Magnetic resonance imaging apparatus and method

Abstract

To improve the quality of imaging data collected by MR imaging. [Solution] The magnetic resonance imaging apparatus according to the embodiment comprises an acquisition unit, a generation unit, a calculation unit, and a determination unit. The acquisition unit performs a first MR imaging on a subject to acquire k-space data. The generation unit generates morphological image data representing the morphology of the subject based on the k-space data. The calculation unit calculates phase distribution data representing the spatial distribution including phase information, based on at least the morphological image data. The determination unit determines, based on the phase distribution data, whether or not a change is necessary for a second MR imaging that follows the first MR imaging.

Description

Technical Field

[0001] The embodiments disclosed in this specification and the drawings relate to magnetic resonance imaging apparatuses and methods.

Background Art

[0002] There is a technique for providing assistance information that helps review parameters related to a second imaging based on the quality of data collected in a first imaging. However, this technique is premised on X-ray CT (Computed Tomography) imaging and does not consider the magnetic field uniformity that affects the quality of data collected in MR (Magnetic Resonance) imaging.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] One of the problems to be solved by the embodiments disclosed in this specification and the drawings is to improve the quality of data collected by MR imaging. However, the problems to be solved by the embodiments disclosed in this specification and the drawings are not limited to the above problems. The problems corresponding to the respective effects of each configuration shown in the embodiments described later can also be regarded as other problems.

Means for Solving the Problems

[0005] The magnetic resonance imaging apparatus according to this embodiment comprises an acquisition unit, a generation unit, a calculation unit, and a determination unit. The acquisition unit performs a first MR imaging on a subject to acquire k-space data. The generation unit generates morphological image data representing the morphology of the subject based on the k-space data. The calculation unit calculates phase distribution data representing the spatial distribution including phase information, based on at least the morphological image data. The determination unit determines, based on the phase distribution data, whether or not a modification is necessary for a second MR imaging that follows the first MR imaging. [Brief explanation of the drawing]

[0006] [Figure 1] Figure 1 shows an example of the configuration of a magnetic resonance imaging apparatus according to this embodiment. [Figure 2] Figure 2 shows the generation process of the phase distribution estimation model according to this embodiment. [Figure 3] Figure 3 shows an example of the input and output of a phase distribution estimation model. [Figure 4] Figure 4 shows other examples of input and output for the phase distribution estimation model. [Figure 5] Figure 5 shows other examples of input and output for the phase distribution estimation model. [Figure 6] Figure 6 shows other examples of input and output for the phase distribution estimation model. [Figure 7] Figure 7 shows a typical flow of an MRI examination using the magnetic resonance imaging apparatus according to this embodiment. [Figure 8] Figure 8 shows an example of a display screen for recommended changes. [Figure 9] Figure 9 shows a typical flow of an MRI examination using a magnetic resonance imaging device according to Example 1. [Figure 10] Figure 10 shows a typical flow of an MRI examination using a magnetic resonance imaging device according to Example 2. [Figure 11] Figure 11 shows the generation process of the phase distribution estimation model according to Example 2. [Figure 12]Figure 12 shows a typical flow of an MRI examination using a magnetic resonance imaging device according to Example 3. [Modes for carrying out the invention]

[0007] The magnetic resonance imaging apparatus and method according to this embodiment will be described in detail below with reference to the drawings.

[0008] Figure 1 shows an example of the configuration of a magnetic resonance imaging apparatus 1 according to this embodiment. As shown in Figure 1, the magnetic resonance imaging apparatus 1 includes a stand 11, a bed 13, a gradient magnetic field power supply 21, a transmitting circuit 23, a receiving circuit 25, a shim coil power supply 26, a bed drive device 27, a sequence control circuit 29, and a host computer 50.

[0009] The mounting base 11 includes a static magnetic field magnet 41, a gradient magnetic field coil 43, and a shim coil 49. The static magnetic field magnet 41, the gradient magnetic field coil 43, and the shim coil 49 are housed in the casing of the mounting base 11. The casing of the mounting base 11 has a hollow bore. A transmitting coil 45 and a receiving coil 47 are arranged inside the bore of the mounting base 11.

[0010] The static magnetic field magnet 41 has a hollow, approximately cylindrical shape and generates a static magnetic field inside the approximately cylindrical body. For example, a permanent magnet, a superconducting magnet, or a normal conducting magnet can be used as the static magnetic field magnet 41. Here, the central axis of the static magnetic field magnet 41 is defined as the Z-axis, the axis perpendicular to the Z-axis is defined as the Y-axis, and the axis perpendicular to the Z-axis horizontally is defined as the X-axis. The X-axis, Y-axis, and Z-axis constitute an orthogonal three-dimensional coordinate system.

[0011] The gradient magnetic field coil 43 is mounted inside the static magnetic field magnet 41 and is a hollow, substantially cylindrical coil unit. The gradient magnetic field coil 43 generates a gradient magnetic field by receiving current from the gradient magnetic field power supply 21. More specifically, the gradient magnetic field coil 43 has three coils corresponding to the mutually orthogonal X, Y, and Z axes. These three coils form a gradient magnetic field in which the magnetic field strength changes along each of the X, Y, and Z axes. The gradient magnetic fields along each of the X, Y, and Z axes are combined to form mutually orthogonal slice-selective gradient magnetic field Gs, phase-encoded gradient magnetic field Gp, and frequency-encoded gradient magnetic field Gr in the desired direction. The slice-selective gradient magnetic field Gs is used to arbitrarily determine the imaging cross-section (slice). The phase-encoded gradient magnetic field Gp is ​​used to change the phase of the magnetic resonance signal (hereinafter referred to as the MR signal) according to the spatial position. The frequency-encoded gradient magnetic field Gr is used to change the frequency of the MR signal according to the spatial position. In the following explanation, the gradient direction of the slice selection gradient magnetic field Gs is assumed to be the Z-axis, the gradient direction of the phase encoding gradient magnetic field Gp is ​​assumed to be the Y-axis, and the gradient direction of the frequency encoding gradient magnetic field Gr is assumed to be the X-axis.

[0012] The gradient power supply 21 supplies current to the gradient coil 43 according to the control signal from the sequence control circuit 29. By supplying current to the gradient coil 43, the gradient power supply 21 generates gradient magnetic fields along the X, Y, and Z axes using the gradient coil 43. These gradient magnetic fields are superimposed on the static magnetic field formed by the static magnetic field magnet 41 and applied to the subject P.

[0013] The transmitting coil 45 is, for example, positioned inside the gradient magnetic field coil 43 and receives current from the transmitting circuit 23 to generate high-frequency pulses (hereinafter referred to as RF pulses).

[0014] The transmission circuit 23 supplies a current to the transmission coil 45 in order to apply an RF pulse for exciting target protons such as hydrogen atomic nuclei existing in the subject P to the subject P via the transmission coil 45. The RF pulse oscillates at the resonance frequency specific to the target protons and excites the target protons. An MR signal is generated from the excited target protons and detected by the reception coil 47. The transmission coil 45 is, for example, a whole-body coil (WB coil). The whole-body coil may be used as a transmit-receive coil.

[0015] The reception coil 47 receives the MR signal emitted from the target protons existing in the subject P under the action of the RF pulse. The reception coil 47 has a plurality of reception coil elements capable of receiving the MR signal. The received MR signal is supplied to the reception circuit 25 via wire or wirelessly. Although not shown in FIG. 1, the reception coil 47 has a plurality of reception channels mounted in parallel. Each reception channel has a reception coil element for receiving the MR signal and an amplifier for amplifying the MR signal, etc. The MR signal is output for each reception channel. The total number of reception channels and the total number of reception coil elements may be the same, or the total number of reception channels may be more or less than the total number of reception coil elements.

[0016] The reception circuit 25 receives the MR signal generated from the excited target protons via the reception coil 47. The reception circuit 25 signal-processes the received MR signal to generate a digital MR signal. The digital MR signal is represented in the k-space defined by the spatial frequency. Hereinafter, the digital MR signal will be referred to as k-space data. The k-space data is supplied to the host computer 50 via wire or wirelessly.

[0017] Note that the above transmission coil 45 and reception coil 47 are merely examples. Instead of the transmission coil 45 and the reception coil 47, a transmit-receive coil having a transmission function and a reception function may be used. Also, the transmission coil 45, the reception coil 47, and the transmit-receive coil may be combined.

[0018] The shim coil 49 is a coil unit mounted inside the static magnetic field magnet 41. The shim coil 49 receives current from the shim coil power supply 26 and generates a corrective magnetic field to compensate for the non-uniformity of the static magnetic field. The non-uniformity of the static magnetic field has a 0th-order component, a 1st-order component, a 2nd-order component, and higher-order components of the 3rd order or higher, and the shim coil 49 generates a corrective magnetic field to compensate for all or part of each of these components.

[0019] The shim coil power supply 26 supplies current to the shim coil 49 according to the control signal from the sequence control circuit 29. Specifically, the shim coil power supply 26 receives data on the magnetic field non-uniformity correction value from the sequence control circuit 29 and supplies current to the shim coil 49 corresponding to each component of the corrected magnetic field according to the said magnetic field non-uniformity correction value. This generates a corrected magnetic field from the shim coil 49. The magnetic field non-uniformity correction value refers to the value of the current supplied to the shim coil 49 in order to make the static magnetic field uniform.

[0020] A bed 13 is installed adjacent to the frame 11. The bed 13 has a top plate 131 and a base 133. The subject P is placed on the top plate 131. The base 133 supports the top plate 131 so that it can slide along the X, Y, and Z axes. A bed drive device 27 is housed in the base 133. The bed drive device 27 moves the top plate 131 under control from the sequence control circuit 29. The bed drive device 27 may include any motor, such as a servo motor or a stepping motor.

[0021] The sequence control circuit 29 has a CPU (Central Processing Unit) or MPU (Micro Processing Unit) processor and memory such as ROM (Read Only Memory) or RAM (Random Access Memory) as hardware resources. The sequence control circuit 29 synchronously controls the gradient magnetic field power supply 21, the transmission circuit 23, and the reception circuit 25 based on the imaging conditions set by the processing circuit 51, and performs MR imaging on the subject P according to the imaging conditions to collect k-space data about the subject P. Possible MR imaging methods include image acquisition for morphological image acquisition, shimming imaging for shimming map acquisition, MRS (Magnetic Resonance Spectroscopy) imaging, and CEST (Chemical Exchange Saturation Transfer) imaging.

[0022] The sequence control circuit 29 performs various MR imaging, generating an MR signal from the imaging area set within the subject P. The receiving circuit 25 receives the MR signal via the receiving coil 47 and processes the received MR signal to collect k-space data.

[0023] As shown in Figure 1, the host computer 50 is a computer having a processing circuit 51, memory 53, display 55, input interface 57, and communication interface 59.

[0024] The processing circuit 51 has a processor such as a CPU as a hardware resource. The processing circuit 51 functions as the central hub of the magnetic resonance imaging apparatus 1. For example, the processing circuit 51 implements imaging condition setting function 511, imaging control function 512, acquisition function 513, reconstruction function 514, phase distribution calculation function 515, judgment function 516, selection function 517, and display control function 518 by executing various programs.

[0025] The imaging condition setting function 511 allows the processing circuit 51 to set imaging conditions related to MR imaging. Imaging conditions include imaging position, magnetic field uniformity correction value, timing of imaging relative to contrast agent administration, data acquisition trajectory, pulse sequence, echo time, voxel size, voxel position, and / or water suppression parameters. The imaging position represents the position of the slice or volume. The magnetic field uniformity correction value is the value of the current supplied to the shim coil 49 to make the static magnetic field in the imaging space uniform. The magnetic field uniformity correction value is calculated based on the shimming map. Timing of imaging relative to contrast agent administration indicates whether the imaging is performed before or after the administration of the contrast agent. The data acquisition trajectory, also called k-space trajectory, can be Cartesian or non-Cartesian. Examples of non-Cartesian acquisition include EPI (Echo Planner Imaging) acquisition, radial acquisition, spiral acquisition, 3D radial acquisition, and stack of stars acquisition. Pulse sequences for MR image acquisition include spin echo systems, FE (GRE) systems, and EPI acquisition, while pulse sequences for spectral acquisition include PRESS and MEGA-PRESS. Echo time is the time from the application of the excitation pulse until the transverse magnetization refocuses. Voxel size is the size of the voxel, which is the spatial region targeted for spectral acquisition. Voxel position is the set position of the voxel. Water suppression parameter is the frequency range of the water suppression pulse applied during MRS imaging. The imaging condition setting function 511 is an example of a setting unit.

[0026] Furthermore, the processing circuit 51 can also change the above imaging conditions using the imaging condition setting function 511. This imaging condition setting function 511 is just one example of a modification unit.

[0027] The imaging control function 512 causes the processing circuit 51 to control the sequence control circuit 29 to perform various MR imaging on the subject P and collect k-space data via the receiving circuit 25. The MR imaging according to this embodiment is mainly classified into shimming imaging, first MR imaging, and second MR imaging. Second MR imaging is MR imaging of an object whose need for modification is determined by the determination function 516. Second MR imaging follows first MR imaging. Second MR imaging is imaging for image acquisition (hereinafter referred to as MR image imaging) or imaging for spectrum acquisition (hereinafter referred to as spectral imaging). First MR imaging is MR imaging for collecting data used to determine whether modification of second MR imaging is necessary. This data is, for example, morphological image data. In other words, first MR imaging is MR image imaging. Shimming imaging is imaging for collecting a shimming map. The first and second MR imaging can be applied to both contrast-enhanced imaging, which targets subjects who have been administered a contrast agent, and non-contrast imaging, which targets subjects who have not been administered a contrast agent. Furthermore, if the imaging condition values ​​are changed by the imaging condition setting function 511, the processing circuit 51 performs the second MR imaging on subject P according to the changed values. The imaging control function 512 is an example of an acquisition unit.

[0028] The processing circuit 51 acquires various types of information through the acquisition function 513. For example, the processing circuit 51 acquires k-space data from the receiving circuit 25. The processing circuit 51 may also acquire medical image data collected by other modalities, non-image data recorded in electronic medical records, and / or contrast agent information. The acquisition function 513 is an example of an acquisition unit.

[0029] The reconstruction function 514 allows the processing circuit 51 to reconstruct MR image data based on k-space data. As an example, the processing circuit 51 generates image data representing the morphology of the subject P (hereinafter referred to as magnetic resonance morphology image data) based on k-space data collected by the first MR imaging. As another example, the processing circuit 51 generates a shimming map based on k-space data collected by shimming imaging. A shimming map is an image that represents the spatial distribution of phase difference, which is the difference between two types of phase images with different echo times. The phase difference is proportional to the static magnetic field strength or the resonance frequency. From another perspective, a shimming map can also be described as an image that represents the spatial distribution of the resonance frequency difference from the center frequency. The reconstruction function 514 is an example of a generation unit.

[0030] The phase distribution calculation function 515 allows the processing circuit 51 to calculate phase distribution data, which includes phase information, based on at least magnetic resonance morphology image data. The phase distribution data represents the spatial distribution of physical quantities such as the phase of the magnetization vector or the magnetic susceptibility related to the phase present in the imaging region. The phase distribution calculation function 515 is an example of a calculation unit.

[0031] The determination function 516 allows the processing circuit 51 to determine whether or not to modify the second MR imaging sequence following the first MR imaging sequence, based on the phase distribution data calculated by the phase distribution calculation function 515. The processing circuit 51 determines that no modification is necessary for the second MR imaging sequence if the representative value of the phase distribution data falls below the reference value. A representative value below the reference value indicates that the magnetic field inhomogeneity is relatively small. On the other hand, the processing circuit 51 determines that a modification is necessary for the second MR imaging sequence if the representative value exceeds the reference value. A representative value above the reference value indicates that the magnetic field inhomogeneity is relatively large. The determination function 516 is an example of a determination unit.

[0032] If the processing circuit 51 determines, based on the determination function 516, that a change to the second MR imaging is necessary, the selection function 517 selects the type and / or recommended value of an imaging condition that is recommended to be changed among the imaging conditions for the second MR imaging, which affect the T2 shortening effect, echo time, and / or magnetic field uniformity. Hereinafter, the imaging conditions that are recommended to be changed will be referred to as the recommended change conditions. The selection function 517 is an example of a selection unit.

[0033] The display control function 518 causes the processing circuit 51 to display various information on the display 55. For example, the processing circuit 51 displays the type and / or recommended value of the recommended condition selected by the selection function 517. As another example, the processing circuit 51 displays the judgment result from the judgment function 516.

[0034] Memory 53 is a storage device such as an HDD (Hard Disk Drive), SSD (Solid State Drive), or integrated circuit memory that stores various types of information. Alternatively, memory 53 may be a drive device that reads and writes various types of information to and from portable storage media such as CD-ROM drives, DVD drives, or flash memory.

[0035] The display 55 displays various information using the display control function 518. As the display 55, for example, a CRT display, liquid crystal display, organic EL display, LED display, plasma display, or any other display known in the art can be used as appropriate.

[0036] The input interface 57 includes input devices that receive various commands from the user. These input devices can include keyboards, mice, various switches, touchscreens, touchpads, etc. However, input devices are not limited to those with physical operating components such as mice and keyboards. For example, an electrical signal processing circuit that receives electrical signals corresponding to input operations from an external input device separate from the magnetic resonance imaging apparatus 1 and outputs the received electrical signals to various circuits is also an example of the input interface 57. Furthermore, the input interface 57 may also be a voice recognition device that converts audio signals collected by a microphone into instruction signals.

[0037] The communication interface 59 is an interface that connects the magnetic resonance imaging device 1 to workstations, PACS (Picture Archiving and Communication System), HIS (Hospital Information System), RIS (Radiology Information System), etc., via a LAN (Local Area Network) or the like. The network interface transmits and receives various types of information between the connected workstation, PACS, HIS, and RIS.

[0038] Next, we will explain the process of calculating phase distribution data using the phase distribution calculation function 515.

[0039] The phase distribution calculation function 515 causes the processing circuit 51 to apply the magnetic resonance morphology image data reconstructed by the reconstruction function 514 to a trained model (hereinafter referred to as the phase distribution estimation model) to generate phase distribution data. The phase distribution estimation model is a machine learning model trained on training samples that include magnetic resonance morphology image data as input data and phase distribution data as output data. The phase distribution estimation model is generated, for example, by the processing circuit 51.

[0040] Phase distribution data can be a magnetic field uniformity map, a susceptibility-weighted imaging (SWI) image, or a quantitative susceptibility mapping (QSM) image. A magnetic field uniformity map, like a shimming map, represents the spatial distribution of phase differences, which are the difference between two phase images with different echo times. A susceptibility-weighted image represents the spatial distribution of susceptibility differences. A quantitative susceptibility mapping image represents the spatial distribution of quantitative susceptibility.

[0041] Figure 2 shows the generation process of the phase distribution estimation model according to this embodiment. As shown in Figure 2, the processing circuit 51 trains the untrained model based on a plurality of training samples, which include magnetic resonance morphology image data as input data and phase distribution data as output data. The phase distribution data is used as ground truth data. Hereinafter, the phase distribution data included in the training samples will be referred to as ground truth phase distribution data. The magnetic resonance morphology image data and phase distribution data are data relating to the same subject, and may be data collected in the same examination or data collected in different examinations. Also, the subject may be the same or different across multiple training samples. The magnetic resonance morphology image data and phase distribution data may not be data collected by MR imaging, but may be artificially generated data.

[0042] An untrained model refers to a machine learning model before its network parameters, such as weight parameters and biases, have been optimized. Specifically, neural networks are used as machine learning models. A machine learning model has a combination of input layers, output layers, fully connected layers, convolutional layers, pooling layers, normalization layers, attention mechanisms, and other arbitrary network layers. The network configuration of a machine learning model is not particularly limited; any network configuration that can input and output image data may be adopted.

[0043] The processing circuit 51 updates the parameters of the untrained model using an arbitrary optimization algorithm through supervised learning based on multiple training samples. Stochastic gradient descent, Adam, and other algorithms can be used as optimization algorithms. For example, the processing circuit 51 generates predicted phase distribution data by applying a forward propagation process to the magnetic resonance morphology image data according to the network configuration of the untrained model. Next, the processing circuit 51 calculates a loss value, which is the error between the predicted phase distribution data and the ground truth phase distribution data, based on a loss function. The processing circuit 51 updates the parameters of the untrained model to minimize the loss value. In this way, the untrained model learns the correlation between the magnetic resonance morphology image data and the phase distribution data.

[0044] The processing circuit 51 repeatedly generates predicted phase distribution data, calculates loss values, and updates parameters until the termination conditions are met. The termination conditions can be set to things like the number of iterations reaching a predetermined number, the loss value converging to less than a predetermined value, or the accuracy of the predicted phase distribution data reaching a predetermined value. The set of parameters when the termination conditions are met is stored in memory 53 as the optimal parameters. The machine learning model to which the optimal parameters are assigned is used as the phase distribution estimation model.

[0045] In the operational phase, the processing circuit 51 applies magnetic resonance morphology image data of the subject to a phase distribution estimation model to generate phase distribution data for the subject. By using the phase distribution estimation model, it becomes possible to generate phase distribution data corresponding to the magnetic resonance morphology image data.

[0046] For example, if a shimming map is used as the ground truth phase distribution data, the phase distribution data output from the phase distribution estimation model in the operational scenario is a magnetic field uniformity map. Generally, shimming maps are generated with a lower spatial resolution than magnetic resonance morphology image data. In cases where the spatial resolution of magnetic resonance morphology image data and the shimming map differs in this way, the spatial resolution of both image data can be made the same. Specifically, the processing circuit 51 may set a shimming map upsampled to the spatial resolution of the magnetic resonance morphology image data as the ground truth shimming map for the training samples. As another example, a network layer may be added to the phase distribution estimation model that upsamples the spatial resolution of the shimming map to the spatial resolution of the magnetic resonance morphology image data.

[0047] Various input and output patterns are possible for the phase distribution estimation model. The input data, magnetic resonance morphology image data, consists of one or more images from among complex images, intensity images, and phase images. The output data, phase distribution data, consists of a magnetic field uniformity map, susceptibility-enhanced image, or quantitative susceptibility mapping image.

[0048] Figure 3 shows an example of the input and output of a phase distribution estimation model. The phase distribution estimation model shown in Figure 3 takes FLAIR images, T1-weighted images, and T2-weighted images of the same imaging area of ​​the same subject as input and outputs a magnetic field uniformity map for that imaging area of ​​the same subject. FLAIR images, T1-weighted images, and T2-weighted images are examples of complex images. It is not limited to using all of the FLAIR images, T1-weighted images, and T2-weighted images as input; two or one of them may be used as input. When the magnetic field uniformity map is the output data, a shimming map is used as the phase distribution data for the training sample.

[0049] Figure 4 shows another example of the input and output of the phase distribution estimation model. The phase distribution estimation model shown in Figure 4 takes intensity images and phase images of the same imaging area of ​​the same subject as input and outputs a magnetic field uniformity map for the same imaging area of ​​the subject. The intensity image can be generated as the real part of any complex image, and the phase image can be generated as the imaginary part of the complex image. It is not limited to using both the intensity image and the phase image as input; either the intensity image or the phase image may be used as input. When the magnetic field uniformity map is the output data, a shimming map is used as the phase distribution data for the training samples.

[0050] Figure 5 shows another example of the input and output of the phase distribution estimation model. The phase distribution estimation model shown in Figure 5 takes intensity images and phase images of the same imaging area of ​​the same subject as input and outputs a susceptibility-weighted image of the same imaging area of ​​the same subject. The intensity image can be generated as the real part of any complex image, and the phase image can be generated as the imaginary part of the complex image. It is not limited to using both the intensity image and the phase image as input; only the intensity image may be used as input. When the susceptibility-weighted image is the output data, the susceptibility-weighted image is used as the phase distribution data for the training samples.

[0051] Furthermore, when intensity and phase images of the same subject and the same imaging area are used as input data, a quantitative susceptibility mapping image of the subject and the same imaging area may be used as output data. In this case, the quantitative susceptibility mapping image is used as the phase distribution data for the training sample.

[0052] The input data for the phase distribution estimation model is not limited to magnetic resonance morphological image data. In addition to magnetic resonance morphological image data, other medical image data collected by other modalities (other modality image data), non-image data recorded in electronic medical records (electronic medical record non-image data), and / or contrast agent information may also be used as input data. This phase distribution estimation model is a machine learning model trained on training samples that include input data (morphological image data, medical image data collected by other modalities, non-image data recorded in electronic medical records, and / or contrast agent information) and output data (phase distribution data).

[0053] Other modality image data refers to image data collected by X-ray computed tomography (CT) scanners, X-ray diagnostic equipment, ultrasound diagnostic equipment, and nuclear medicine diagnostic equipment. Ideally, magnetic resonance imaging data and other modality image data should be image data of the same subject and the same imaging site. Electronic medical record non-image data refers to string data related to disease names, blood information, genes, etc., recorded in the electronic medical record. Contrast agent information refers to information regarding whether a contrast agent was administered to the subject during the acquisition of magnetic resonance imaging data or other modality image data, and if so, the type of contrast agent and injection rate.

[0054] Figure 6 shows another example of input and output for the phase distribution estimation model. The phase distribution estimation model shown in Figure 6 takes magnetic resonance morphological image data, medical image data collected by other modalities (other modality image data), non-image data recorded in the electronic medical record (electronic medical record non-image data), and contrast agent information for the same subject as input and outputs phase distribution data for that subject. The magnetic resonance morphological image data, other modality image data, and phase distribution data are image data for the same imaging area of ​​the subject. Note that the input to the phase distribution estimation model does not need to include all of the magnetic resonance morphological image data, other modality image data, electronic medical record non-image data, and contrast agent information; some of the other modality image data, electronic medical record non-image data, and contrast agent information may be input.

[0055] Next, we will explain an example of the operation of an MR examination using the magnetic resonance imaging device 1.

[0056] Figure 7 shows a typical flow of an MR examination using the magnetic resonance imaging apparatus 1 according to this embodiment. It is assumed that the phase distribution estimation model has already been generated and stored in memory 53, etc., before the start of the MR examination. The following embodiment is applicable to any phase distribution estimation model described above, but for the purpose of providing a concrete explanation, the phase distribution estimation model shown in Figure 2 will be used as an example.

[0057] As shown in Figure 7, first, the processing circuit 51 controls the sequence control circuit 29 via the imaging control function 512 to perform shimming imaging on the subject P (step S1). k-space data is collected by the receiving circuit 25 through shimming imaging. The collected k-space data is acquired by the processing circuit 51 via the acquisition function 513. After step S1 is performed, the processing circuit 51 generates a shimming map based on the k-space data collected in step S1 via the reconstruction function 514 (step S2). As an example, in step S1, the sequence control circuit 29 executes a low-resolution gradient echo pulse sequence twice, and the receiving circuit 25 collects two sets of k-space data. In step S2, the processing circuit 51 generates a shimming map, which is the spatial distribution of the static magnetic field shift, based on the two sets of k-space data. The static magnetic field shift can be determined, for example, as phase difference / time difference [rad / s]. The processing circuit 51 calculates a magnetic field uniformity correction value based on the shimming map, using the imaging condition setting function 511, to make the static magnetic field displacement spatially uniform.

[0058] When step S2 is performed, the processing circuit 51 controls the sequence control circuit 29 with the imaging control function 512 and performs a first MR imaging of the subject P according to the magnetic field uniformity correction value (step S3). k-space data is collected by the receiving circuit 25 as a result of the first MR imaging. The collected k-space data is acquired by the processing circuit 51 with the acquisition function 513. When step S3 is performed, the processing circuit 51 generates magnetic resonance morphological image data based on the k-space data collected in step S3 with the reconstruction function 514 (step S4).

[0059] In this embodiment, the shimming imaging and the first MR imaging are the same, regardless of whether they are contrast-enhanced or non-contrast imaging. Similarly, the first MR imaging and the second MR imaging are the same, regardless of whether they are contrast-enhanced or non-contrast imaging.

[0060] When step S4 is performed, the processing circuit 51 uses the phase distribution calculation function 515 to apply the magnetic resonance morphology image data generated in step S4 to the phase distribution estimation model to estimate the phase distribution data (step S5). In step S5, the processing circuit 51 reads the phase distribution estimation model from the memory 53, inputs the magnetic resonance morphology image data to the read phase distribution estimation model, performs forward propagation processing according to the network configuration of the phase distribution estimation model to generate phase distribution data, and outputs the phase distribution data from the phase distribution estimation model. The spatial range of the phase distribution data can be set arbitrarily. For example, in the case of MR image acquisition, phase distribution data may be generated for the entire imaging range, or it may be generated only for the imaging target area. In the case of MRS imaging, phase distribution data may be generated only within the VOI (Volume Of Interest), or it may be generated for the VOI and its surrounding area. The surrounding area may be an area within a predetermined number of voxels from the VOI, or it may be an area arbitrarily specified by the user. Furthermore, the phase distribution data can be arbitrarily selected from magnetic field uniformity maps, susceptibility-weighted images, and quantitative susceptibility mapping images.

[0061] When step S5 is performed, the processing circuit 51 uses the determination function 516 to determine whether the change to the second MR imaging is appropriate based on the phase distribution data calculated in step S5 (step S6). Specifically, the processing circuit 51 first calculates representative values ​​for the phase distribution data. The method for calculating representative values ​​differs depending on whether the second MR imaging is contrast-enhanced or non-contrast imaging. If the second MR imaging is contrast-enhanced, representative values ​​are calculated based on the data values ​​of the phase distribution data.

[0062] Specifically, if the data value of the phase distribution data is not a quantitative value, the processing circuit 51 sets the relative value, specifically the difference value, between the data value of pixels inside the imaging target area and the data value of pixels outside the area as the representative value. Specifically, "if the data value of the phase distribution data is not a quantitative value" refers to cases where a magnetic field uniformity map or susceptibility-enhanced image is used as the phase distribution data. If the data value of the phase distribution data is a quantitative value, the processing circuit 51 sets the data value of pixels inside the imaging target area as the representative value. Specifically, "if the data value of the phase distribution data is a quantitative value" refers to cases where a quantitative susceptibility mapping image is used as the phase distribution data.

[0063] On the other hand, if the second MR imaging is non-contrast imaging, the processing circuit 51 calculates a representative value based on the difference between the data value of the phase distribution data and the data value of the shimming map collected by shimming imaging performed before or after the first MR imaging. For example, the representative value is calculated based on the difference between the data value of the phase distribution data generated in step S5 and the data value of the shimming map generated in step S2.

[0064] Note that "phase distribution data values" refer to the statistical values ​​of all or some pixels inside or outside the imaging target area. The statistical values ​​may be the average, median, minimum, maximum, or other statistical values ​​of multiple pixels. As another example, the pixel value of any one pixel inside or outside the imaging target area may be used as a representative value.

[0065] Once a representative value of the phase distribution data is calculated, the processing circuit 51 compares this representative value with a reference value. The reference value can be set to any value. If the representative value of the phase distribution data is lower than the reference value, the processing circuit 51 determines that no change is needed for the second MR imaging. If the representative value is higher than the reference value, the processing circuit 51 determines that a change is needed for the second MR imaging.

[0066] If it is determined in step S6 to change the second MR imaging (step S7: YES), the processing circuit 51 selects the type and / or recommended value of the recommended change conditions using the selection function 517 (step S8). In step S8, the processing circuit 51 selects the type and / or recommended value of the recommended change conditions. The processing circuit 51 selects the imaging position, magnetic field uniformity correction value, the timing of imaging in relation to contrast agent administration, and imaging parameters as the recommended change conditions. The imaging parameters may be selected from data acquisition trajectory, pulse sequence, echo time, voxel size, voxel position, and / or water suppression parameters.

[0067] The types of recommended changes should differ depending on whether the second MR imaging is contrast-enhanced or non-contrast-enhanced. If the second MR imaging is contrast-enhanced, the types of recommended changes should be selected from the sequence of imaging relative to contrast agent administration, data acquisition trajectory, pulse sequence, echo time, voxel size, voxel position, and water suppression parameters. If the second MR imaging is MR image imaging, the types of imaging parameters should be selected from the data acquisition trajectory, pulse sequence, and echo time. If the second MR imaging is MRS imaging, the types of imaging parameters should be selected from the pulse sequence, voxel size, voxel position, echo time, and water suppression parameters. If the second MR imaging is CEST imaging, the types of imaging parameters should be selected from the pulse sequence, voxel size, voxel position, echo time, and water suppression parameters. If the second MR imaging is non-contrast-enhanced, the types of recommended changes should be selected from the imaging position and magnetic field uniformity correction value.

[0068] The recommended values ​​for the recommended change conditions may be manually determined by the user via the input interface 57, or they may be automatically determined according to conditions empirically determined in terms of T2 shortening effect, echo time, and / or magnetic field uniformity. If it is not necessary to provide recommended values, then recommended values ​​do not need to be determined.

[0069] As an example, the processing circuit 51 determines whether the values ​​of each imaging parameter of the second MR imaging satisfy the judgment conditions related to the T2 shortening effect, echo time, and / or magnetic field uniformity (hereinafter referred to as the change judgment conditions), and selects imaging parameters with values ​​that do not satisfy the change judgment conditions as recommended change conditions. The change judgment conditions are registered in a LUT (Look Up Table) that associates the type of imaging parameter with the recommended values ​​that the imaging parameter should have from the standpoint of the T2 shortening effect, echo time, and / or magnetic field uniformity. The recommended values ​​may be registered as one or more discrete values, or as a range from an upper limit to a lower limit.

[0070] The processing circuit 51 compares the initial value of each imaging parameter for the second MR imaging with the recommended value registered in the LUT. If the initial value matches the recommended value, the processing circuit 51 does not select the imaging parameter as a recommended change condition. On the other hand, if the initial value does not match the recommended value, the processing circuit 51 selects the imaging parameter as a recommended change condition. In this case, the processing circuit 51 outputs the type of the recommended change condition and the recommended value.

[0071] Below, we will explain specific examples of the change determination conditions, divided into cases where the second MR imaging is MR image acquisition, MRS imaging, and CEST imaging.

[0072] [MR Image Acquisition] When the type of imaging parameter is data acquisition trajectory, non-Cartesian acquisition methods such as EPI acquisition and spiral acquisition are sensitive to magnetic fields, so the recommended value is set to Cartesian acquisition, which is not sensitive to magnetic fields. When the type of imaging parameter is pulse sequence, the FE system is sensitive to magnetic fields, so the recommended value is set to the SE system, which is not sensitive to magnetic fields. When the type of imaging parameter is echo time, a shorter echo time has less of an effect on T2 shortening, so the recommended value is set to a relatively short value that has been determined to have a small effect on T2 shortening.

[0073] [MRS Imaging] When the imaging parameter type is pulse sequence, the MEGA system is sensitive to magnetic fields, so the recommended value is set to the PRESS system without MEGA pulses. When the imaging parameter type is voxel size, a smaller value has less effect from magnetic field uniformity, so the recommended value is set to a relatively small value that is defined as having low magnetic field heterogeneity. When the imaging parameter type is voxel position, magnetic field uniformity changes depending on the position, so the recommended value is set to a position that is defined as having low magnetic field heterogeneity. When the imaging parameter type is echo time, a shorter time has less effect from the T2 shortening effect, so the recommended value is set to a relatively short value that is defined as having low effect from the T2 shortening effect. When the imaging parameter type is water suppression parameter, if magnetic field uniformity deteriorates, it is better to expand the range of the applied frequency of the water suppression pulse, so the recommended value is set to a range that can ensure relatively good accuracy even if magnetic field uniformity deteriorates.

[0074] [CEST Imaging] When the imaging parameter type is pulse sequence, the FE system is sensitive to magnetic fields, so the recommended value is set to the SE system, which is not sensitive to magnetic fields. When the imaging parameter type is voxel size, a smaller value has less effect from magnetic field uniformity, so the recommended value is set to a relatively small value that is defined as having low magnetic field heterogeneity. When the imaging parameter type is voxel position, magnetic field uniformity changes depending on the position, so the recommended value is set to a position that is defined as having low magnetic field heterogeneity. When the imaging parameter type is echo time, a shorter time has less effect from the T2 shortening effect, so the recommended value is set to a relatively short value that is defined as having low effect from the T2 shortening effect. When the imaging parameter type is water suppression parameter, if magnetic field uniformity deteriorates, it is better to expand the range of the applied frequency of the water suppression pulse, so the recommended value is set to a range that can ensure relatively good accuracy even if magnetic field uniformity deteriorates.

[0075] When step S8 is performed, the processing circuit 51 uses the display control function 518 to display the type and / or recommended value of the recommended change condition selected in step S8 on the display 55 (step S9). The layout of the type and / or recommended value of the recommended change condition can be arbitrarily set. The user checks the displayed type and / or recommended value of the recommended change condition and decides whether to accept or reject the change to the recommended change condition. The type and / or recommended value of the recommended change condition is an example of support information related to the second MR imaging.

[0076] Figure 8 shows an example of the display screen I1 for recommended changes. As shown in Figure 8, the display screen I1 includes a display area I11, an approve button I12, and a reject button I13. Display area I11 is a display area that shows whether or not the second MR imaging needs to be changed, the type of recommended change condition, and / or the recommended value. For example, display area I11 displays the result of the determination of whether or not the second MR imaging needs to be changed, such as "The magnetic field uniformity did not meet the criteria." This allows the user to know that the magnetic field uniformity of the second MR imaging is not good under the current imaging conditions. Also, display area I11 displays a message indicating that the type of recommended change condition is a pulse sequence and the recommended value for the recommended change condition is PRESS, such as "Sequence: MEGA-PREE → PRESS". This allows the user to understand the type of recommended change condition and / or the recommended value without having to find it themselves. In addition, display area I11 may also display the echo time at the recommended value, such as "TE: 68 → 25ms". This allows users to understand how changing the set value to a recommended value can alter the echo time, which is one of the key indicators of magnetic field uniformity.

[0077] The approval button I12 is a GUI button that outputs a signal indicating approval of the change in imaging conditions. When the approval button I12 is pressed, the processing circuit 51 determines that the change has been approved. The rejection button I13 is a GUI button that outputs a signal indicating rejection of the change in imaging conditions. When the rejection button I13 is pressed, the processing circuit 51 determines that the change has not been approved.

[0078] If the change is approved (step S10: YES), the processing circuit 51 changes the imaging conditions for the second MR imaging using the imaging condition setting function 511 (step S11). In step S11, if a recommended value was selected in step S8, the processing circuit 51 changes the values ​​of the imaging conditions to the recommended value. If the changed recommended conditions and magnetic field uniformity correction value are the imaging position, the processing circuit 51 changes the slice position to an arbitrary position according to instructions from the user via the input interface 57. Then, the processing circuit 51 performs shimming imaging for the changed slice position using the imaging control function 512 and recalculates the magnetic field uniformity correction value based on the collected k-space data. The changes to the slice position, shimming imaging, and recalculation of the magnetic field uniformity correction value are repeated until the user determines that the quality of the imaging position and magnetic field uniformity correction value is appropriate.

[0079] If step S11 is performed, and the change is not approved (step S10: NO), or if it is determined that the second MR imaging will not be changed (step S7: NO), the processing circuit 51 performs the second MR imaging using the imaging control function 512 (step S12). If the change is not approved (step S10: NO) and it is determined that the second MR imaging will not be changed (step S7: NO), the second MR imaging is performed using the imaging conditions before the change. If step S11 is performed, the second MR imaging is performed using the recommended conditions after the change. The second MR imaging may be MR image imaging, MRS imaging, or CEST imaging.

[0080] This concludes the MRI examination according to this embodiment.

[0081] As described above, the magnetic resonance imaging apparatus 1 according to this embodiment has a processing circuit 51. The processing circuit 51 performs a first MR imaging on the subject P and collects k-space data. Based on the collected k-space data, the processing circuit 51 generates magnetic resonance morphological image data representing the morphology of the subject P. Based on at least the magnetic resonance morphological image data, the processing circuit 51 calculates phase distribution data representing the spatial distribution including phase information. Based on the phase distribution data, the processing circuit 51 determines whether or not it is necessary to modify the second MR imaging that follows the first MR imaging.

[0082] According to the above configuration, the magnetic resonance imaging apparatus 1 calculates phase distribution data based on the magnetic resonance morphological image data acquired in the first MR imaging (first MR imaging), and determines whether or not to modify the second MR imaging (second MR imaging) based on the phase distribution data. In this way, the magnetic resonance imaging apparatus 1 can determine whether or not to modify the third MR imaging by considering the phase distribution data, which affects the quality of the data acquired by MR imaging, and consequently, it becomes possible to improve the quality of the data acquired by MR imaging.

[0083] Next, several examples of MRI examinations according to this embodiment will be described.

[0084] <Example 1> In Example 1, both the first and second MR imaging were non-contrast imaging. The imaging area is assumed to be a region where the magnetic resonance morphological image data contains a relatively large number of boundaries between the air and the imaging area. For example, the breast and abdomen contain more boundaries compared to the head, where the brain parenchyma is the imaging target. When there are many boundaries in this way, static magnetic field shifts are likely to occur, so the optimal value for magnetic field uniformity correction with good static magnetic field uniformity is determined by repeatedly performing shimming imaging by changing the imaging position.

[0085] Figure 9 shows a typical flow of an MR examination using the magnetic resonance imaging apparatus 1 according to Example 1. As shown in Figure 9, first, the processing circuit 51 controls the sequence control circuit 29 with the imaging control function 512 to perform shimming imaging on the subject P (step SA1). Shimming imaging is performed without contrast. After step SA1 is performed, the processing circuit 51 generates a shimming map based on the k-space data collected in step SA1 with the reconstruction function 514, and calculates a magnetic field uniformity correction value based on the generated shimming map (step SA2).

[0086] When step SA2 is performed, the processing circuit 51 determines whether the number of shimming images performed in the MR examination is more than one (step SA3). In the first step SA3 of the MR examination, only one shimming image was performed, so it is determined that the number of shimming images is not more than one (step SA3: NO). In this case, the processing circuit 51 controls the sequence control circuit 29 with the imaging control function 512 and performs a first non-contrast image on the subject P (step SA4). A non-contrast image is defined as MR image acquisition on the subject P who has not been administered a contrast agent. The first non-contrast image is an example of the first MR image. k-space data is collected by the receiving circuit 25 during the first non-contrast image. The collected k-space data is acquired by the processing circuit 51 with the acquisition function 513. When step SA4 is performed, the processing circuit 51 generates magnetic resonance morphological image data based on the k-space data collected in step SA4 with the reconstruction function 514 (step SA5).

[0087] When step SA5 is performed, the processing circuit 51 uses the phase distribution calculation function 515 to apply the magnetic resonance morphology image data generated in step SA5 to the phase distribution estimation model to estimate the phase distribution data (step SA6). If step SA6 is performed or if it is determined in step SA3 that the number of shimming imagings is more than one (step SA3: YES), the processing circuit 51 uses the determination function 516 to determine whether the representative value of the phase distribution data calculated in step SA6 is smaller than the reference value (step SA7). In Example 1, the difference between the shimming map and the phase distribution data is used as the representative value of the phase distribution data. As described above, if the representative value of the phase distribution data is smaller than the reference value, it means that the uniformity of the magnetic field is good and therefore the imaging conditions for the second non-contrast imaging will not be changed. If the representative value of the phase distribution data is not smaller than the reference value, it means that the uniformity of the static magnetic field is not good and therefore the imaging conditions for the second non-contrast imaging will be changed.

[0088] If, in step SA7, it is determined that the representative value of the phase distribution data is not smaller than the reference value (step SA7: NO), the processing circuit 51 uses the selection function 517 to select the imaging position and / or magnetic field uniformity correction value as recommended change conditions (step SA8). After step SA8 is performed, the processing circuit 51 uses the display control function 518 to display the type and / or recommended value of the recommended change conditions selected in step SA8 on the display 55 (step SA9). The user checks the displayed type and / or recommended value of the recommended change conditions and decides whether to accept or reject the change to the recommended change conditions.

[0089] If the changes to the recommended conditions are approved (step SA10: YES), the processing circuit 51 changes the imaging position for the second non-contrast imaging and shimming imaging using the imaging condition setting function 511 (step SA11). Then the processing circuit 51 returns to step SA1 and performs a second shimming imaging at the changed imaging position (step SA1), updating the shimming map and magnetic field uniformity correction values ​​(step SA2). The second shimming map and magnetic field uniformity correction values ​​are stored in memory 53. The first shimming map and magnetic field uniformity correction values ​​may be discarded from memory 53.

[0090] In the second step SA3, the number of shimming images is two, which is determined to be more than one (step SA3: NO). The processing circuit 51 then uses the determination function 516 to determine whether the representative value of the phase distribution data is smaller than the reference value (step SA7). In the second step SA7, the difference between the phase distribution data based on the first non-contrast image and the shimming map based on the second shimming image is used as the representative value. Steps SA1 to SA11 are repeated in this manner, changing the imaging position, until the representative value of the shimming map is determined to be smaller than the reference value in step SA7. It is also possible to exit the repetition of steps SA1 to SA11 if approval for changing the recommended conditions is rejected in step SA10.

[0091] Then, if in step SA7 it is determined that the representative value of the phase distribution data is smaller than the reference value (step SA7: YES) or if in step SA10 approval for changing the recommended conditions is rejected (step SA10: NO), the processing circuit 51 controls the sequence control circuit 29 based on the latest magnetic field uniformity correction value using the imaging control function 512, and performs a second non-contrast imaging (step SA12). The second non-contrast imaging is an example of a second MR imaging.

[0092] This concludes the MR examination according to Example 1.

[0093] As described above, according to Example 1, when both the first and second MR imaging are non-contrast imaging, and the target area requires repeated shimming imaging while changing the imaging position, it becomes possible to determine whether or not repeated shimming imaging is necessary based on the phase distribution data. Therefore, since shimming imaging is repeated only when it is necessary, the throughput of the MR examination is improved. In addition, since shimming imaging is repeated appropriately when the magnetic field uniformity is poor, it becomes possible to ensure the quality of the data collected by the second MR imaging.

[0094] <Example 2> In Example 2, the first MR imaging is contrast-enhanced, and the second MR imaging is non-contrast-enhanced. Furthermore, the recommended change is the order of imaging in relation to contrast agent administration.

[0095] Figure 10 shows a typical flow of an MR examination using the magnetic resonance imaging apparatus 1 according to Example 2. As shown in Figure 10, first, the processing circuit 51 controls the sequence control circuit 29 with the imaging control function 512 to perform shimming imaging on the subject P (step SB1), and the reconstruction function 514 generates a shimming map based on the k-space data collected in step SB1, and calculates a magnetic field uniformity correction value based on the generated shimming map (step SB2).

[0096] When step SB2 is performed, the processing circuit 51 controls the sequence control circuit 29 with the imaging control function 512 to perform non-contrast imaging on the subject P (step SB3), and the reconstruction function 514 generates magnetic resonance morphological image data based on the k-space data collected in step SB3 (step SB4). The non-contrast imaging in step SB3 is an example of the first MR imaging.

[0097] When step SB4 is performed, the processing circuit 51 uses the phase distribution calculation function 515 to apply the magnetic resonance morphology image data generated in step SB4 to the phase distribution estimation model to estimate the phase distribution data after contrast agent administration (step SB5). The phase distribution estimation model according to Example 2 is a machine learning model trained to take magnetic resonance morphology image data before contrast agent administration as input and output phase distribution data after contrast agent administration. The phase distribution estimation model according to Example 2 will be described below. Note that the same matters as those described in the phase distribution estimation model according to this embodiment shown in Figure 2 will be omitted.

[0098] Figure 11 shows the generation process of the phase distribution estimation model according to Example 2. As shown in Figure 11, the processing circuit 51 trains the untrained model based on multiple training samples, which include magnetic resonance morphological image data as input data and phase distribution data as output data. Magnetic resonance morphological image data is collected by MR imaging of subjects who have not been administered contrast agent. As phase distribution data, a shimming map collected by shimming imaging of subjects who have been administered contrast agent is used. The phase distribution data is used as ground truth data.

[0099] The processing circuit 51 updates the parameters of the untrained model using an arbitrary optimization algorithm through supervised learning based on multiple training samples. For example, the processing circuit 51 generates predicted phase distribution data by applying a forward propagation process to the magnetic resonance morphology image data according to the network configuration of the untrained model. Next, the processing circuit 51 calculates a loss value, which is the error between the predicted phase distribution data and the ground truth phase distribution data, based on a loss function. The processing circuit 51 updates the parameters of the untrained model so that the loss value is reduced. In this way, the untrained model learns the correlation between magnetic resonance morphology image data before contrast agent administration and phase distribution data after contrast agent administration.

[0100] The processing circuit 51 repeatedly generates predicted phase distribution data, calculates loss values, and updates parameters until the termination condition is met. The set of parameters when the termination condition is met is stored in memory 53 as the optimal parameters. The machine learning model to which the optimal parameters are assigned is used as the phase distribution estimation model according to Example 2.

[0101] In the operational phase, the processing circuit 51 applies the magnetic resonance morphology image data of the subject before contrast agent administration to the phase distribution estimation model according to Example 2 to estimate the phase distribution data of the subject after contrast agent administration. By using the phase distribution estimation model, it becomes possible to estimate the phase distribution data after contrast agent administration from magnetic resonance morphology image data before contrast agent administration, without administering a contrast agent to the subject.

[0102] When step SB5 is performed, the processing circuit 51 uses the determination function 516 to determine whether the representative value of the phase distribution data estimated in step SB5 is smaller than the reference value (step SB6). In the case of Example 2, the statistical value of the phase distribution data is used as the representative value of the phase distribution data.

[0103] If, in step SB6, it is determined that the representative value of the phase distribution data is not smaller than the reference value (step SB6: NO), the processing circuit 51 uses the selection function 517 to select the imaging order of contrast-enhanced imaging as the recommended change condition (step SB7). Contrast-enhanced imaging is an example of a second MR imaging. After step SB7 is performed, the processing circuit 51 uses the display control function 518 to display the type of recommended change condition selected in step SB7 and / or the recommended value on the display 55 (step SB8).

[0104] Here, the processing circuit 51 determines and displays the recommended imaging order for unperformed MR scans in the MR examination. If the representative value of the phase distribution data is not small compared to the reference value, it is expected that the static magnetic field shift due to the administration of the contrast agent will be large, so it is desirable to perform contrast-enhanced imaging after non-contrast imaging. Therefore, the processing circuit 51 rearranges the imaging order so that unperformed contrast-enhanced imaging is performed after unperformed non-contrast imaging. The processing circuit 51 displays the rearranged imaging order as the recommended value on the display 55. The user checks the recommended value and decides whether to approve or reject the change in imaging order.

[0105] If the change to the recommended conditions is approved (Step B9: YES), the processing circuit 51 changes the imaging order of contrast-enhanced imaging to after other non-contrast imaging using the imaging condition setting function 511 (Step SB10). Note that the user may manually change the imaging order via the input interface 57. Once Step SB10 is performed, the processing circuit 51 performs MR imaging according to the changed imaging order using the imaging control function 512 (Step SB11).

[0106] On the other hand, if it is determined that the representative value of the phase distribution data is smaller than the reference value (Step SB6: YES), the processing circuit 51 performs MR imaging according to the initial imaging order using the imaging control function 512 (Step SB12). If it is determined that the representative value of the phase distribution data is smaller than the reference value, it is considered that the shift in the static magnetic field due to contrast agent administration is small, so it is not necessary to change the order of contrast-enhanced imaging and non-contrast imaging. For this reason, MR imaging can be performed according to the initial imaging order. Furthermore, even if it is determined that the representative value of the phase distribution data is not smaller than the reference value (Step SB6: NO), approval for changing the recommended conditions for change is rejected (Step B9: NO), MR imaging may still be performed according to the initial imaging order.

[0107] This concludes the MR examination according to Example 2.

[0108] As described above, according to Example 2, when the first MR imaging is non-contrast imaging and the second MR imaging is contrast imaging, it becomes possible to calculate contrast agent-administered phase distribution data based on the magnetic resonance morphological image data collected by the first MR imaging, and to determine whether or not to change the imaging order of the second MR imaging based on said phase distribution data. In other words, it becomes possible to determine whether to perform the second MR imaging according to the original order or to perform it after other non-contrast imaging before actually administering the contrast agent to subject P. This makes it possible to perform contrast imaging and other non-contrast imaging at appropriate timings. It also becomes possible to reduce the occurrence of re-administration of contrast agent due to errors in the timing of contrast agent administration, etc.

[0109] <Example 3> In Example 3, the first MR imaging is contrast-enhanced imaging, the second MR imaging is non-contrast-enhanced imaging, and the recommended imaging for modification is imaging parameters.

[0110] Figure 12 shows a typical flow of an MR examination using the magnetic resonance imaging apparatus 1 according to Example 3. Steps SC1 to SC6 are the same as steps SB1 to SB6 according to Example 2.

[0111] If, in step SC6, it is determined that the representative value of the phase distribution data is not smaller than the reference value (step SC6: NO), the processing circuit 51 uses the selection function 517 to select imaging parameters as recommended change conditions for contrast-enhanced imaging (step SC7). Contrast-enhanced imaging is an example of a second MR imaging procedure.

[0112] The types of imaging parameters that can be selected as recommended change conditions are data acquisition trajectory, pulse sequence, echo time, voxel size, voxel position, and water suppression parameters. If contrast-enhanced imaging is MR imaging, the types of imaging parameters should be selected from data acquisition trajectory, pulse sequence, and echo time. If contrast-enhanced imaging is MRS imaging, the types of imaging parameters should be selected from pulse sequence, voxel size, voxel position, echo time, and water suppression parameters. If contrast-enhanced imaging is CEST imaging, the types of imaging parameters should be selected from pulse sequence, voxel size, voxel position, echo time, and water suppression parameters. The method for selecting the types of recommended change conditions and / or recommended values ​​is as described above.

[0113] When step SC7 is performed, the processing circuit 51, using the display control function 518, displays the type of imaging parameter selected in step SC7 and / or the recommended value on the display 55 (step SC8). The user checks the displayed type of recommended change condition and / or the recommended value and decides whether to accept or reject the change to the recommended change condition.

[0114] If the change in imaging parameters is approved (Step C9: YES), the processing circuit 51 changes the imaging parameters for contrast-enhanced imaging using the imaging condition setting function 511 (Step SC10). After Step SC10 is performed, the processing circuit 51 performs contrast-enhanced imaging according to the changed imaging parameters using the imaging control function 512 (Step SC11).

[0115] On the other hand, if it is determined that the representative value of the phase distribution data is smaller than the reference value (step SC6: YES), the processing circuit 51 performs MR imaging according to the initial imaging parameters using the imaging control function 512 (step SC12). In addition, even if it is determined that the representative value of the phase distribution data is not smaller than the reference value (step SC6: NO), but approval for changing the imaging parameters is refused (step SC9: NO), MR imaging may still be performed according to the initial imaging parameters.

[0116] This concludes the MR examination according to Example 3.

[0117] As described above, according to Example 3, when the first MR imaging is non-contrast imaging and the second MR imaging is contrast imaging, it becomes possible to calculate contrast-enhanced phase distribution data based on magnetic resonance morphological image data collected by the first MR imaging, and to determine whether or not it is necessary to change the imaging parameters of the second MR imaging based on said phase distribution data. In other words, it is possible to determine whether or not it is necessary to change the imaging parameters of the second MR imaging before actually administering the contrast agent to subject P. This makes it possible to perform contrast imaging under appropriate imaging parameters.

[0118] According to at least one embodiment described above, the quality of imaging data acquired by MR imaging can be improved.

[0119] In the above description, the term "processor" refers to circuits such as CPUs, GPUs, or Application Specific Integrated Circuits (ASICs), programmable logic devices (e.g., Simple Programmable Logic Devices (SPLDs), Complex Programmable Logic Devices (CPLDs), and Field Programmable Gate Arrays (FPGAs)). A processor functions by reading and executing a program stored in a memory circuit. Alternatively, instead of storing the program in a memory circuit, the program may be directly incorporated into the processor's circuitry. In this case, the processor functions by reading and executing the program incorporated into the circuitry. 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's circuitry as a logic circuit. In this embodiment, each processor is not limited to being configured as a single circuit; multiple independent circuits may be combined to form a single processor and realize its functions. Furthermore, the multiple components shown in Figure 1 may be integrated into a single processor to realize its functions.

[0120] While several embodiments have been described, these embodiments are presented as examples only and are not intended to limit the scope of the invention. These embodiments can be implemented in a variety of other forms, and various omissions, substitutions, modifications, and combinations of embodiments are possible without departing from the spirit of the invention. These embodiments and their variations are included in the scope and spirit of the invention, as well as in the claims and their equivalents. [Explanation of symbols]

[0121] 1. Magnetic Resonance Imaging System 11. Stand 13 berths 21 Gradient magnetic field power supply 23 Transmitter Circuit 25 Receiving Circuit 26 Shim coil power supply 27 Bed drive mechanism 29 Sequence control circuit 41 Static magnetic field magnet 43. Gradient field coil 45 Transmitter coil 47 Receiving coil 49 Shim Coil 50 Host Computers 51 Processing Circuit 53 memory 55 displays 57 Input Interfaces 59 Communication Interface 131 Top plate 133 Base 511 Imaging Condition Setting Function 512 Imaging control function 513 Acquisition function 514 Reconfiguration function 515 Phase distribution calculation function 516 Judgment Function 517 Selection function 518 Display control function

Claims

1. A data acquisition unit that performs a first MR imaging on the subject to collect k-space data, A generation unit that generates morphological image data representing the subject's morphology based on the k-space data, A calculation unit that calculates phase distribution data representing a spatial distribution including phase information based on at least the morphological image data, A determination unit that determines whether or not to modify the second MR imaging that follows the first MR imaging based on the phase distribution data, A magnetic resonance imaging apparatus equipped with the following:

2. The calculation unit applies the morphological image data to the trained model to generate the phase distribution data, The aforementioned trained model is a machine learning model trained on training samples that include morphological image data as input data and phase distribution data as output data. The magnetic resonance imaging apparatus according to claim 1.

3. The aforementioned morphological image data is one or more images selected from a complex image, an intensity image, and a phase image. The phase distribution data is a magnetic field uniformity map, a susceptibility-enhanced image, or a quantitative susceptibility mapping image. The magnetic resonance imaging apparatus according to claim 2.

4. The calculation unit inputs the morphological image data, medical image data collected by other modalities, non-image data recorded in the electronic medical record and / or contrast agent information into a trained model and outputs the phase distribution data. The aforementioned trained model is a machine learning model trained on training samples that include input data consisting of morphological image data, medical image data collected by other modalities, non-image data and / or contrast agent information recorded in electronic medical records, and output data consisting of phase distribution data. The magnetic resonance imaging apparatus according to claim 1.

5. The magnetic resonance imaging apparatus according to claim 1, wherein the determination unit determines that if the representative value of the phase distribution data is below a reference value, it is unnecessary to change the second MR imaging, and determines that if the representative value is above the reference value, it is necessary to change the second MR imaging.

6. If it is determined that a change to the second MR imaging is necessary, a selection unit selects the type and / or recommended value of an imaging condition that is recommended to be changed, which is an imaging condition that affects the T2 shortening effect, echo time, and / or magnetic field uniformity among the imaging conditions for the second MR imaging. The system further comprises a display control unit that displays the type of the recommended change conditions and / or the recommended value on a display device, The magnetic resonance imaging apparatus according to claim 5.

7. If it is determined that a change to the second MR imaging is necessary, the system further includes a modification unit for changing the value of the recommended change condition. The acquisition unit performs the second MR imaging on the subject according to the modified values ​​of the recommended change conditions. The magnetic resonance imaging apparatus according to claim 6.

8. The magnetic resonance imaging apparatus according to claim 6, wherein the selection unit selects, as the recommended change conditions, the imaging position, the magnetic field uniformity correction value, the timing of imaging relative to contrast agent administration, the data acquisition trajectory, the pulse sequence, the echo time, the voxel size, the voxel position, and / or the water suppression parameter.

9. The aforementioned recommended modification conditions are imaging parameters including data acquisition trajectory, pulse sequence, echo time, voxel size, voxel position and / or water suppression parameters. The magnetic resonance imaging apparatus according to claim 6, wherein the selection unit determines whether each of the values ​​of the imaging parameters for the second MR imaging satisfies the conditions relating to the T2 shortening effect, echo time, and / or magnetic field uniformity, and selects the imaging parameters having values ​​that do not satisfy the conditions as the recommended change conditions.

10. The first MR imaging and the second MR imaging are imaging without the use of contrast agent. The selection unit selects the imaging position and / or the magnetic field uniformity correction value as the recommended change conditions. The aforementioned modification unit changes the value of the aforementioned recommended modification condition. The magnetic resonance imaging apparatus according to claim 8.

11. The first MR imaging described above is imaging without the use of contrast agent. The second MR imaging described above is imaging using a contrast agent, The selection unit selects, as the recommended change condition, the order of imaging in relation to the administration of the contrast agent. The modification unit rearranges the order of the second MR imaging and the other MR imaging so that the second MR imaging is performed after other MR imaging that does not use contrast agent. The magnetic resonance imaging apparatus according to claim 8.

12. The first MR imaging described above is imaging without the use of contrast agent. The second MR imaging described above is imaging using a contrast agent, The selection unit selects the data acquisition trajectory, pulse sequence, echo time, voxel size, voxel position and / or water suppression parameter as the recommended change conditions. The aforementioned modification unit changes the value of the aforementioned recommended modification condition. The magnetic resonance imaging apparatus according to claim 8.

13. The magnetic resonance imaging apparatus according to claim 5, wherein the determination unit calculates the representative value based on the data value of the phase distribution data, or the difference between the data value of the phase distribution data and the data value of the shimming map collected by shimming imaging performed before or after the first MR imaging.

14. The magnetic resonance imaging apparatus according to claim 1, wherein the calculation unit estimates the phase distribution data relating to the subject to which the contrast agent has been administered based on the morphological image data relating to the subject to which the contrast agent has not been administered.

15. The magnetic resonance imaging apparatus according to claim 1, wherein the second MR imaging is imaging for image acquisition or imaging for spectral acquisition.

16. The magnetic resonance imaging apparatus according to claim 1, wherein the spatial range of the phase distribution data is the imaging target area in the case of MR image acquisition, and in the case of MRS imaging, it is within the VOI, or within the VOI and the surrounding area of ​​the VOI.

17. An acquisition step of acquiring morphological image data representing the morphology of the subject obtained by a first MR imaging of the subject, A calculation step of calculating phase distribution data representing the spatial distribution of the phase of the static magnetic field based on at least the morphological image data, A determination step, based on the phase distribution data, determines whether or not it is necessary to change the second MR imaging that follows the first MR imaging, A magnetic resonance imaging support method comprising the following.

Images