Medical data processing device, medical data processing method, medical data processing program, and magnetic resonance imaging device

Abstract

To perform estimation quickly with high reliability.SOLUTION: A medical data processing device according to an embodiment comprises: an acquisition unit; a generation unit; and a reconstruction unit. The acquisition unit acquires imaging data obtained by executing spoiled gradient echo imaging and coherent gradient echo imaging at a plurality of flip angles. The generation unit generates a low-rank approximate image set being a set of low-rank approximate images from the imaging data. The reconstruction unit reconstructs one or more parameter maps using a multi-pool model including a plurality of free waters related to water exchange of a biological tissue and a bound water subjected to water exchange with the free water and the low-rank approximate image set.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] An embodiment of the present invention relates to a medical data processing device, a medical data processing method, a medical data processing program, and a magnetic resonance imaging device. [Background technology]

[0002] For example, in the brain, depending on the structure of biological tissue, there is free water that can move relatively freely, such as intracellular fluid and intercellular fluid, and free water whose movement is restricted, such as myelin water confined in the myelin sheath. Note that below, free water with different characteristics will be collectively referred to as multiple free water. Multiple free water types alternately exchange water, and the proportion of each type of free water reflects the structure of biological tissue. There are multi-compartment microstructure imaging techniques that take into account parameters related to biological tissue structure. For example, there is a two-pool model that considers two free water pools. However, these techniques have problems with low accuracy and reliability because they do not take into account the magnetization transfer effect (MT effect) or B1 distribution and use imaging data in a steady state. Therefore, there is also a four-pool model, which is a multi-compartment microstructure model that takes the MT effect into account. However, the four-pool model has a large number of parameters, which is expected to take a long time to analyze, and it is unclear whether quantitative values ​​for each parameter map can be obtained. Furthermore, there are problems such as unknown imaging and analysis methods for imaging parameter values. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent Publication No. 2021-10408 [Non-patent literature]

[0004] [Non-Patent Document 1] JESSE I. HAMILTON, “MEASURING CARDIAC RELAXATION TIMES AND MULTI-COMPARTMENT WATER EXCHANGE WITH MAGNETIC RESONANCE FINGERPRINTING”, May 2018. [Non-patent document 2] TA Bjarnason et al, “Characterization of the NMR Behavior of White Matter in Bovine Brain”, Magnetic Resonance in Medicine 54, 1072-1081, 2005. Summary of the Invention [Problem to be solved by the invention]

[0005] The problem that the present invention aims to solve is to provide fast and reliable estimation. [Means for solving the problem]

[0006] The medical data processing device according to this embodiment includes an acquisition unit, a generation unit, and a reconstruction unit. The acquisition unit acquires imaging data obtained by performing spoiled gradient echo imaging and coherent gradient echo imaging at multiple flip angles. The generation unit generates a low-rank approximation image set, which is a set of low-rank approximation images from the imaging data. The reconstruction unit reconstructs one or more parameter maps using the low-rank approximation image set and a multi-pool model including multiple free waters related to water exchange in biological tissues and bound waters that exchange with the free waters. [Brief explanation of the drawings]

[0007] [Figure 1] FIG. 1 is a block diagram showing the configuration of a medical data processing device according to the first embodiment. [Figure 2] FIG. 2 is a conceptual diagram of the multi-pool model assumed in the first embodiment. [Figure 3] FIG. 3 is a flowchart showing the operation of the medical data processing device according to the first embodiment. [Figure 4] FIG. 4 is a diagram showing an example of an imaging sequence according to the first embodiment. [Figure 5] FIG. 5 is a diagram showing another example of the imaging sequence according to the first embodiment. [Figure 6] FIG. 6 is a diagram showing an example of radial scanning assumed in this embodiment. [Figure 7] FIG. 7 is a diagram illustrating the concept of a sparsely sampled low-rank approximate image and an example of its generation. [Figure 8] FIG. 8 is a diagram illustrating the concept of a fully sampled low-rank approximation image and an example of its generation. [Figure 9] FIG. 9 is a diagram showing an example of the first neural network during learning. [Figure 10] FIG. 10 is a diagram showing an example of the first neural network during inference (use). [Figure 11] FIG. 11 is a conceptual diagram of generating a low-rank approximation set from a transient signal. [Figure 12] FIG. 12 is a diagram showing an example of the second neural network during learning. [Figure 13] FIG. 13 is a diagram showing an example of the second neural network during inference (use). [Figure 14] FIG. 14 is a diagram showing an example of a calibration curve relating to the estimated values ​​of each parameter value. [Figure 15] FIG. 15 is a diagram showing a parameter map generated from estimated parameter values. [Figure 16] FIG. 16 is an example of another image based on low-rank approximations. [Figure 17] FIG. 17 is a diagram showing the overall configuration of a magnetic resonance imaging apparatus according to the second embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0008] A model that takes into account two free water pools is the two-pool model. The two free water pools are called the slow pool and the fast pool. The longitudinal relaxation time of the slow pool is defined as T1s [s], and the transverse relaxation time as T2s [s]. The longitudinal relaxation time of the fast pool is defined as T1f [s], and the transverse relaxation time as T2f [s]. The ratio of the fast pool to the total is defined as "f". Therefore, the ratio of the slow pool can be defined as "1-f". The slow pool and fast pool exchange water, and the exchange rate from the slow pool to the fast pool can be defined as ksf [Hz], and the exchange rate from the slow pool to the fast pool can be defined as kfs [Hz]. However, ksf and kfs have the following relationship based on the detailed balance conditions: kfs·f=ksf·(1-f)

[0009] Therefore, the number of independent parameters in the two-pool model is six. Separating and quantifying these parameters is useful for understanding the microstructure of biological tissues, and is known as multicompartment microstructure imaging. However, due to the technical difficulties involved, current MRI (Magnetic Resonance Imaging) techniques generally use a single-compartment model (T1 map, T2 map, etc.) that assumes a single pool of water rather than separating multiple water components, even when calculating quantitative values.

[0010] However, because the analysis is based on a single compartment model that differs from reality, the results are "apparent quantitative values" and are dependent on the imaging and analysis methods, resulting in limited versatility. In some cases, there may be a large error with the real biological model, which may not match values ​​obtained by other methods, making it impossible to compare with other methods or perform combined analyses. Therefore, to obtain versatile quantitative parameters, it is desirable to perform multicompartment microstructure quantitative imaging, which is more consistent with reality. For example, mcDESPOT (multi-component driven equilibrium single pulse observation of T1 and T2) is a known high-speed imaging method for a two-pool model. This method involves imaging SP-GRE and bSSFP at different flip angles, and determining the parameters of the two-pool model from the steady-state signal values.

[0011] However, since mcDESPOT does not take into account the B1 distribution or the magnetization transfer (MT) effect, it is known that errors due to the B1 distribution and errors due to the on-resonance MT effect caused by the excitation RF cannot be ignored. When obtaining T2 information with bSSFP, increasing the flip angle as much as possible improves the SNR and increases accuracy, but on the other hand, the MT effect becomes larger, making it impossible to ignore it.

[0012] Another recent technique proposed is a two-pool model analysis using MR Fingerprinting imaging. One proposed MR Fingerprinting method involves performing FISP imaging and matching it with a two-pool model dictionary. However, because the number of parameters is six, the dictionary becomes huge, and dictionary matching takes time. Furthermore, due to sparsity and artifacts during imaging, the required accuracy cannot be achieved, so full sampling imaging is required. These issues mean that the imaging time, analysis time, and accuracy obtained are not at a practical level. Furthermore, like the mcDESPOT technique mentioned above, this method has the problem of ignoring the MT effect.

[0013] However, because the MT effect itself actually reflects the microstructure of living organisms, it is desirable to simultaneously map the MT effect, since this would allow for more detailed information on biological tissues. Furthermore, considering the MT effect is desirable because it improves the agreement between reality and the model and is expected to improve the accuracy, reliability, and versatility of other parameters. Furthermore, our research has shown that when imaging with different MT effects is added, the free water content is also affected through the MT effect, making it possible to collect more detailed information on the entire biological tissue. Therefore, analysis taking the MT effect into account allows for more accurate overall analysis, including other parameters. Therefore, in the following embodiment, a four-pool model that takes the MT effect into account is used as an example.

[0014] The medical data processing device, medical data processing method, medical data processing program, and magnetic resonance imaging device according to the present embodiment will be described below with reference to the drawings. In the following embodiments, parts with the same reference numerals perform similar operations, and redundant explanations will be omitted as appropriate.

[0015] (First embodiment) A medical data processing apparatus according to a first embodiment will be described with reference to the block diagram of FIG. The medical data processing device 1 includes a processing circuit 2, an input interface 4, a communication interface 6, and a memory 8.

[0016] The processing circuit 2 includes an acquisition function 21, a generation function 22, a reconstruction function 23, and an optimization function 24. The processing circuit 2 has a processor (not shown) as a hardware resource.

[0017] The acquisition function 21 acquires imaging data obtained by performing spoiled gradient echo imaging and coherent gradient echo imaging with varying flip angles. The generation function 22 generates a low-rank approximate image set, which is a set of low-rank approximated images, from the imaging data. The reconstruction function 23 reconstructs one or more parameter maps using a multi-pool model including multiple free waters and bound waters that exchange with the free waters, relating to water exchange in biological tissues, and a low-rank approximation image set. The optimization function 24 performs optimization processing related to data consistency. The learning function 25 learns a first neural network and a second neural network, which will be described later, and generates a first trained model and a second trained model.

[0018] The input interface 4 has circuits for receiving various instructions and information input from a user. The input interface 4 has circuits related to input devices such as a pointing device such as a mouse or a keyboard. Note that the circuits included in the input interface 4 are not limited to circuits related to physical operation components such as a mouse and a keyboard. For example, the input interface 4 may have an electrical signal processing circuit that receives electrical signals corresponding to input operations from an external input device provided separately from the medical data processing device 1 and outputs the received electrical signals to various circuits. The communication interface 6 exchanges data with external devices via wire or wirelessly.

[0019] The memory 8 stores imaging data, a low-rank approximation image set, weighting coefficients for neural networks, trained models, parameter maps, and the like, which are used or generated by the medical data processing device 1. The memory 8 is a semiconductor memory element such as a random access memory (RAM), a flash memory, a hard disk drive (HDD), a solid state drive (SSD), an optical disk, or the like. The memory 8 may also be a drive or the like that reads and writes various information from and to a portable storage medium such as a CD-ROM drive, a DVD drive, or a flash memory.

[0020] The various functions of the processing circuit 2 are stored in the memory 8 in the form of programs executable by a computer. The processing circuit 2 is a processor that realizes the functions corresponding to the respective programs by reading the programs corresponding to the various functions from the memory 8 and executing them. In other words, the processing circuit 2 in a state in which the respective programs have been read out has the multiple functions shown in the processing circuit 2 of FIG. 1.

[0021] 1, it has been explained that these various functions are realized by a single processing circuit 2, but it is also possible to configure the processing circuit 2 by combining multiple independent processors, and have each processor execute a program to realize the function. In other words, it is possible that each of the above functions is configured as a program and one processing circuit executes each program, or that a specific function is implemented in a dedicated, independent program execution circuit.

[0022] Next, the concept of the multi-pool model assumed in the first embodiment, which is characterized by including multiple free water pools and bound water pools that exchange with them, will be explained with reference to Fig. 2. Here, bound water refers to water that is bound to biomolecules such as proteins and lipids, has an extremely fast T2 relaxation time of several tens of microseconds or less, and induces the MT effect by exchanging with free water pools. Figure 2 is a conceptual diagram of a multi-pool model, here a four-pool model, that includes bound water in addition to free water, which can move relatively freely within biological tissues, and free water, whose movement is restricted. Hereinafter, we will consider a four-pool model as a multi-pool model characterized by the inclusion of multiple free water molecules and bound water molecules that exchange water with them. Specifically, we will consider two free water molecules and two bound water molecules that exchange water with them. For ease of explanation, we will refer to the two free water molecules as the slow pool and fast pool, the bound water that exchanges water with the slow pool as the MT-Slow pool, and the bound water that exchanges water with the fast pool as the MT-Fast pool.

[0023] It is assumed that water exchange occurs between the MT-Slow pool and the Slow pool, between the Slow pool and the Fast pool, between the Fast pool and the Slow pool and the MT-Fast pool, and between the MT-Fast pool and the Fast pool. According to this four-pool model, parameters such as the amount of bound water in the MT pools (MT-Slow pool and MT-Fast pool) and the water exchange rate are set, taking into account the MT effect.

[0024] Here, the relaxation times T1 and T2 of the Slow pool are defined as "T1s" and "T2s." The relaxation times T1 and T2 of the Fast pool are defined as "T1f" and "T2f." The relaxation times T1 and T2 of the MT-Slow pool are defined as "T1MTs" and "T2MTs." The relaxation times T1 and T2 of the MT-Fast pool are defined as "T1MTf" and "T2MTf."

[0025] In addition, the proton ratio of the MT-Slow pool is defined as "fMTs" and the proton ratio of the MT-Fast pool is defined as "fMTf" when the total is set to 1. Therefore, the proton ratio of the Slow pool can be expressed as "1-f-fMTs-fMTf".

[0026] Furthermore, the exchange rate from the Slow pool to the Fast pool is defined as "ksf". The exchange rate from the Fast pool to the Slow pool is defined as "kfs". The exchange rate from the Slow pool to the MT-Slow pool is defined as "kMTs", and the exchange rate from the Fast pool to the MT-Fast pool is defined as "kMTf". From detailed balancing, the exchange rate from the MT-Slow pool to the Slow pool and the exchange rate from the MT-Fast pool to the Fast pool can be expressed as "kMTs × (1 - f - fMTs - fMTf) / fMTs" and "kMTf f / fMTf", respectively.

[0027] Thus, the four-pool model has 14 independent parameters, but it is generally known that these can be approximated as "T1MTs = T1s" and "T1MTf = T1f." In the case of the brain, the MT pools are water molecules tightly bound to macromolecules, and their movement is strongly restricted. Therefore, their T2 values, i.e., T2MTs and T2MTf, can be set as fixed values ​​and excluded from the independent parameters. In this case, the total number of parameters can be reduced by four, bringing the total to 10. In this embodiment, the plurality of free water bodies is not limited to the two types of free water bodies that are relatively free to move and the two types of free water bodies whose movement is restricted, but may include other types of free water bodies. For example, free water bodies in the head can be broadly divided into three types: intercellular fluid, intracellular fluid, and myelin water, and the plurality of free water bodies may include these three types. Furthermore, in this embodiment, a four-pool model is assumed as the multi-pool model, but this is not limited to this, and it goes without saying that the number of pools can be increased or decreased depending on how many types of tissues with significantly different shapes are present in the biological tissue.

[0028] Next, the operation of the medical data processing apparatus according to the first embodiment will be described with reference to the flowchart of FIG. In step S301, the processing circuitry 2 acquires imaging data using the acquisition function 21. The imaging data is collected by successively performing a plurality of spoiled gradient echo imaging operations and coherent gradient echo imaging operations while changing the flip angle. Here, the imaging data is considered to be N pieces of imaging data generated by sparsely sampling the object for each RF shot, where N is an integer equal to or greater than 2. In this embodiment, sparse sampling refers to sampling at intervals that are thinner than the normal sampling interval, or sampling with a number of samples that is smaller than the normal number of samples that should be acquired.

[0029] In step S302, the processing circuit 2 generates a sparsely sampled low-rank approximate image set from the imaging data using the generation function 22. The generation function 22 uses M preset sets of N weighting coefficients to multiply the N first data by the corresponding weighting coefficient for each set and add the results, thereby performing low-rank approximation on the N first data sets and generating M sparsely sampled low-rank approximate image sets, where M is an integer less than N.

[0030] In step S303, the processing circuit 2 uses the generation function 22 to estimate a full-sampled low-rank approximate image from the sparse-sampled low-rank approximate image using the first trained model. The first trained model is a model in which a first neural network is trained so that a set of M sparse-sampled low-rank approximate images is input and a set of M full-sampled low-rank approximate images is output, obtained by low-rank approximating N full-sampled data generated by full-sampling the subject in the same manner as the imaging data. Note that, hereinafter, when there is no distinction between full sampling and sparse sampling, the term may simply be referred to as a low-rank approximate image set. In this embodiment, full sampling refers to sampling at a sampling interval that should normally be performed or sampling with the number of samples that should normally be obtained.

[0031] In step S304, the processing circuitry 2 performs optimization processing for data consistency on the full-sampled low-rank approximation image using the optimization function 24. Data consistency refers to the consistency between the k-space data obtained when the estimated full-sampled low-rank approximation image is captured and the actually measured k-space data. Specifically, the optimization processing can be performed by alternately repeating estimation of the full-sampled low-rank approximation image using the first trained model and estimation of the full-sampled low-rank approximation image from the captured data by a back-projection method using the conjugate gradient (CG) method. More specifically, the obtained full-sampled low-rank approximation image is inversely transformed to generate check k-space data, which is sparsely sampled k-space data for consistency check, and it is determined whether the value of an error function including an evaluation of the difference between the check k-space data and the k-space data of the imaging data acquired in step S301 is equal to or less than a threshold. If the value of the error function is equal to or less than the threshold, it is determined that the convergence condition is satisfied, and the process proceeds to step S305. If the value of the error function is greater than the threshold, the full-sampled low-rank approximation image is corrected and the same process is repeated until the convergence condition is satisfied.

[0032] Furthermore, for example, convergence can be achieved by repeating the estimation of a full-sampling low-rank approximation image set using the above-mentioned convolutional neural network and the estimation of a full-sampling low-rank approximation image set from imaging data using the back projection method with the CG method, using the ADMM (Alternating Direction Method of Multipliers) method. This method also optimizes data consistency, enabling more reliable estimation of a full-sampling low-rank approximate image. For details of the ADMM method or an approximate ADMM method similar to the ADMM method executed by the optimization function 24, see, for example, the method described in Japanese Patent Application Laid-Open No. 2021-10408.

[0033] In step S305, the processing circuit 2 sets the pixel values ​​(voxel values) of the full-sampling low-rank approximation image set as a low-rank approximation value set using the generation function 22, and estimates parameter values ​​using the second trained model based on the low-rank approximation value set. Details of the second trained model will be described later with reference to Figures 12 and 13, etc. In step S306, the processing circuit 2 determines whether or not the processing of step S305 has been completed for all voxels, for example, by using the reconstruction function 23. If the processing has been completed for all voxels, the process proceeds to step S307, and if there are unprocessed voxels, the process returns to step S305 and repeats the same processing. In step S307, the processing circuit 2 generates a parameter map by causing the reconstruction function 23 to estimate parameters for all voxels.

[0034] In step S303, it is desirable that the number of fully sampled low-rank approximate image sets to be output is the same M (rank M) for the number M (rank M) of sparsely sampled low-rank approximate image sets to be input. However, the output accuracy can be improved by increasing the number of input ranks relative to the number of output ranks. Furthermore, by reducing the number of input ranks relative to the number of output ranks, the output accuracy will be slightly reduced, but it will be possible to increase the memory amount and calculation speed. In this way, the number of input ranks and the number of output ranks can be adjusted appropriately depending on the purpose of use. In other words, it is not limited to outputting all M items. In the following examples, a case will be described in which the number of input ranks and the number of output ranks are the same. Furthermore, the optimization process in step S304 is not essential, and step S305 may be executed after step S303.

[0035] Next, an example of an imaging sequence according to the first embodiment will be described with reference to FIG. The upper part of Fig. 4 shows the time series changes in flip angle according to the number of shots, and the lower part of Fig. 4 shows the time series changes in MR signal value according to the number of shots.

[0036] Here, spoiled gradient echo (hereinafter also referred to as spoiled GRE) and coherent gradient echo (hereinafter also referred to as coherent GRE) are performed as imaging sequences. In spoiled GRE, the transverse magnetization is spoiled after each RF excitation (each shot), but RF spoiling can also be combined. RF spoiling is a technique that spoils the transverse magnetization by modulating the RF phase.

[0037] In spoiled GRE imaging, the transverse magnetization is spoiled for each shot, so the MR signal contains almost no information on transverse magnetization relaxation, but mainly information on longitudinal magnetization relaxation and B1 distribution. On the other hand, coherent GRE imaging contains all information on longitudinal magnetization relaxation, transverse magnetization relaxation, B1 distribution, and in some cases B0 distribution. Therefore, by combining these two, it is possible to accurately separate longitudinal magnetization relaxation, transverse magnetization relaxation, B1 distribution, and in some cases B0 distribution.

[0038] As an example of coherent GRE, FISP (Fast Imaging with Steady-state free Precession) is assumed. Coherent GRE is not limited to FISP; it can also be an imaging sequence that performs full rewind, such as bSSFP (Balanced Steady-State Free Precession). However, in this case, the static magnetic field distribution (B0 distribution) is affected, so the B0 distribution itself must also be mapped. To eliminate the influence of the B0 distribution, a method in which gradient spoiling is applied in one direction, such as FISP, can be used. For these reasons, the following describes the use of FISP as coherent GRE imaging.

[0039] In the imaging sequence shown in Figure 4, the first half is imaged using spoiled GRE and the second half is imaged using FISP. An IR pulse (adiabatic 180-degree pulse) is applied at the very beginning to invert the longitudinal magnetization by 180 degrees. At the end of the spoiled GRE, a short-time image is taken with a large flip angle to saturate the longitudinal magnetization and bring it close to zero. This suppresses magnetization oscillations at the start of FISP imaging, improving analysis accuracy. The TR / TE ratio can be fixed throughout the entire sequence. For example, TR = 7.0 ms and TE = 3.5 ms can be set. Note that the TR / TE ratio may be changed for each image.

[0040] In comparison, analysis is not possible with spoiled GRE imaging alone, as transverse magnetization relaxation information is not available, and analysis is difficult with coherent GRE imaging alone, as it is difficult to separate the information.

[0041] Information about the MT effect can be obtained from the on-resonance MT effect induced by RF pulses. In spoiled GRE and coherent GRE, imaging is performed with different flip angles. Changing the flip angle requires changing the B1 intensity, which naturally results in imaging with different on-resonance MT effect intensities. To obtain more detailed information about the MT effect, imaging that actively changes the MT effect can be performed. In both cases, the magnetization state of free water undergoing water exchange can also be changed through the MT effect, allowing for more accurate parameter estimation of the entire multicompartment microstructure model. This achieves accuracy that cannot be achieved without considering the MT effect. Furthermore, actively changing the MT effect increases the amount of data obtained, enabling even greater accuracy.

[0042] Next, another example of the imaging sequence according to the first embodiment will be described with reference to FIG. The top row of Figure 5 shows the time series changes in flip angle depending on the number of shots. The middle row of Figure 5 shows the time series changes in RF pulse width depending on the number of shots. The bottom row of Figure 5 shows the time series changes in MR signal value depending on the number of shots. The imaging sequence is the same as in Figure 4.

[0043] To perform imaging that alters the influence of the MT effect, imaging can be performed by changing the RF pulse width. While the flip angle is proportional to the integral of the RF magnetic field strength, the on-resonance MT effect is proportional to the integral of the RF power. On the other hand, because RF power is proportional to the square of the RF magnetic field strength, halving the RF pulse width doubles the MT effect even with the same flip angle. Thus, even with the same flip angle, imaging with different RF pulse widths can yield different MT effects. This allows for more sensitive extraction of information about the MT effect. To obtain more detailed MT information, an off-resonance MT pulse can be inserted before the on-resonance RF pulse used for excitation. Furthermore, by combining imaging with different off-resonance frequencies of the applied off-resonance MT pulse, even more detailed information about the MT effect can be obtained. For this reason, using an off-resonance MT pulse is desirable, especially when it is necessary to independently obtain the T1 and T2 parameter maps of the MT pool, i.e., bound water.

[0044] Next, an example of radial scanning assumed in this embodiment will be described with reference to FIG. Figure 6 shows the radial scan trajectory in k-space, with the horizontal axis being kx and the vertical axis being ky. For each shot, the angle of the radial scan spokes is increased (rotated) by the golden angle (2π / (1+√5)). However, this is just an example, and the way the spoke angle is set is not limited to the golden angle. The spoke angle may also be changed in other ways.

[0045] It is desirable for the k-space trajectory to have the spokes pass through the center of k-space for each shot. This is because, in order to obtain a low-rank approximation image set from the acquired imaging data, it is desirable for the signal quality for each shot to be as consistent as possible. Therefore, it is desirable to use a radial trajectory, a spiral trajectory, a variable density spiral (VDS) trajectory, etc., but this is not limited to these, and other scanning methods can also be used. Note that scanning methods and k-space trajectories in which the spokes do not pass through the center of k-space for each shot can also be used.

[0046] Next, the concept of sparsely sampled low-rank approximate images and an example of their generation will be described with reference to Fig. 7. Here, an image is assumed as data. In each shot (1, 2, 3, 4, . . .), MR signals are collected along the trajectory of one spoke in the radial scan, and k-space data is acquired by filling the k-space. Specifically, in the first shot, MR signals are collected along the spoke sloping downward to the right, and in the second shot, MR signals are collected along the spoke sloping upward to the right. In this way, sparsely sampled k-space data 601 is acquired for each shot.

[0047] For sparsely sampled k-space data 701 for each shot, a weighting factor 702, W ij and add them together to generate i k-space data 703. i and j are integers equal to or greater than 1. ij A plurality of values ​​may be prepared corresponding to the number of ranks, and represents a weighting coefficient by which the k-space data of the j-th shot is multiplied when creating a low-rank approximation image of rank i.

[0048] By performing an inverse Fourier transform using an IFFT (Inverse Fast Fourier Transform) or the like on each of the i k-space data 703, i sparsely sampled low-rank approximation images 704 are generated.

[0049] To calculate the weighting coefficients 702, a set of transient signals obtained by imaging using the aforementioned sequence for various parameter values ​​is first simulated based on a four-pool model. The set of transient signals thus created can be subjected to principal component analysis to calculate the weighting coefficients for the required ranks. Ranks are numbered 1, 2, 3, and so on, from largest to smallest singular value in principal component analysis (PCA). Instead of principal component analysis, multivariate analysis such as singular value decomposition (SVD) and dimensionality reduction processing such as nonnegative matrix factorization (NMF) may also be applied. This allows the i-set of weighting coefficients for low-rank approximation to be calculated.

[0050] Although Figure 7 shows an example of a radial scan, even for trajectories using other scanning methods, a similar method can be used to obtain i k-space data by multiplying all k-space data obtained for each shot by a weighting coefficient and adding them together.

[0051] However, the low-rank approximation image created by the above method is essentially a sparsely sampled low-rank approximation image because it contains only one trajectory signal (spoke) per shot. As a result, the image contains a large number of artifacts, and even if you try to obtain a parameter map directly from the sparsely sampled low-rank approximation image, you may not be able to achieve the required accuracy. Therefore, it is preferable to use a fully sampled low-rank approximation image to obtain a parameter map.

[0052] Next, the concept of a fully sampled low-rank approximation image and an example of its generation will be described in detail with reference to FIG. Figure 8 shows a conceptual diagram of a method for creating a fully sampled low-rank approximation image. In radial imaging, for example, if the image of each shot (1, 2, 3, 4, . . .) can be fully sampled with 400 spokes, imaging based on the above sequence is performed 400 times while changing the angle of the initial spoke. By combining these, 400 trajectory signals (spokes) are collected for each shot, and fully sampled k-space data 801 for each shot can be obtained.

[0053] For the full sampling k-space data 801 of each shot, a weighting coefficient 802, W ij 6. By multiplying and adding the weighting coefficients 802, i pieces of k-space data 803 are generated. The weighting coefficients 802 are the same as the weighting coefficients 702 in Fig. 6. An inverse Fourier transform such as IFFT is performed on each of the i pieces of k-space data 803, thereby generating i pieces of full-sampling low-rank approximation images 804.

[0054] However, in this case, the imaging time is 400 times longer than in the sparse sampling case of Fig. 7. Therefore, in this embodiment, a trained model such as a neural network is used to estimate a fully sampled low-rank approximate image from a sparsely sampled low-rank approximate image.

[0055] Next, the learning of the first neural network will be described with reference to FIG. A convolutional neural network is assumed as the first neural network. When training the convolutional neural network, a simulation is performed in which a large number of numerical phantoms configured with a four-pool model are imaged using the above-mentioned sequence. Through this simulation, a large number of pairs of fully sampled low-rank approximation image sets and sparsely sampled low-rank approximation image sets with the same rank number are prepared as training data. Note that the sparsely sampled low-rank approximation image sets can be generated through a simulation that assumes imaging one spoke in one shot.

[0056] The processing circuit 2 uses the learning function 25 to train a convolutional neural network using the sparsely sampled low-rank approximate image set from the training data as input data and the full-sampled low-rank approximate image set as ground truth data. In the following embodiment, the spoiled GRE imaging data and the FISP imaging data are individually subjected to principal component analysis, and weighting coefficients of ranks 1 to 13 are used to create a total of 26 low-rank approximate image sets. However, principal component analysis may also be performed on all the data at once. In the example of FIG. 9 , a set of 13 sparsely sampled low-rank approximate images from spoiled GRE imaging and 13 sparsely sampled low-rank approximate images from FISP imaging are used as input data. A set of 13 full-sampled low-rank approximate images from spoiled GRE imaging and 13 full-sampled low-rank approximate images from FISP imaging are used as ground truth data. Note that in the example of FIG. 9 , pixel values ​​rapidly decrease as the number of ranks increases, so data normalized so that the variance of pixel values ​​at each rank is 1 is used. In addition, the sparsely sampled low-rank approximation image set is normalized to include spatial noise, so it is displayed with an appropriate scaling factor. It is recommended to normalize appropriately during training.

[0057] In the training, the parameters of the convolutional neural network may be updated and the network may be optimized so that a loss function designed using, for example, the mean squared error between the correct data and the first neural network or cross entropy is minimized. Note that a general machine learning training method may be used as the neural network training method, and therefore a detailed description thereof will be omitted here. Upon completion of the training, a first trained model of the first neural network is generated. In addition, U-Net, DenseNet, etc. are assumed as examples of convolutional neural networks, but any network structure may be used as long as it is a neural network used in the field of machine learning.

[0058] Next, the inference (use) of the first neural network will be described with reference to FIG. During inference, a sparsely sampled low-rank approximate image set is input to the trained model, and a fully sampled low-rank approximate image set is output, as shown in Figure 10. Although each image in the sparsely sampled low-rank approximate image set has high noise, using the trained model makes it possible to obtain a fully sampled low-rank approximate image set with reduced noise.

[0059] Next, the process of generating a parameter map from a set of fully sampled low-rank approximation images will be described with reference to FIGS. The reconstruction function 23 generates parameter values ​​by applying the second trained model to each pixel value, i.e., each voxel, of the fully sampled low-rank approximation image. A parameter map can be generated based on the parameter values ​​of all voxels.

[0060] First, we will explain the training process of the second neural network to obtain the second trained model. FIG. 11 is a conceptual diagram of generating a low-rank approximation set from a transient signal. In Fig. 11, a four-pool model is used to perform a single voxel simulation using the imaging sequence shown in Fig. 4 or 5. Transient signals are generated for a large number of parameter value sets obtained by the simulation.

[0061] The transient signal obtained in the simulation is calculated using the weighting coefficient W ij Using this, the MR signal value of each shot is weighted by a weighting factor W ij By multiplying and adding them together, we obtain a low-rank approximation of the required number of ranks. Here, we use W as the weighting factor for rank i. ijThe low-rank approximation process is performed by multiplying the corresponding shot j by (j=1, 2...) and adding them up to obtain a low-rank approximation value. In other words, the set of voxel values ​​of 13 (ranks 1 to 13) full-sampling low-rank approximation images matches the set of 13 (ranks 1 to 13) low-rank approximations. Figure 11 shows an example graph of a low-rank approximation value set. In the graph, the vertical axis represents the low-rank approximation value, and the horizontal axis represents the rank number. A graph of a set of low-rank approximations of ranks 1 to 13 for spoiled GRE (SP-GRE) and a graph of a set of low-rank approximations of ranks 1 to 13 for FISP are shown. Together, they are simply referred to as the low-rank approximation value set. A large number of low-rank approximation value sets and parameter value sets generated in this way are prepared.

[0062] Next, FIG. 12 shows the learning process of the second neural network. The processing circuit 2 uses the low-rank approximation value set as input data and the parameter value set as correct answer data to train the second neural network by the learning function 25. It is desirable to properly normalize the input and output before training.

[0063] The second neural network is assumed to be, for example, a tightly coupled neural network, but is not limited to this and any network in the field of machine learning may be used. Furthermore, the training method for the convolutional neural network may be a general machine learning training method. Upon completion of the training, a second trained model of the second neural network is generated.

[0064] Next, FIG. 13 shows the second neural network during inference (use). During inference, as shown in Fig. 13, a pixel value set of a fully sampled low-rank approximation image based on imaging data (i.e., a low-rank approximation value set) is input to the second trained model, and an estimated parameter value set is output. This places the parameter values ​​of the four-pool model in one-to-one correspondence, allowing the parameter values ​​to be estimated.

[0065] Next, an example of a calibration curve for the estimated values ​​of each parameter is shown in FIG. Figure 14 shows calibration curves for 11 parameter values ​​(T1s, T1f, T2s, T2f, ksf, f, kMTs, fMTs, kMTf, fMTf, B1), which are the 10 parameters mentioned above in the four-pool model shown in Figure 2 plus the B1 value. Here, B1 is the RF magnetic field distribution. In each calibration curve graph, the vertical axis represents the estimated value and the horizontal axis represents the correct value. As shown in Figure 14, it can be seen that good estimation was achieved overall.

[0066] Next, an example of the generated parameter map will be described with reference to FIG. In addition to the parameter maps for the parameter values ​​shown in Figure 14, Figure 15 shows a proton density map, a parameter map for the proton ratio of the slow pool (1-f-fMTs-fMTf), and a parameter map for the proton ratio outside the slow pool (f+fMTs+fMTf). A parameter value set is generated by inputting the pixel value set for each voxel of the fully sampled low-rank approximation image (i.e., the low-rank approximation value set) obtained by the imaging sequence shown in Figure 4 into the second trained model. A parameter map corresponding to 11 parameters is generated using the parameter value sets for all voxels. This generates the parameter map shown in Figure 15. The example in Figure 15 is an example in which imaging was performed in approximately one minute and a 256 x 256 pixel two-dimensional image was analyzed in approximately 10 seconds, enabling high-speed processing.

[0067] Although the above description has been given using a two-dimensional image as an example, a parameter map can be generated in the same manner for three-dimensional imaging.

[0068] Furthermore, other images can be generated from the full-sampled low-rank approximate image. An example of generating other images based on the full-sampled low-rank approximate image is shown in FIG. FIG. 16 shows an example of generating a T1 map obtained by IR-GRE imaging and a T2 map obtained by CPMG (Carr Purcell Meiboom Gill) imaging. In IR-GRE, longitudinal magnetization relaxation curves can be obtained for each voxel by performing multiple scans with different IR times, and single-compartment T1 can be obtained by fitting these curves to a single exponential function. In CPMG imaging, transverse magnetization relaxation curves can be obtained for each voxel by performing multiple scans with different TE, and single-compartment T2 can be obtained by fitting these curves to a single exponential function.

[0069] These are the apparent relaxation times, T1a and T2a, obtained by single compartment analysis. However, since single compartment analysis is currently the mainstream method for quantifying T1 and T2 using MRI imaging, it may be desirable to obtain T1a and T2a for comparison with conventional techniques.

[0070] In this case, if the parameter map of the 4-pool model has been obtained, it is possible to use the 4-pool model and parameter map to simulate multiple IR-GRE scans with different IR times to obtain relaxation curves for each voxel, and then fit the curves to a single exponential function to obtain T1a.Similarly, it is possible to simulate multiple CPMG scans with different TEs to obtain relaxation curves for each voxel, and then fit the curves to a single exponential function to obtain T2a. In other words, once the parameter values ​​of the 4-pool model are determined, it is possible to calculate T1a and T2a for a specific sequence, such as T1 calculated by IR-GRE and T2 calculated by CPMG.On the other hand, as already mentioned, the parameter values ​​of the 4-pool model can be calculated for each pixel from a fully sampled low-rank approximation image. Therefore, a neural network can be trained so that the pixel value set (low-rank approximation value set) of the fully sampled low-rank approximation image is used as input and T1a and T2a are used as outputs.

[0071] That is, a large number of low-rank approximation value sets and T1a and T2a data sets are created, and the second neural network is made to learn the correspondence between them. By using the second trained model, it is possible to directly create a T1 map when imaging is performed with MP2REGE or IR-GRE, a T2 map when imaging is performed with CPMG, and the like, from imaging data of the imaging sequence according to the first embodiment.

[0072] By doing the same, the medical data processing apparatus according to the first embodiment can also create various weight images when imaging is performed using a specific imaging method and imaging parameters.

[0073] According to the first embodiment described above, multiple spoiled GRE imaging and coherent GRE imaging are performed consecutively while changing the flip angle to obtain imaging data. A first trained model is used to generate a fully sampled low-rank approximation image from a sparsely sampled low-rank approximation image. A second trained model is used to estimate parameter values ​​from a low-rank approximation value set corresponding to a pixel value set for each voxel of the fully sampled low-rank approximation image. A parameter map is reconstructed by estimating parameter values ​​for all voxels. This allows us to obtain an accurate quantitative map that takes into account the MT effect and B1 distribution, as well as a quantitative map of the MT effect itself. Furthermore, we can perform high-speed, reliable estimation with a practical resolution.

[0074] (Modification of the first embodiment) In the first embodiment, a parameter map of a multi-pool model characterized by containing multiple free water molecules and bound water molecules that exchange water with the free water molecules is generated by obtaining a low-rank approximation image set based on an imaging sequence such as that shown in Figure 4 or Figure 5.

[0075] On the other hand, in a modification of the first embodiment, spoiled GRE imaging and coherent GRE imaging are performed as an imaging sequence while continuously changing the flip angle for each shot. Even when imaging is performed using a method in which the flip angle is changed for each shot, as in the first embodiment, a parameter map can be generated using a multi-pool model characterized by including multiple free water molecules and bound water molecules that exchange water with the free water molecules.

[0076] (Second embodiment) In the second embodiment, the overall configuration of a magnetic resonance imaging apparatus including the medical data processing apparatus according to the above-described embodiment will be described with reference to Fig. 17. Fig. 17 is a diagram showing the configuration of a magnetic resonance imaging apparatus 100 in this embodiment.

[0077] 17 , the magnetic resonance imaging apparatus 100 includes a static magnetic field magnet 101, a gradient magnetic field coil 103, a gradient magnetic field power supply 105, a bed 107, a bed control circuit 109, a transmitting coil 113, a transmitting circuit 115, a receiving coil 117, a receiving circuit 119, a sequence control circuit 121, a bus 123, an interface 125, a display 127, a storage device 129, and a processing circuit 131. The magnetic resonance imaging apparatus 100 may include a hollow cylindrical shim coil between the static magnetic field magnet 101 and the gradient magnetic field coil 103. A group of units, represented by the transmitting circuit 115, the receiving circuit 119, and the sequence control circuit 121, that capture images of a subject based on an imaging sequence and collect imaging data is also referred to as an acquisition unit.

[0078] The static magnetic field magnet 101 is a magnet formed in a hollow, approximately cylindrical shape. Note that the static magnetic field magnet 101 is not limited to an approximately cylindrical shape, and may be configured in an open shape. The static magnetic field magnet 101 generates a uniform static magnetic field in the internal space. For example, a superconducting magnet or the like is used as the static magnetic field magnet 101.

[0079] The gradient magnetic field coil 103 is a coil formed in a hollow cylindrical shape. The gradient magnetic field coil 103 is placed inside the static magnetic field magnet 101. The gradient magnetic field coil 103 is formed by combining three coils corresponding to the X, Y, and Z axes that are orthogonal to each other. The Z-axis direction is the same as the direction of the static magnetic field. The Y-axis direction is the vertical direction, and the X-axis direction is the direction perpendicular to the Z and Y axes. The three coils in the gradient magnetic field coil 103 are individually supplied with current from a gradient magnetic field power supply 105, and generate gradient magnetic fields whose magnetic field strength changes along each of the X, Y, and Z axes.

[0080] The gradient magnetic fields of the X, Y, and Z axes generated by the gradient coil 103 form, for example, a frequency encoding gradient magnetic field (also called a readout gradient magnetic field), a phase encoding gradient magnetic field, and a slice selection gradient magnetic field. The slice selection gradient magnetic field is used to determine the imaging cross section. The phase encoding gradient magnetic field is used to change the phase of the MR signal depending on the spatial position. The frequency encoding gradient magnetic field is used to change the frequency of the MR signal depending on the spatial position.

[0081] The gradient magnetic field power supply 105 is a power supply device that supplies current to the gradient magnetic field coil 103 under the control of the sequence control circuit 121 .

[0082] The bed 107 is a device equipped with a tabletop 1071 on which the subject P is placed. Under the control of a bed control circuit 109, the bed 107 inserts the tabletop 1071 on which the subject P is placed into a bore 111. The bed 107 is installed in an examination room in which the magnetic resonance imaging apparatus 100 is installed, for example, so that the longitudinal direction is parallel to the central axis of the static magnetic field magnet 101.

[0083] The bed control circuit 109 is a circuit that controls the bed 107, and drives the bed 107 in response to an instruction from the operator via the interface 125, thereby moving the tabletop 1071 in the longitudinal direction and the up-down direction.

[0084] The transmitting coil 113 is an RF coil arranged inside the gradient magnetic field coil 103. The transmitting coil 113 receives RF (Radio Frequency) pulses from a transmitting circuit 115 and generates a transmitting RF wave corresponding to a high frequency magnetic field. The transmitting coil 113 is, for example, a whole body coil (hereinafter referred to as a WBC). The WBC may be used as a transmitting / receiving coil. A cylindrical RF shield is installed between the WB coil and the gradient magnetic field coil 103 to magnetically separate these coils.

[0085] The transmission circuit 115 supplies an RF pulse corresponding to the Larmor frequency or the like to the transmission coil 113 under the control of the sequence control circuit 121 .

[0086] The receiving coil 117 is an RF coil arranged inside the gradient magnetic field coil 103. The receiving coil 117 receives MR signals emitted from the subject P by a high frequency magnetic field. The receiving coil 117 outputs the received MR signals to a receiving circuit 119. The receiving coil 117 is, for example, a coil array having one or more, typically a plurality of coil elements. The receiving coil 117 is, for example, a phased array coil.

[0087] The receiving circuit 119 generates a digital MR signal, which is digitized complex data, based on the MR signal output from the receiving coil 117 under the control of the sequence control circuit 121. Specifically, the receiving circuit 119 performs various signal processing on the MR signal output from the receiving coil 117, and then performs analog-to-digital (A / D) conversion on the data that has been subjected to various signal processing. The receiving circuit 119 samples the A / D converted data. As a result, the receiving circuit 119 generates a digital MR signal (hereinafter referred to as MR data). The receiving circuit 119 outputs the generated MR data to the sequence control circuit 121.

[0088] The sequence control circuit 121 controls the gradient magnetic field power supply 105, the transmission circuit 115, the reception circuit 119, etc. in accordance with the examination protocol output from the processing circuit 131, and performs imaging of the subject P. For example, spoiled GRE imaging and FISP imaging shown in FIG. 4 or 5 are alternately and repeatedly performed. Also, spoiled GRE imaging and FISP imaging with a different flip angle for each shot are alternately and repeatedly performed.

[0089] The examination protocol has various pulse sequences according to the examination, and defines the magnitude of the current supplied to the gradient magnetic field coil 103 by the gradient magnetic field power supply 105, the timing at which the gradient magnetic field power supply 105 supplies the current to the gradient magnetic field coil 103, the magnitude of the RF pulse supplied to the transmission coil 113 by the transmission circuit 115, the timing at which the RF pulse is supplied to the transmission coil 113 by the transmission circuit 115, the timing at which the MR signal is received by the reception coil 117, etc.

[0090] The bus 123 is a transmission path for transmitting data among the interface 125, the display 127, the storage device 129, and the processing circuit 131. Various biosignal measuring devices, external storage devices, various modalities, etc. may be appropriately connected to the bus 123 via a network, etc. For example, an electrocardiograph (not shown) is connected to the bus as a biosignal measuring device.

[0091] The interface 125 has circuits for receiving various instructions and information input from an operator. The interface 125 has circuits related to input devices such as a pointing device such as a mouse or a keyboard. Note that the circuits included in the interface 125 are not limited to circuits related to physical operating components such as a mouse and a keyboard. For example, the interface 125 may have an electrical signal processing circuit that receives electrical signals corresponding to input operations from an external input device provided separately from the magnetic resonance imaging apparatus 100 and outputs the received electrical signals to various circuits.

[0092] The display 127 displays various magnetic resonance images (MR images) generated by the image generation function, various information related to imaging and image processing, and the like, under the control of the system control function 1311 in the processing circuitry 131. The display 127 is, for example, a CRT display, a liquid crystal display, an organic EL display, an LED display, a plasma display, or any other display or monitor known in the art.

[0093] The storage device 129 stores MR data filled in the k-space via the image generation function 1313, image data generated by the image generation function 1313, etc. The storage device 129 stores various examination protocols, imaging conditions including a plurality of imaging parameters defining the examination protocols, etc. The storage device 129 stores programs corresponding to various functions executed by the processing circuitry 131. The storage device 129 is, for example, a random access memory (RAM), a semiconductor memory element such as a flash memory, a hard disk drive, a solid state drive, an optical disk, etc. The storage device 129 may also be a drive or the like that reads and writes various information from and to a portable storage medium such as a CD-ROM drive, a DVD drive, or a flash memory.

[0094] The processing circuitry 131 has hardware resources such as a processor, a read-only memory (ROM), a RAM, and the like, which are not shown, and performs overall control of the magnetic resonance imaging apparatus 100. The processing circuitry 131 has a system control function 1311, an image generation function 1313, an acquisition function 21, a generation function 22, a reconstruction function 23, and an optimization function 24. The acquisition function 21, the generation function 22, the reconstruction function 23, and the optimization function 24 are similar to the functions included in the processing circuitry 2 of the medical data processing apparatus 1 according to the above-described embodiment, and therefore description thereof will be omitted here.

[0095] The various functions of the processing circuitry 131 are stored in the storage device 129 in the form of programs executable by a computer. The processing circuitry 131 is a processor that realizes the functions corresponding to these various functions by reading and executing the programs corresponding to these various functions from the storage device 129. In other words, the processing circuitry 131 in a state in which each program has been read out has the multiple functions shown in the processing circuitry 131 in FIG. 1.

[0096] 1, it has been explained that these various functions are realized by a single processing circuit 131, but it is also possible to configure the processing circuit 131 by combining multiple independent processors, and have each processor execute a program to realize the function. In other words, it is possible that each of the above-mentioned functions is configured as a program and one processing circuit executes each program, or that a specific function is implemented in a dedicated, independent program execution circuit.

[0097] The term "processor" used in the above description refers to circuits such as a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), an Application Specific Integrated Circuit (ASIC), a programmable logic device (e.g., a Simple Programmable Logic Device (SPLD), a Complex Programmable Logic Device (CPLD), and a Field Programmable Gate Array (FPGA)).

[0098] If the processor is a CPU, for example, it performs its functions by reading and executing a program stored in a memory circuit. On the other hand, if the processor is an ASIC, for example, the program is not stored in a memory circuit, but the function is directly incorporated into the processor circuit as a logic circuit. Note that the bed control circuit 109, transmission circuit 115, reception circuit 119, sequence control circuit 121, etc. are also similarly configured by electronic circuits such as the processor.

[0099] The processing circuitry 131 controls the magnetic resonance imaging apparatus 100 using a system control function 1311. Specifically, the processing circuitry 131 reads out a system control program stored in the storage device 129, loads it on memory, and controls each circuit of the magnetic resonance imaging apparatus 100 in accordance with the loaded system control program. For example, the processing circuitry 131 reads out an examination protocol from the storage device 129 using the system control function 1311 based on imaging conditions input by the operator via the interface 125. The processing circuitry 131 may also generate the examination protocol based on the imaging conditions. The processing circuitry 131 transmits the examination protocol to the sequence control circuit 121 and controls imaging of the subject P.

[0100] The processing circuitry 131 applies excitation pulses and applies gradient magnetic fields in accordance with an excitation pulse sequence using a system control function 1311. After executing the excitation pulse sequence using the system control function 1311, the processing circuitry 131 collects MR signals from the subject P in accordance with a data collection sequence, which is a pulse sequence for collecting various types of data, and generates MR data.

[0101] The processing circuitry 131 fills the MR data along the readout direction of the k-space according to the strength of the readout gradient magnetic field using the image generation function 1313. The processing circuitry 131 generates an MR image by performing a Fourier transform on the MR data filled in the k-space. For example, the processing circuitry 131 can generate an absolute magnitude image from complex MR data. The processing circuitry 131 can also generate a phase image using real and imaginary part data of the complex MR data. The processing circuitry 131 outputs MR images such as the absolute magnitude image and the phase image to the display 127 or the storage device 129.

[0102] According to the second embodiment described above, low-rank approximations equivalent to full sampling can be generated from imaging data acquired by performing spoiled GRE imaging and coherent GRE imaging, and thus, similar to the first embodiment, quantitative values ​​and parameter maps of various parameters can be provided quickly and with high reliability.

[0103] According to at least one of the embodiments described above, it is possible to perform high-speed and highly reliable estimation.

[0104] In addition, each function according to the embodiments can be realized by installing a program for executing the corresponding process on a computer such as a workstation and expanding the program in memory. In this case, the program for causing a computer to execute the corresponding method can be stored and distributed on a storage medium such as a magnetic disk (such as a hard disk), an optical disk (such as a CD-ROM, DVD, or Blu-ray (registered trademark) disk), or a semiconductor memory.

[0105] Although several embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their modifications are included within the scope and spirit of the invention, as well as within the scope of the invention described in the claims and their equivalents. [Explanation of symbols]

[0106] 1 Medical data processing device 2,131 processing circuits 4 Input Interface 6 Communication Interface 8. Memory 21 Acquisition Function 22 Generation function 23 Reconfiguration function 24 Optimization Function 25 Learning Function 100 Magnetic resonance imaging device 101 Static Magnetic Field Magnet 103 Gradient magnetic field coil 105 Gradient magnetic field power supply 107 Sleeper 109 Bed control circuit 111 Bore 113 Transmitting circuit 115 Transmitting Coil 117 Receiving Coil 119 Receiving circuit 121 Sequence control circuit 123 Bus 125 Interface 127 Display 129 Storage device 701 Sparsely sampled k-space data 702,802 weighting factors 703,803 k-space data 704 Sparsely Sampling Low-Rank Approximation Images 801 fully sampled k-space data 804 Fully Sampling Low-Rank Approximation Images 1071 Top plate 1311 System Control Functions 1313 Image generation function

Claims

1. an acquisition unit that acquires a plurality of imaging data obtained by performing spoiled gradient echo imaging and coherent gradient echo imaging at a plurality of flip angles; a generation unit that generates a low-rank approximation image set, which is a set of images obtained by performing a low-rank approximation process that generates an image set the number of which is less than the number of the plurality of imaging data by multiplying each of the plurality of imaging data by the weighting coefficient for each set, the weighting coefficient being less than the number of the plurality of imaging data, and adding the results; and a reconstruction unit that reconstructs a parameter map, which is a distribution image of at least one parameter value among the pool abundance and water exchange rate, proton density, and magnetic field distribution in the multipool model, using the multipool model including a plurality of free water molecules related to water exchange in biological tissues and bound water molecules that exchange with the free water molecules and the low-rank approximation image set; A medical data processing device comprising:

2. The generation unit performing the low-rank approximation process on the plurality of imaging data to generate a first sparsely sampled low-rank approximation image set; 2. The medical data processing device of claim 1, wherein a second sparsely sampled low-rank approximation image set for training is input, and a first trained model is used to train the first trained model to output a full-sampled low-rank approximation image set generated by performing the low-rank approximation processing on fully sampled imaging data, and the full-sampled low-rank approximation image set estimated from the first sparsely sampled low-rank approximation image set is used as the low-rank approximation image set.

3. 3. The medical data processing device according to claim 2, further comprising an optimization unit that performs optimization processing by alternately repeating estimation of a full-sampled low-rank approximation image using the first trained model and estimation of a full-sampled low-rank approximation image from the imaging data.

4. The medical data processing device according to claim 3 , wherein the optimization unit executes the optimization process using an ADMM (Alternating Direction Method of Multipliers) method.

5. the generation unit receives a fully sampled low-rank approximation image for training and uses a second trained model trained to output quantitative values ​​for one or more parameters from the fully sampled low-rank approximation image estimated by the generation unit, to estimate quantitative values ​​for one or more parameters; The medical data processing device according to claim 2 , wherein the reconstruction unit generates the one or more parameter maps based on quantitative values ​​relating to the one or more parameters.

6. 6. The medical data processing device according to claim 2, wherein the reconstruction unit calculates T1 and T2 values ​​based on a single compartment analysis using a sequence different from imaging sequences for the spoiled gradient echo imaging and the coherent gradient echo imaging from the full-sampling low-rank approximation image.

7. 7. The medical data processing device according to claim 1, wherein the imaging data is data obtained by incorporating a technique for modifying a magnetization transfer effect into the spoiled gradient echo imaging and the coherent gradient echo imaging.

8. The medical data processing device according to claim 7 , wherein the technique is to change the pulse width of an RF pulse.

9. The medical data processing device according to claim 7 , wherein the technique is to apply an off-resonance MT pulse.

10. The medical data processing device according to claim 9 , wherein the technique is to change an off-resonance frequency of the off-resonance MT pulse.

11. an acquisition unit that acquires a plurality of imaging data obtained by performing spoiled gradient echo imaging and coherent gradient echo imaging while changing the flip angle for each shot; a generation unit that generates a low-rank approximation image set, which is a set of images obtained by performing a low-rank approximation process that generates an image set the number of which is less than the number of the plurality of imaging data by multiplying each of the plurality of imaging data by the weighting coefficient for each set, the weighting coefficient being less than the number of the plurality of imaging data, and adding the results; and a reconstruction unit that reconstructs a parameter map, which is a distribution image of at least one parameter value among the pool abundance and water exchange rate, proton density, and magnetic field distribution in the multipool model, using the multipool model including a plurality of free water molecules related to water exchange in biological tissues and bound water molecules that exchange with the free water molecules and the low-rank approximation image set; A medical data processing device comprising:

12. acquiring a plurality of imaging data obtained by performing spoiled gradient echo imaging and coherent gradient echo imaging at a plurality of flip angles; generating a low-rank approximation image set, which is a set of images obtained by performing a low-rank approximation process for generating an image set the number of which is less than the number of the plurality of imaging data, by multiplying each of the plurality of imaging data by the weighting coefficient for each set, the weighting coefficient being less than the number of the plurality of imaging data, and adding the results; using a multi-pool model including a plurality of free water molecules related to water exchange in biological tissues and bound water molecules that exchange with the free water molecules and the set of low-rank approximation images, reconstructing a parameter map that is a distribution image of at least one parameter value among the pool abundance and water exchange rate, proton density, and magnetic field distribution in the multi-pool model; Medical data processing method.

13. Computer, an acquisition function for acquiring multiple imaging data obtained by performing spoiled gradient echo imaging and coherent gradient echo imaging at multiple flip angles; a generation function for generating a low-rank approximation image set, which is a set of images obtained by performing a low-rank approximation process for generating an image set the number of which is less than the number of the plurality of imaging data, by using a weighting coefficient for each set that is less than the number of the plurality of imaging data, and adding the results together; and a multi-pool model including a plurality of free water molecules related to water exchange in biological tissues and bound water molecules exchanging with the free water molecules, and the low-rank approximation image set, to realize the reconstruction function of reconstructing a parameter map, which is a distribution image of at least one parameter value selected from the pool abundance and water exchange rate, proton density, and magnetic field distribution in the multi-pool model; Medical data processing program.

14. an acquisition unit that performs spoiled gradient echo imaging and coherent gradient echo imaging on a subject at a plurality of flip angles to acquire a plurality of imaging data; a generation unit that generates a low-rank approximation image set, which is a set of images obtained by performing a low-rank approximation process that generates an image set the number of which is less than the number of the plurality of imaging data by multiplying each of the plurality of imaging data by the weighting coefficient for each set, the weighting coefficient being less than the number of the plurality of imaging data, and adding the results; and a reconstruction unit that reconstructs a parameter map, which is a distribution image of at least one parameter value among the pool abundance and water exchange rate, proton density, and magnetic field distribution in the multipool model, using the multipool model including a plurality of free water molecules related to water exchange in biological tissues and bound water molecules that exchange with the free water molecules and the low-rank approximation image set; A magnetic resonance imaging apparatus comprising:

Images