magnetic resonance scanning data, and the movement distances between different levels are determined by the correlation coefficients; motion correction is performed on the magnetic resonance scanning data of the different levels according to the movement distances between the levels, and image reconstruction is performed based on the corrected magnetic resonance scanning data to obtain magnetic resonance image data, thereby greatly reducing the occurrence of artifacts, improving the image clarity of magnetic resonance imaging, and thus improving the accuracy of generating magnetic resonance image data.

[0046] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0048] Figure 1 A flow chart of a method for generating magnetic resonance image data provided by an embodiment of the present invention is shown;

[0049] Figure 2 A schematic diagram of a rotation scan provided by an embodiment of the present invention is shown;

[0050] Figure 3 A schematic diagram of two narrowband motion distance curves provided by an embodiment of the present invention is shown;

[0051] Figure 4 A flow chart of another method for generating magnetic resonance image data provided by an embodiment of the present invention is shown;

[0052] Figure 5 A flow chart of another method for generating magnetic resonance image data provided by an embodiment of the present invention is shown;

[0053] Figure 6 A flow chart of another method for generating magnetic resonance image data provided by an embodiment of the present invention is shown;

[0054] Figure 7 A block diagram showing the composition of a magnetic resonance image data generating device provided by an embodiment of the present invention is shown;

[0055] Figure 8 A schematic diagram of the structure of a terminal provided by an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0056] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0057] When acquiring the PROPELLER sequence, all layer data of a narrow band with a specific rotation angle is usually acquired at the end of each exhalation. During the image reconstruction process, the appearance of artifacts is reduced by configuring weights between narrow band layers. However, configuring weights cannot completely reduce the appearance of artifacts, and radial artifacts or noise ratio loss may be introduced during the weight configuration process, resulting in unclear images and greatly reducing the accuracy of magnetic resonance image data generation. An embodiment of the present invention provides a method for generating magnetic resonance image data, such as Figure 1 As shown, the method includes:

[0058] 101. Obtain magnetic resonance scan data at different levels.

[0059] In the embodiment of the present invention, the magnetic resonance scanning device acquires magnetic resonance scanning data through the PROPELLER sequence, so as to obtain a fully sampled K-space signal by rotating the narrow band angle in different respiratory cycles, which is used as the magnetic resonance scanning data. The magnetic resonance scanning data is acquired by rotating the narrow band in different cycles. In the same cycle, the narrow bands of the same rotation angle at different levels are acquired. The magnetic resonance scanning data at any level includes multiple narrow bands with different rotation angles. The cycle in the embodiment of the present invention can be a respiratory cycle or a cardiac cycle (in this case, the cardiac imaging data is obtained by scanning the narrow band rotated during the cardiac cycle), and there is no specific limitation. Figure 2As shown, during the scanning process according to the PROPELLER sequence, the magnetic resonance device covers the scanned object with multiple evenly distributed narrow-band blades. There is a specific angle between each narrow band, and the multiple evenly distributed narrow-band blades are collected in different breathing cycles. In each breathing cycle, multiple levels of narrow bands with the same angle are collected at the same time, and two-dimensional magnetic resonance scanning data is obtained at each level. As the magnetic resonance scanning data collected by each narrow band at the corresponding level, it is the magnetic resonance scanning data of the PROPELLER sequence.

[0060] 102. Analyze the correlation coefficients of different narrow bands at different levels based on the magnetic resonance scanning data, and determine the movement distances between different levels through the correlation coefficients.

[0061] In the embodiment of the present invention, the scanning object cannot ensure that the breathing state is the same every time during the scanning process, so the state of the region of interest is different at different end-expirations. Since narrow bands of specific angles of multiple levels are collected at each end-expiration, the different states of the region of interest at different end-expirations will cause the positions of the narrow bands collected at different end-expirations of the same level to be different relative to the region of interest. That is, at different end-expirations, the level position changes, and artifacts are likely to appear during image reconstruction. Therefore, the correlation coefficients of different narrow bands at different levels are analyzed based on the magnetic resonance scanning data, so as to determine the inter-level movement distances of each narrow band at different levels. Among them, the correlation coefficient is used to describe the numerical value of the linear correlation degree of the magnetic resonance scanning data between any two narrow bands at the corresponding levels, so that a movement distance curve between the levels corresponding to the correlation coefficient can be obtained, such as Figure 3 As shown, the horizontal axis is the distance in the hierarchical direction and the vertical axis is the correlation coefficient. Since the thickness of each narrow band is the same, and the obtained inter-level distance does not necessarily match the integer multiple of the thickness, the curve is derived, and the point of the minimum value is the movement value. Because the distance calculated between the two narrow bands is only the relative distance between the two narrow bands, if the data of one of the narrow bands is contaminated due to various occurrences, the calculated movement distance value will not be good. However, the probability that all narrow band data are contaminated is very low, so by calculating the coefficients of all narrow band pairs (blade pairs), the solution method of the simultaneous equations will be more stable. Therefore, based on this curve, a solution equation for calculating the inter-level movement distance can be constructed, such as constructing a first-order equation group based on this curve, solving the inter-level movement distance as a variable, and obtaining a solution matrix for the inter-level movement distance.

[0062] 103. Perform motion correction on the magnetic resonance scan data of the different levels according to the inter-level motion distance, and perform image reconstruction based on the corrected magnetic resonance scan data to obtain magnetic resonance image data.

[0063] In the embodiment of the present invention, since the calculated inter-layer motion distance may be a non-integer multiple of the thickness, when the magnetic resonance scan data is corrected based on the inter-layer motion distance, the inter-layer motion distance is used to perform motion correction on the layer dimension frequency domain matrix of the magnetic resonance scan data. The frequency domain matrix of the magnetic resonance scan data is obtained by performing a one-dimensional inverse Fourier transform on the magnetic resonance scan data according to the layer dimension, and then the frequency domain matrix is ​​multiplied by the linear phase of the corresponding inter-layer motion distance according to the layer dimension to achieve motion correction.

[0064] It should be noted that since the corrected magnetic resonance data is a frequency domain matrix with a hierarchical dimension, at this time, the frequency domain matrix can be converted back to a signal in the K space through a one-dimensional Fourier transform, so as to be used as the corrected magnetic resonance scan data for image reconstruction to obtain magnetic resonance image data. Among them, since the hierarchical dimension is that the k space is uniform, it can be carried out through a Cartesian reconstruction method, and since the acquisition within the layer is a non-Cartesian acquisition, the magnetic resonance scan data is not on a uniformly spaced grid point, and the image cannot be directly reconstructed using a Fourier transform. Therefore, the magnetic resonance scan data needs to be interpolated to the Cartesian grid point, that is, gridding. In addition, since the acquisition density of different k spaces of non-Cartesian acquisition is also different, contrast correction can also be performed based on a density compensation function, which is not specifically limited in the embodiment of the present invention.

[0065] In another embodiment of the present invention, in order to further define and illustrate, as Figure 4 As shown, step 102 analyzes the correlation coefficients between different narrow bands at different levels based on the magnetic resonance scan data, and determines the motion distances between different levels through the correlation coefficients, including:

[0066] 1021. Construct a linear correlation coefficient between any two narrow bands at different levels, and generate a movement distance curve corresponding to the linear correlation coefficient;

[0067] 1022. Perform a second-order fitting on the movement distance curve to obtain second-order term coefficients and first-order term coefficients, and construct a first-order linear equation group for solving the movement distance at corresponding levels for any two narrow bands based on the second-order term coefficients and the first-order term coefficients.

[0068] In order to solve the problem of artifacts or blurring caused by different breathing states of the scanned object, which leads to mismatch of magnetic resonance scan data corresponding to different narrow bands between layers, a linear correlation coefficient between any two narrow bands at different layers is constructed, and a motion distance curve is generated with the linear correlation coefficient and the motion distance as the image axis. The first-order derivative of the curve is obtained, and its minimum point is the motion distance value of the two narrow bands at this layer. Specifically, the linear correlation function γ(X,Y)= 构建任意两个窄带处于不同层级间的线性相关系数，其中，X为第i个窄带的磁共振扫描数据，Y为第j个窄带blade的磁共振扫描数据，i＝1,2,3....M，j＝2,3....M，M为窄带blade的个数，此时说明在生成相关系数时，是指两个窄带中一个窄带的指定层对于另一个窄带的±S层的相关系数。同时，由于采集的磁共振扫描数据为在层级间进行的，选取±S作为层级间运动的搜索范围，在构建相关系数时，则在n±S范围内构建层级对应的相关系数，n∈[1,N]为层级数，N为层级总数。

[0069] 需要说明的是，由于相关系数用于描述任意两个窄带blade之间在对应层级间的磁共振扫描数据的线性相关程度的数值，因此可以绘制一条相关系数对应不同层级间的运动距离曲线，如 图3所示，其中，横轴为层级方向的距离，纵轴为相关系数。进而通过运动距离曲线进行二阶拟合，得到二阶项系数和一阶项系数，并基于所述二阶项系数和所述一阶项系数构建任意两窄带在对应层级处对运动距离求解的一阶项线性方程组，以进行求解得到对应层级间所有窄带的运动距离，其中，一阶项线性方程组中的各项系数为基于二阶拟合得到的系数矩阵。例如，第i个窄带在第n层的层级间运动距离为δ i,n，对应的二阶项线性方程组为：

[0070] 其中，a ij为二阶项系数，b ij为一阶项系数。

[0071] 在另一个本发明实施例中，为了进一步限定及说明，步骤构建任意两个窄带处于不同层级间的线性相关系数包括：在多个不同旋转角度的窄带中，选取任意两个窄带；获取所述任意两个窄带的夹角，按照所述夹角对所述任意两个窄带中的一个窄带进行角度旋转，使得两个窄带的读出方向一致，以基于角度旋转后的所述窄带构建不同层级间的线性相关系数。

[0072] 由于相邻窄带blade之间的夹角是固定的，且通过 计算得到，并且，不同窄带blade之间的读取磁共振扫描数据的方向不同，即读取频率编码方向不同，因此，为了进行统一的线性相关系数的构建，则基于 可以计算得到任意两个窄带的夹角，即计算第i个窄带blade和第j个窄带blade之间的夹角，并对第j个窄带blade进行旋转，此时，旋转的角度即为第i个窄带blade和第j个窄带blade之间的夹角θ(j-i)，然后，基于第i个窄带blade和旋转角度后的第j个窄带blade构建不同层级间的线性相关系数。

[0073] 在另一个本发明实施例中，为了进一步限定及说明，如 图5所示，步骤103根据所述层级间运动距离对所述不同层级的所述磁共振扫描数据的进行运动校正包括：

[0074] 1031、计算所述层级间运动距离匹配不同层级的线性相位；

[0075] 1032、按照不同层级将所述磁共振扫描数据的频域矩阵与所述线性相位进行相乘计算，得到校正后的频域矩阵；

[0076] 1033、对校正后的所述频域矩阵进行傅里叶变换，得到校正后的磁共振扫描数据。

[0077] 为了基于层级间运动距离进行伪影消除，从而提高磁共振扫描数据的成像准确性以及清晰度，基于层级间运动距离进行运动校正，具体是通过磁共振扫描数据的频域矩阵与层级间运动距离的线性相位进行相乘实现。其中，线性相位为基于层级间运动距离进行线性相位计算得到的，线性相位的计算方式为exp(I*2π / N*δ i,n)，I为单位虚数，π为圆周率，N为层级总数。另外，由于频域矩阵为按照层级构建的，在按照不同层级将频域矩阵与线性相位相乘计算时，具体通过公式exp(l i*2p i / N*δ i,n)*(p-N / 2)计算得到，其中，p为层级数，p＝1,2,3....N，此时，即为p层级的频域矩阵与p层级的线性相位进行相乘计算，得到校正后的频域矩阵。

[0078] 需要说明的是，由于步骤1032中频域矩阵为在层级维度进行逆傅里叶变换得到的，因此，在得到校正后的频域矩阵后进行傅里叶变换，得到校正后待进行图像重建的磁共振扫描数据，即为第i个窄带blade在第n层进行层面间运动校正后的磁共振扫描数据。

[0079] 在另一个本发明实施例中，为了进一步限定及说明，如 图6所示，步骤103根据所述层级间运动距离对所述不同层级的所述磁共振扫描数据的进行运动校正之前，所述方法还包括：

[0080] 201、获取磁共振成像参数，并解析所述磁共振成像参数中的层级维度；

[0081] 202、按照所述层级维度构建所述磁共振扫描数据的三维矩阵，并以所述层级维度对所述三维矩阵进行一维逆傅里叶变换得到频域矩阵。

[0082] 为了实现从不同层级维度进行运动校正，从而伪影消除，首先要构建磁共振扫描数据的频域矩阵。具体的，获取磁共振设备中的磁共振成像参数，此时，磁共振成像参数包括M为窄带blade的个数，层级间运动的搜索范围±S，层级总数N，相邻窄带blade之间的夹角 从而控制磁共振设备进行扫描。解析磁共振成像参数中的层级维度即为确定层级总数，以按照此层级总数的维度构建包含磁共振扫描数据的三维矩阵。此时，由于磁共振扫描数据包括相位编码、频率编码，层级维度在图像域，因此，三维矩阵中的三个维度分别为层级、相位编码、频率编码。然后，按照层级维度对此三维矩阵进行一维逆傅里叶变换得到频域矩阵，此时频域矩阵包括不同层级维度的相位编码、频率编码，从而通过此频域矩阵进行运动校正。

[0083] 在另一个本发明实施例中，为了进一步限定及说明，步骤1021构建任意两个窄带处于不同层级间的线性相关系数，并生成所述线性相关系数对应的运动距离曲线之前，所述方法还包括：配置层级间运动搜索范围，并按照所述层级间运动搜索范围，得到磁共振扫描数据，所述磁共振成像参数包括窄带个数、任意两个窄带的夹角、层级总数、层级间厚度。

[0084] 由于本发明实施例中采用的是磁共振在呼吸门控下进行PROPELLER采样扫描，得到PROPELLER序列的磁共振扫描数据，为了准确进行运动过程中的伪影消除，配置层级间运动搜索范围，从而在磁共振设备按照磁共振成像参数进行扫描获取的数据中选出磁共振扫描数据。其中，磁共振成像参数包括窄带个数M、任意两个窄带的夹角 层级总数N、层级间厚度K，配置±S作为层级间运动搜索范围，优选为，±S可以为运动距离与层级间厚度的比值，例如，呼吸时肝脏上下运动的距离为±10cm，采集层级间厚度K为5cm，则搜索范围±S为±2层。还可以不设定层级间运动搜索范围为固定值，而在扫描过程中在层级总数的范围内进行搜索确定，本发明实施例不做具体限定。

[0085] 本发明实施例提供了一种磁共振图像数据生成方法，与现有技术相比，本发明实施例通过获取不同层级的磁共振扫描数据，所述磁共振扫描数据为窄带在不同周期进行旋转采集得到的，在同一周期内，采集所述不同层级的同一旋转角度的窄带，任一层级的所述磁共振扫描数据包括多个不同旋转角度的窄带；基于所述磁共振扫描数据解析不同窄带在不同层级间的相关系数，并通过所述相关系数确定不同层级间的运动距离；根据所述层级间运动距离对所述不同层级的所述磁共振扫描数据的进行运动校正，并基于校正后的所述磁共振扫描数据进行图像重建得到磁共振图像数据，大大减少了伪影的出现，提高磁共振成像的图像清晰度，从而提高磁共振图像数据生成准确性。

[0086] 进一步的，作为对上述 图1所示方法的实现，本发明实施例提供了一种磁共振图像数据生成装置，如 图7所示，该装置包括：

[0087] 获取模块31，用于获取不同层级的磁共振扫描数据，所述磁共振扫描数据为窄带在不同周期进行旋转采集得到的，在同一周期内，采集所述不同层级的同一旋转角度的窄带，任一层级的所述磁共振扫描数据包括多个不同旋转角度的窄带；

[0088] 确定模块32，用于基于所述磁共振扫描数据解析不同窄带在不同层级间的相关系数，并通过所述相关系数确定不同层级间的运动距离；

[0089] 校正模块33，用于根据所述层级间运动距离对所述不同层级的所述磁共振扫描数据的进行运动校正，并基于校正后的所述磁共振扫描数据进行图像重建得到磁共振图像数据。

[0090] 进一步地，所述确定模块包括：

[0091] 构建单元，用于构建任意两个窄带处于不同层级间的线性相关系数，并生成所述线性相关系数对应的运动距离曲线；

[0092] 拟合单元，用于对所述运动距离曲线进行二阶拟合，构建对层级间运动距离求解的二阶项线性方程组，以进行求解得到层级间运动距离，所述二阶项线性方程组中的各项系数为基于二阶拟合得到的系数矩阵。

[0093] 进一步地，所述构建单元，具体用于获取任意两个窄带的夹角，按照所述夹角对所述任意两个窄带中的一个窄带进行角度旋转，以基于角度旋转后的所述窄带构建不同层级间的线性相关系数。

[0094] 进一步地，所述校正模块包括：

[0095] 第一计算单元，用于计算所述层级间运动距离匹配不同层级的线性相位；

[0096] 第二计算单元，用于按照不同层级将所述磁共振扫描数据的频域矩阵与所述线性相位进行相乘计算，得到校正后的频域矩阵；

[0097] 处理单元，用于对校正后的所述频域矩阵进行傅里叶变换，得到校正后的磁共振扫描数据。

[0098] 进一步地，所述装置还包括：解析模块，

[0099] 所述解析模块，用于获取磁共振成像参数，并解析所述磁共振成像参数中的层级维度；

[0100] 所述构建模块，还用于按照所述层级维度构建所述磁共振扫描数据的三维矩阵，并以所述层级维度对所述三维矩阵进行逆傅里叶变换得到频域矩阵，所述频域矩阵包括不同层级维度的相位编码、频率编码。

[0101] 进一步地，所述确定模块还包括：

[0102] 配置单元，用于配置层级间运动搜索范围，并按照所述层级间运动搜索范围，得到磁共振扫描数据，所述磁共振成像参数包括窄带个数、任意两个窄带的夹角、层级总数、层级间厚度。

[0103] 本发明实施例提供了一种磁共振图像数据生成装置，与现有技术相比，本发明实施例通过获取不同层级的磁共振扫描数据，所述磁共振扫描数据为窄带在不同周期进行旋转采集得到的，在同一周期内，采集所述不同层级的同一旋转角度的窄带，任一层级的所述磁共振扫描数据包括多个不同旋转角度的窄带；基于所述磁共振扫描数据解析不同窄带在不同层级间的相关系数，并通过所述相关系数确定不同层级间的运动距离；根据所述层级间运动距离对所述不同层级的所述磁共振扫描数据的进行运动校正，并基于校正后的所述磁共振扫描数据进行图像重建得到磁共振图像数据，大大减少了伪影的出现，提高磁共振成像的图像清晰度，从而提高磁共振图像数据生成准确性。

[0104] 根据本发明一个实施例提供了一种存储介质，所述存储介质存储有至少一可执行指令，该计算机可执行指令可执行上述任意方法实施例中的磁共振图像数据生成方法。

[0105] 图8示出了根据本发明一个实施例提供的一种终端的结构示意图，本发明具体实施例并不对终端的具体实现做限定。

[0106] 如 图8所示，该终端可以包括：处理器(processor)402、通信接口(CommunicationsInterface)404、存储器(memory)406、以及通信总线408。

[0107] 其中：处理器402、通信接口404、以及存储器406通过通信总线408完成相互间的通信。

[0108] 通信接口404，用于与其它设备比如客户端或其它服务器等的网元通信。

[0109] 处理器402，用于执行程序410，具体可以执行上述磁共振图像数据生成方法实施例中的相关步骤。

[0110] 具体地，程序410可以包括程序代码，该程序代码包括计算机操作指令。

[0111] 处理器402可能是中央处理器CPU，或者是特定集成电路ASIC(ApplicationSpecific Integrated Circuit)，或者是被配置成实施本发明实施例的一个或多个集成电路。终端包括的一个或多个处理器，可以是同一类型的处理器，如一个或多个CPU；也可以是不同类型的处理器，如一个或多个CPU以及一个或多个ASIC。

[0112] 存储器406，用于存放程序410。存储器406可能包含高速RAM存储器，也可能还包括非易失性存储器(non-volatile memory)，例如至少一个磁盘存储器。

[0113] 程序410具体可以用于使得处理器402执行以下操作：

[0114] 获取不同层级的磁共振扫描数据，所述磁共振扫描数据为窄带在不同周期进行旋转采集得到的，在同一周期内，采集所述不同层级的同一旋转角度的窄带，任一层级的所述磁共振扫描数据包括多个不同旋转角度的窄带；

[0115] 基于所述磁共振扫描数据解析不同窄带在不同层级间的相关系数，并通过所述相关系数确定不同层级间的运动距离；

[0116] 根据所述层级间运动距离对所述不同层级的所述磁共振扫描数据的进行运动校正，并基于校正后的所述磁共振扫描数据进行图像重建得到磁共振图像数据。

[0117] 显然，本领域的技术人员应该明白，上述的本发明的各模块或各步骤可以用通用的计算装置来实现，它们可以集中在单个的计算装置上，或者分布在多个计算装置所组成的网络上，可选地，它们可以用计算装置可执行的程序代码来实现，从而，可以将它们存储在存储装置中由计算装置来执行，并且在某些情况下，可以以不同于此处的顺序执行所示出或描述的步骤，或者将它们分别制作成各个集成电路模块，或者将它们中的多个模块或步骤制作成单个集成电路模块来实现。这样，本发明不限制于任何特定的硬件和软件结合。

[0118] 以上所述仅为本发明的优选实施例而已，并不用于限制本发明，对于本领域的技术人员来说，本发明可以有各种更改和变化。凡在本发明的精神和原则之内，所作的任何修改、等同替换、改进等，均应包括在本发明的保护范围之内。

Claims

1. A method for generating magnetic resonance image data, characterized in that: include: Acquire magnetic resonance scanning data of different levels, wherein the magnetic resonance scanning data is acquired by rotating narrow bands in different periods, and within the same period, narrow bands of the same rotation angle of the different levels are acquired, and the magnetic resonance scanning data of any level includes multiple narrow bands of different rotation angles; Analyzing the correlation coefficients between different narrow bands at different levels based on the magnetic resonance scan data, and determining the movement distances between different levels through the correlation coefficients; Performing motion correction on the magnetic resonance scan data of the different levels according to the inter-level motion distance, and performing image reconstruction based on the corrected magnetic resonance scan data to obtain magnetic resonance image data; The step of analyzing the correlation coefficients between different narrow bands at different levels based on the magnetic resonance scan data and determining the movement distances between different levels by the correlation coefficients includes: Constructing a linear correlation coefficient between any two narrow bands at different levels, and generating a motion distance curve corresponding to the linear correlation coefficient; Performing a second-order fitting on the movement distance curve to obtain second-order term coefficients and first-order term coefficients, and constructing a first-order term linear equation group for solving the movement distance of any two narrow bands at corresponding levels based on the second-order term coefficients and the first-order term coefficients, so as to obtain the movement distances of all narrow bands between corresponding levels by solving the first-order term linear equation group, wherein each term coefficient in the first-order term linear equation group is a coefficient matrix obtained based on the second-order fitting; The construction of the linear correlation coefficient between any two narrow bands at different levels includes: Select any two narrow bands from a plurality of narrow bands with different rotation angles; Acquire an angle between the two narrow bands, and perform angular rotation on one of the two narrow bands according to the angle, so as to construct a linear correlation coefficient between different levels based on the narrow band after the angular rotation; The performing motion correction on the magnetic resonance scan data according to the inter-layer motion distance comprises: Calculating the inter-level motion distance to match the linear phases of different levels; Multiplying the frequency domain matrix of the magnetic resonance scanning data by the linear phase according to different levels to obtain a corrected frequency domain matrix; The corrected frequency domain matrix is ​​subjected to Fourier transformation to obtain corrected magnetic resonance scanning data.

2. The method according to claim 1, characterized in that Before performing motion correction on the magnetic resonance scan data according to the inter-layer motion distance, the method further includes: Acquiring magnetic resonance imaging parameters, and parsing hierarchical dimensions in the magnetic resonance imaging parameters; A three-dimensional matrix of the magnetic resonance scanning data is constructed according to the hierarchical dimension, and a one-dimensional inverse Fourier transform is performed on the three-dimensional matrix according to the hierarchical dimension to obtain a frequency domain matrix, wherein the frequency domain matrix includes phase encoding and frequency encoding of different hierarchical dimensions.

3. The method according to claim 2, characterized in that Before constructing the linear correlation coefficient between any two narrow bands at different levels and generating the movement distance curve corresponding to the linear correlation coefficient, the method further includes: The inter-layer motion search range is configured, and magnetic resonance scanning data is obtained according to the inter-layer motion search range. The magnetic resonance imaging parameters include the number of narrow bands, the angle between any two narrow bands, the total number of layers, and the thickness between layers.

4. A magnetic resonance image data generating device, characterized in that: include: An acquisition module, used for acquiring magnetic resonance scanning data of different levels, wherein the magnetic resonance scanning data is acquired by rotating a narrow band in different periods, and within the same period, narrow bands of the same rotation angle of the different levels are acquired, and the magnetic resonance scanning data of any level includes a plurality of narrow bands of different rotation angles; A determination module, configured to analyze correlation coefficients between different narrow bands at different levels based on the magnetic resonance scan data, and determine movement distances between different levels through the correlation coefficients; a correction module, configured to perform motion correction on the magnetic resonance scan data of the different levels according to the inter-level motion distance, and to perform image reconstruction based on the corrected magnetic resonance scan data to obtain magnetic resonance image data; Wherein, the determination module includes: A construction unit, used to construct a linear correlation coefficient between any two narrow bands at different levels, and generate a motion distance curve corresponding to the linear correlation coefficient; A fitting unit, used for performing a second-order fitting on the movement distance curve, constructing a second-order linear equation group for solving the inter-level movement distance, so as to solve and obtain the inter-level movement distance, wherein each coefficient in the second-order linear equation group is a coefficient matrix obtained based on the second-order fitting; The construction unit is specifically used to select any two narrow bands from a plurality of narrow bands with different rotation angles; obtain the angle between the two narrow bands, and perform angle rotation on one of the two narrow bands according to the angle, so as to construct linear correlation coefficients between different levels based on the narrow band after the angle rotation; The correction module comprises: A first calculation unit, used for calculating the linear phase of different levels of the inter-level motion distance matching; A second calculation unit is used to multiply the frequency domain matrix of the magnetic resonance scanning data with the linear phase according to different levels to obtain a corrected frequency domain matrix; The processing unit is used to perform Fourier transform on the corrected frequency domain matrix to obtain corrected magnetic resonance scanning data.

5. A storage medium storing at least one executable instruction, wherein the executable instruction enables a processor to execute an operation corresponding to the method for generating magnetic resonance image data according to any one of claims 1 to 4.

6. A terminal, comprising: A processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform operations corresponding to the magnetic resonance image data generating method according to any one of claims 1 to 4.

Images