JP7337603B2 - Magnetic resonance imaging system - Google Patents

Description

Embodiments of the present invention relate to magnetic resonance imaging apparatus.

In magnetic resonance imaging, MR (Magnetic Resonance) signals may be acquired using downsampling. At this time, navigator data and MR signals are acquired, and the navigator data may be used to reacquire the MR signals and perform motion compensation on the MR signals.

Such an MR signal acquisition method does not take into account the amount of MR signal data required for reconstruction of an MR image. Therefore, more MR signals are acquired in order to compensate for the deterioration in image quality due to the lack of the amount of data required for reconstruction. If the movement state of the subject changes when the MR signals are acquired, the amount of data required for reconstruction is not known before reconstruction, so the image quality of the reconstructed MR image cannot be guaranteed.

Japanese Patent Publication No. 2017-533811 Japanese Patent Application Publication No. 2012-505709 Japanese Patent Publication No. 2013-544552 U.S. Pat. No. 6,292,684

The problem to be solved by the present invention is to ensure the quality of MR images by additional acquisition.

A magnetic resonance imaging apparatus according to an embodiment includes a first acquisition unit that acquires MR signals according to a first pulse sequence set in an imaging protocol; a determination unit that determines whether or not additional acquisition of MR signals is necessary based on image quality determination; a second acquisition unit for acquiring MR signals according to the added second pulse sequence; and a reconstruction of an MR image based on the MR signals by the first pulse sequence and the second pulse sequence. and a reconstructing unit for configuring.

FIG. 1 is a diagram showing an example of the configuration of a magnetic resonance imaging apparatus according to this embodiment. FIG. 2 is a flowchart showing an example of the procedure of additional collection execution processing according to this embodiment. FIG. 3 is a flowchart showing an example of the procedure of additional collection execution processing according to this embodiment. FIG. 4 is a diagram showing an example of a first acquisition position and a second acquisition position identified in k-space. FIG. 5 is a diagram showing an example of an angle map used in the additional function. FIG. 6 is a diagram illustrating an example of a first trajectory and a second trajectory in k-space for the expiratory phase and k-space for the inspiratory phase in relation to the present embodiment. FIG. 7 is a diagram showing an example of a first trajectory and a second trajectory in the k-space for the expiratory phase and the k-space for the inspiratory phase in relation to the first modification. FIG. 8 is a flow chart showing an example of the procedure of additional collection execution processing in the first modified example. FIG. 9 is a flowchart showing an example of the procedure of additional collection execution processing in the first modified example. FIG. 10 is a diagram showing an example of temporal changes in the position of the diaphragm, reference periods, and partial periods used in the second modification. FIG. 11 is a diagram showing an example of a projection image generated in the third modified example. FIG. 12 is a diagram showing an example of a region to which the navigation echo method is applied and which is shown in the positioning image used in the third modified example. FIG. 13 is a diagram showing an example of body motion artifacts rendered in a partial image when the scanning method is radial scanning in the fifth modification. FIG. 14 is a diagram showing an example of body movement artifacts rendered in a partial image when the scanning method is Cartesian scanning in the fifth modification. FIG. 15 shows the collection start time, the predetermined time, the end time of the predetermined time, the determination start limit time, the determination additional time, and the additional collection start time for the additional collection execution process in the sixth modification. It is a figure which shows an example with. FIG. 16 is a diagram showing an example of a respiratory waveform and an electrocardiographic waveform in the eighth modified example. FIG. 17 is a diagram showing an example of a configuration of a processing circuit in an application example. FIG. 18 is a diagram schematically showing an example of inputs and outputs of trained models according to another embodiment.

An embodiment of a magnetic resonance imaging (MRI) apparatus will be described below with reference to the drawings. In the following description, components having substantially the same functions and configurations are denoted by the same reference numerals, and redundant description will be given only when necessary.

FIG. 1 is a diagram showing the configuration of an MRI apparatus 100 according to this embodiment. As shown in FIG. 1, the MRI 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 transmission circuit (transmission unit) 113, and a transmission A coil 115, a receiving coil 117, a receiving circuit (receiving unit) 119, an imaging control circuit (collecting unit) 121, an interface (input unit) 125, a display (display unit) 127, and a storage device (storage unit) 129 and a processing circuit (processing unit) 131 . The MRI apparatus 100 may have a hollow cylindrical shim coil between the static magnetic field magnet 101 and the gradient magnetic field coil 103 .

The static magnetic field magnet 101 is a magnet formed in a hollow, substantially cylindrical shape. Note that the static magnetic field magnet 101 is not limited to a substantially 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 .

The gradient magnetic field coil 103 is a hollow cylindrical coil. The gradient magnetic field coil 103 is arranged inside the static magnetic field magnet 101 . The gradient magnetic field coil 103 is formed by combining three coils corresponding to the mutually orthogonal X, Y, and Z axes. It is assumed that 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-axis and the Y-axis. The three coils in the gradient magnetic field coil 103 are individually supplied with current from the gradient magnetic field power source 105 to generate gradient magnetic fields whose magnetic field strengths vary along the X, Y, and Z axes.

Here, the gradient magnetic fields of the X, Y, and Z axes generated by the gradient magnetic field coil 103 are, for example, a slice selection gradient magnetic field, a phase encoding gradient magnetic field, and a frequency encoding gradient magnetic field (also referred to as a readout gradient magnetic field). They correspond to each other. A slice selection gradient magnetic field is used to arbitrarily determine an imaging section. A phase-encoding gradient magnetic field is used to change the phase of the MR signal depending on the spatial position. A frequency-encoding gradient magnetic field is used to vary the frequency of the MR signal according to spatial location.

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 imaging control circuit 121 .

The bed 107 is an apparatus having a top plate 1071 on which the subject P is placed. The bed 107 inserts the top plate 1071 on which the subject P is placed into the bore 111 under the control of the bed control circuit 109 . The bed 107 is installed, for example, in the examination room where the MRI apparatus 100 is installed so that the longitudinal direction is parallel to the central axis of the static magnetic field magnet 101 .

A bed control circuit 109 is a circuit that controls the bed 107 . The bed control circuit 109 drives the bed 107 according to an operator's instruction via the interface 125, thereby moving the top plate 1071 in the longitudinal direction and the vertical direction, and in some cases in the horizontal direction.

The transmission circuit 113 supplies radio frequency (RF) pulses modulated with the Larmor frequency to the transmission coil 115 under the control of the imaging control circuit 121 .

The transmission coil 115 is an RF coil arranged inside the gradient magnetic field coil 103 . The transmission coil 115 receives an RF pulse from the transmission circuit 113 and generates a transmission RF wave corresponding to a high frequency magnetic field. The transmission coil 115 is, for example, a whole body coil (WB coil: Whole Body Coil) having a plurality of coil elements. A WB coil may be used as a transmit/receive coil. Also, the transmission coil 115 may be a WB coil formed by one coil.

Receive coil 117 is an RF coil positioned inside gradient coil 103 in bore 111 . The receiving coil 117 receives MR signals emitted from the subject P by the high frequency magnetic field. Receiving coil 117 outputs the received MR signal to receiving circuit 119 . The receive coil 117 is, for example, a coil array having one or more, typically a plurality of coil elements. It should be noted that although transmit coil 115 and receive coil 117 are depicted as separate RF coils in FIG. 1, transmit coil 115 and receive coil 117 may be implemented as an integrated transmit and receive coil. The transmission/reception coil corresponds to the imaging target of the subject P, and is, for example, a local transmission/reception RF coil such as a head coil.

The receiving circuit 119 generates a digital MR signal (hereinafter referred to as MR data) based on the MR signal output from the receiving coil 117 under the control of the imaging 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/digital conversion (A/D conversion) on the data subjected to the various signal processing. ) and sampling. Thereby, MR data is generated. The receiving circuit 119 outputs the generated MR data to the imaging control circuit 121 .

The imaging control circuit 121 controls the gradient magnetic field power supply 105, the transmission circuit 113, the reception circuit 119, etc. according to the imaging protocol output from the processing circuit 131, and performs MR imaging of the subject P. FIG.

In this embodiment, the imaging control circuit 121 acquires MR signals according to the first pulse sequence set in the imaging protocol. A first pulse sequence is a pulse sequence initially set by the setting function 1313 of the processing circuit 131 . In addition, the imaging control circuit 121 acquires MR signals according to the second pulse sequence set in the imaging protocol. The second pulse sequence is the pulse sequence added by add function 1319 of processing circuitry 131 .

In this embodiment, the pulse sequence is a series of various RF pulses, and includes any parameter that defines the RF pulse as a parameter of the pulse sequence. The pulse sequence includes, for example, 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, and the transmission circuit 113 to supply the transmission coil 115 with the current. The magnitude and time width of the RF pulse to be applied, the timing of supplying the RF pulse to the transmitting coil 115 by the transmitting circuit 113, the timing of receiving the MR signal by the receiving coil 117, and the like are defined. The parameters of the pulse sequence include a trajectory (k-space filling trajectory) indicating the acquisition position of the MR signal on the k-space, flip angle, transmission frequency of various RF pulses, echo order indicating the order of the trajectory, and the like. . Trajectory types include Stack-Of-Stars, Cartesian, Radial, PROPELLAR (Periodically Rotated Overlapping Parallel Lines with Enhanced Reconstruction), and Spiral scanning methods. are applicable.

Specifically, the imaging control circuit 121 sequentially acquires MR signals (referred to as first MR signals) regarding the subject P according to the first trajectory set by the setting function 1313 in the processing circuit 131 . The first trajectory corresponds to an acquisition trajectory indicating acquisition positions of a plurality of first MR signals in k-space (hereinafter referred to as first acquisition positions), that is, a filling trajectory of the first MR signals in k-space. For example, the first trajectory is preset by the setting function 1313 according to an operator's instruction via the interface 125 prior to execution of MR imaging.

The interface 125 has a circuit for receiving various instructions and information input from the operator. The interface 125 has, for example, circuitry relating to a pointing device such as a mouse or an input device such as a keyboard. Note that the circuits included in the interface 125 are not limited to circuits related to physical operation parts such as a mouse and keyboard. For example, the interface 125 receives an electrical signal corresponding to an input operation from an external input device provided separately from the MRI apparatus 100, and outputs the received electrical signal to various circuits. may have Under the control of the processing circuit 131, the interface 125 acquires various data from various biological signal measuring instruments, optical cameras, external storage devices, various modalities, etc., which are directly connected via a network or the like. good too.

The display 127 displays MR images generated by the reconstruction function 1321 under the control of the system control function 1311 in the processing circuit 131, various information related to MR imaging and image processing, and the like. Display 127 is, for example, a CRT (Cathode-Ray Tube) display, a liquid crystal display, an organic EL (Electro Luminescence) display, an LED (Light-Emitting Diode) display, a plasma display, or any other known in the art. is a display device such as a display, monitor, or the like.

The storage device 129 stores MR data filled into the k-space via the reconstruction function 1321, MR image data generated by the reconstruction function 1321, and the like. The storage device 129 stores various imaging protocols, imaging conditions including a plurality of imaging parameters defining the imaging protocols, a trajectory corresponding to a scanning method for the subject P, and the like. The storage device 129 also stores programs related to various reconstruction methods used in the reconstruction function 1321 . Reconstruction methods include, for example, parallel imaging, compressed sensing, and learned neural networks.

The storage device 129 stores programs corresponding to various functions executed by the processing circuit 131 . The storage device 129 is, for example, a RAM (Random Access Memory), a semiconductor memory device such as a flash memory, a hard disk drive, a solid state drive, an optical disc, or the like. Also, the storage device 129 may be a drive device or the like that reads and writes various information with a portable storage medium such as a CD-ROM drive, a DVD drive, and a flash memory.

The processing circuit 131 has a processor (not shown) as hardware resources, memories such as ROM (Read Only Memory) and RAM, and the like, and controls the MRI apparatus 100 in an integrated manner. The processing circuit 131 has a system control function 1311 , a setting function 1313 , an acquisition function 1315 , a determination function 1317 , an addition function 1319 and a reconstruction function 1321 . The functions performed by the system control function 1311, the setting function 1313, the acquisition function 1315, the determination function 1317, the addition function 1319, and the reconstruction function 1321 are stored in the storage device 129 in the form of computer-executable programs. The processing circuit 131 is a processor that reads programs corresponding to these functions from the storage device 129 and executes them, thereby implementing functions corresponding to each program. In other words, the processing circuit 131 in a state where each program is read has a plurality of functions shown in the processing circuit 131 of FIG.

In FIG. 1, the single processing circuit 131 has been described as realizing these various functions. It does not matter if the function is realized by In other words, each function described above may be configured as a program, and one processing circuit may execute each program, or a specific function may be implemented in a dedicated independent program execution circuit. may The system control function 1311, the setting function 1313, the acquisition function 1315, the determination function 1317, the additional function 1319, and the reconfiguration function 1321 of the processing circuit 131 are respectively a system control unit, a setting unit, an acquisition unit, a determination unit, and an addition unit. , which is an example of a reconstruction unit.

The term "processor" used in the above description includes, for example, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), an application specific integrated circuit (ASIC), or a programmable logic device (for example, a simple Programmable Logic Device (SPLD), Complex Programmable Logic Device (CPLD), Field Programmable Gate Array (FPGA), and other circuits.

The processor implements various functions by reading and executing programs stored in the storage device 129 . Note that instead of storing the program in the storage device 129, the program may be configured to be directly installed in the circuit of the processor. In this case, the processor realizes its function by reading and executing the program embedded in the circuit. The bed control circuit 109, the transmission circuit 113, the reception circuit 119, the imaging control circuit 121, and the like are similarly configured by electronic circuits such as the processor.

The processing circuit 131 centrally controls the MRI apparatus 100 through a system control function 1311 . Specifically, the processing circuit 131 reads the system control program stored in the storage device 129, develops it on the memory, and controls each circuit of the MRI apparatus 100 according to the developed system control program. For example, the processing circuit 131 uses the system control function 1311 to read the imaging protocol from the storage device 129 based on the imaging conditions input by the operator via the interface 125 . Note that the processing circuit 131 may generate an imaging protocol based on imaging conditions. The processing circuit 131 transmits an imaging protocol to the imaging control circuit 121 and controls imaging of the subject P. FIG.

The processing circuitry 131 fills the k-space with MR data along the readout direction by the reconstruction function 1321, for example, according to the strength of the readout gradient magnetic field. The processing circuitry 131 generates an MR image by performing a Fourier transform on the MR data filled in k-space. It should be noted that the generation of MR images is not limited to the above-described procedure, and methods of reconstructing MR images by regularization using MR data with missing data such as parallel imaging and compressed sensing, or methods involving missing data It may be executed using a trained deep neural network or the like trained using MR data. The processing circuit 131 outputs the MR image to the display 127 and storage device 129 .

The above is the description of the overall configuration of the MRI apparatus 100 according to the present embodiment. The additional collection execution process in this embodiment will be described below. Additional acquisition corresponds to additional imaging performed on the subject P by the determination function 1317 in the processing circuitry 131 . The additional acquisition execution process is a process of executing additional acquisition when it is determined that additional acquisition is necessary based on the amount of data of the first MR signal and the body motion state of the subject P.

The storage device 129 stores thresholds for discriminating the body motion state of the subject P into a plurality of classes. For example, if the body motion is caused by the respiratory motion of the subject P, the multiple classes correspond to, for example, two classes, an expiratory phase and an inspiratory phase. At this time, the threshold corresponds to, for example, the position of the diaphragm. The plurality of classes is not limited to the above two classes, and may be three or more classes including at least one intermediate state between the expiratory phase and the inspiratory phase. It may be a continuous stepless class adapted to changes. At this time, the storage device 129 stores two or more thresholds corresponding to three or more classes, or continuous thresholds corresponding to stepless classes.

For the sake of specificity, the body movement of the subject P is assumed to be caused by breathing. In addition, it is assumed that there are two body motion states, an expiratory phase and an inspiratory phase. Note that the state of body motion is not limited to the above respiration, which is a substantially periodic body motion, and is caused by the body motion of the subject P, such as the beating of the heart of the subject P or the vocalization of the subject P. , it is applicable to any body movement of the subject P.

The storage device 129 stores a predetermined time from the acquisition start time of the first MR signal to the start time of determining whether or not additional acquisition is necessary. For example, the predetermined period of time corresponds to a certain period of time for acquiring the amount of data necessary for determining whether additional collection is necessary. It should be noted that the predetermined time may be appropriately input or changed by an operator's instruction via the interface 125 . Also, the predetermined time may be zero. The storage device 129 stores a reference value used for determining whether or not additional acquisition is necessary according to the amount of data of the first MR signals belonging to each of the plurality of classes. The reference value is set in advance so as to ensure the quality of the MR image reconstructed by the reconstruction function 1321, for example.

2 and 3 are flowcharts showing an example of the procedure of the additional collection execution process.

In the following description, it is assumed that the setting function 1313 and the adding function 1319 mainly set and add trajectories at the same acquisition position or non-identical acquisition positions as the pulse sequence.

(Step Sa1)
The processing circuit 131 uses the setting function 1313 to set a first trajectory indicating the first acquisition position of the first MR signal in the k-space according to the operator's instruction via the interface 125 . To make the description concrete, the processing circuit 131 will be described below assuming that the filling trajectory of the first MR signal in the k-space by radial scanning is set as the first trajectory. The interface 125 inputs imaging conditions for the subject P according to an operator's instruction.

(Step Sa2)
The imaging control circuit 121 acquires the first MR signal according to the set first trajectory. More specifically, the imaging control circuit 121 performs MR imaging of the subject P according to a first pulse sequence including a first trajectory, and acquires first MR signals regarding the subject P. FIG. For example, the imaging control circuit 121 acquires the first MR signal by performing MR imaging on the subject P under free breathing, that is, asynchronously. The imaging control circuit 121 outputs the acquired first MR signal together with the acquisition time to the storage device 129 . The storage device 129 stores the first MR signal in association with the acquisition time.

(Step Sa3)
The processing circuit 131 acquires the body motion state of the subject P during acquisition of the first MR signal by the acquisition function 1315 . Specifically, the processing circuit 131 acquires the data output from the breathing belt attached to the subject P via the interface 125 as the body motion state. The data output from the breathing belt corresponds to, for example, the respiration level of the subject P, that is, the position of the diaphragm in the subject P. The processing circuit 131 outputs the position of the diaphragm as the state of body motion to the storage device 129 together with the detection time when the position of the diaphragm is detected. The storage device 129 stores the position of the diaphragm together with the detection time. The processing circuit 131 may store the position of the diaphragm together with the detection time in its own memory.

Instead of acquiring the respiration level, the processing circuit 131 may acquire data output from an optical camera that optically photographs the subject P as the body motion state. At this time, the optical camera is, for example, a camera that captures the whole body of the subject P or a part of the subject P (eyes, hands, face, abdomen, etc.). For the sake of specificity, the processing circuit 131 will be described as obtaining the position of the diaphragm from the breathing belt.

(Step Sa4)
The processing circuit 131 uses the determination function 1317 to determine whether or not a predetermined period of time has elapsed from the acquisition start time of the first MR signal. If the predetermined time has not elapsed since the acquisition start time (No in step Sa4), the imaging control circuit 121 repeatedly acquires the first MR signal according to the first trajectory and imaging conditions. When the predetermined time has passed since the collection start time (Yes in step Sa4), the processing circuit 131 executes the process of step Sa5.

(Step Sa5)
The processing circuit 131 classifies the first MR signal into a plurality of classes by the decision function 1317 according to the body motion state value. More specifically, the processing circuit 131 classifies the first MR signals acquired within a predetermined period of time into a plurality of classes (expiratory phase and inspiratory phase) based on the acquired body motion state and threshold. Specifically, the processing circuit 131 associates, among the plurality of first MR signals, a first MR signal corresponding to the same acquisition time as the detection time corresponding to the position of the diaphragm equal to or greater than the threshold value with the exhalation phase. Note that the processing circuit 131 may associate, among the plurality of first MR signals, the first MR signal corresponding to the same acquisition time as the detection time corresponding to the position of the diaphragm below the threshold with the inspiratory phase.

(Step Sa6)
The processing circuit 131 uses the determination function 1317 to identify the maximum class having the maximum data amount of the first MR signal among the plurality of classes. When the body motion is respiratory motion, the time interval of the expiratory phase is basically longer than the time interval of the inspiratory phase. become more. Therefore, processing circuitry 131 identifies the expiratory phase as the largest class.

(Step Sa7)
The processing circuit 131 compares the data amount of the maximum class with the reference value by the determination function 1317 . Note that the amount of data in a predetermined class may be used instead of the reference value. At this time, the data amount of the predetermined class is stored in the storage device 129 instead of the reference value. Specifically, the processing circuit 131 determines that the data amount of the first MR signal classified into a plurality of classes using a threshold for discriminating the body movement state into a plurality of classes is the maximum data amount among the plurality of classes. If the amount of data in the largest class is less than the reference value, it is determined that additional collection is required. More specifically, when the expiratory phase data amount is less than the reference value (Yes in step Sa7), the processing circuit 131 executes the process of step Sa8. If the expiratory phase data amount is greater than or equal to the reference value (No in step Sa7), the processing circuit 131 executes the process of step Sa9. Through the processing of steps Sa4 to Sa7, the processing circuit 131 performs the first MR signal related to the subject P based on the amount of data of the first MR signal corresponding to the state of body motion after a predetermined time has elapsed from the acquisition start time of the first MR signal. Determine whether additional acquisition of 2MR signals is necessary.

(Step Sa8)
The processing circuit 131 adds a second trajectory different from the first MR signal belonging to the largest class to the imaging protocol by means of the addition function 1319 . More specifically, the processing circuit 131 adds a trajectory different from the first trajectory for the first MR signal belonging to the largest class to the imaging protocol as a second trajectory indicating the second acquisition position. The second acquisition position is the acquisition position on the k-space of the second MR signal acquired during the additional acquisition. Specifically, processing circuitry 131 identifies a first acquisition location for a first MR signal belonging to the expiratory phase; A second trajectory indicating the position is added to the imaging protocol. Note that the processing circuit 131 may determine the second collection positions based on the first collection positions in the k-space so that the first collection positions and the second collection positions are evenly distributed in the k-space. Further, when the change in the body motion state acquired by the acquisition function 1315 is substantially periodic, the processing circuit 131 specifies the second collection position in the k-space, taking into consideration the periodicity of the change in the body motion state. may

FIG. 4 shows an example of the region 13 occupied by the first trajectory of the first acquired MR signal identified in k-space 11 . In FIG. 4 the trajectory is drawn as a straight line passing through the origin. As shown in FIG. 4, the area other than area 13 in k-space 11 is set as area 15 . A second trajectory of the second MR signal, which is an acquisition target in the additional imaging, is set in the region 15 . For example, a second trajectory is set in the region 15 where the first MR signal is not acquired during the expiratory phase.

Processing circuitry 131 may use the angle map to identify the second trajectory by additional function 1319 . An angle map is a map in which MR signal trajectories are arranged in a one-dimensional space defined by angles around the origin of the k-space.

FIG. 5 shows an example of an angle map. As shown in FIG. 5, the processing circuit 131 determines the area 11 in the angle map where the first trajectory does not exist as the area 15 in which the second trajectory can be set. For example, processing circuitry 131 places the second trajectory in region 15 at the same spacing as the first trajectory in region 11, as shown in FIG. Furthermore, the second trajectory may be determined such that the interval between the first trajectory and the second trajectory is uniform. Note that when the scanning method for the subject P is Cartesian, the processing circuit 131 may determine the second trajectory using a map in which the first trajectory is arranged in the phase encoding direction instead of the angle map. . Similarly for other scanning schemes, the processing circuit 131 can determine the second trajectory.

FIG. 6 is a diagram showing an example of the first trajectory P1 and the second trajectory P2 in the k-space 11E for the expiratory phase EA and the k-space 11I for the inspiratory phase IA. As shown in FIG. 6, the expiratory phase EA is assumed to be the maximum class. In a later step Sa11, a second MR signal corresponding to a second trajectory P2 in k-space 11E for the expiratory phase EA is acquired. The first MR signal in the inspiratory phase IA that is not the maximum class may be discarded.

(Step Sa9)
The processing circuit 131 causes the display 127 to display at least one of the execution of additional acquisition, the imaging extension time, and the scheduled imaging end time by the system control function 1311 . Specifically, the processing circuitry 131 calculates the imaging extension time based on the total number of additionally acquired second trajectories P2. The processing circuit 131 calculates the scheduled imaging end time by adding the imaging extension time to the imaging time set according to the imaging conditions before execution of acquisition of the first MR signal. The display 127 indicates that additional acquisitions will be performed, the calculated image extension time, and the calculated scheduled imaging end time.

(Step Sa10)
The processing circuit 131 uses the determination function 1317 to determine whether or not the imaging time set as the imaging condition has elapsed. If the imaging time has not elapsed (No in step Sa10), the imaging control circuit 121 executes the process of step Sa2. Next, the processing circuit 131 repeats the processing from step Sa3 to step Sa10. At this time, the maximum class data amount in step Sa7 and the second trajectory in step Sa8 are updated. If the imaging time has elapsed (Yes in step Sa10), the imaging control circuit 121 executes the process of step Sa11.

(Step Sa11)
The imaging control circuit 121 performs additional acquisition according to the second trajectory to acquire a second MR signal. More specifically, the imaging control circuit 121 performs MR imaging of the subject P according to a second pulse sequence including a second trajectory, and acquires second MR signals regarding the subject P. FIG. The imaging control circuit 121 causes the acquired second MR signal to be attached to the maximum class and stored in the storage device 129 . At this time, the MR signal belonging to the maximum class has the first MR signal and the second MR signal. The processing circuit 131 may classify the second MR signal into a plurality of classes by the determination function 1317 and perform the determination described in step Sa7 again. Step Sa11 may be executed before the process of step Sa10.

In step Sa11, the interface 125 may input an instruction to stop the additional collection according to an operator's instruction. Triggered by the input of the instruction to stop the additional acquisition, the imaging control circuit 121 stops the additional acquisition. In addition, the processing circuit 131, triggered by the input of the stop instruction, reconstructs the MR image based on the first MR signal and the second MR signal acquired until the input of the stop instruction.

(Step Sa12)
The processing circuit 131 uses the reconstruction function 1321 to reconstruct an MR image based on the first MR signal and the second MR signal belonging to the largest class. Specifically, the processing circuit 131 reconstructs an MR image corresponding to the expiratory phase EA based on the first MR signal and the second MR signal arranged in the k-space KspE for the expiratory phase EA. Processing circuitry 131 outputs the reconstructed MR image to display 127 and storage device 129 . A display 127 displays the reconstructed MR image.

According to the configuration and the additional collection execution process described above, the following effects can be obtained.

According to the MRI apparatus 100 of this embodiment, the first trajectory indicating the first acquisition position of the first MR signal in the k-space is set, the first MR signal related to the subject P is sequentially acquired according to the first trajectory, and the first MR signal acquires the body motion state of the subject P during the acquisition of the first MR signal, and after a predetermined time has passed since the acquisition start time of the first MR signal, based on the data amount of the first MR signal corresponding to the body motion state, Determining whether additional acquisition of the second MR signal is necessary, and if it is determined that the additional acquisition is required, adding a second trajectory indicating the second acquisition position for the additional acquisition in k-space to the imaging protocol, and adding the first MR signal and a second MR signal acquired by performing an additional acquisition, an MR image can be reconstructed. At this time, the MRI apparatus 100 can reconstruct MR images using, for example, parallel imaging, compressed sensing, trained deep neural networks, or combinations thereof.

Further, according to the MRI apparatus 100, regarding the amount of data of the first MR signals classified into a plurality of classes using a threshold for discriminating body movement states into a plurality of classes, the maximum data amount among the plurality of classes is If the amount of data in the class is less than the reference value, it is determined that additional acquisition is required, and a trajectory different from the first trajectory for the first MR signal belonging to the maximum class is added to the imaging protocol as a second trajectory, and the maximum An MR image can be reconstructed using the first MR signal and the second MR signal belonging to the class. Further, according to the MRI apparatus 100, the first trajectory is set as the filling trajectory of the first MR signal in k-space by radial scanning, and the second trajectory is set so as not to overlap the first trajectory before acquiring the first MR signal. can do.

Further, according to the MRI apparatus 100, when it is determined that additional acquisition is necessary, the additional acquisition is performed, the imaging time increased due to the execution of the additional acquisition, and the scheduled imaging end time including the additional acquisition. At least one of them can be displayed on the display 127 . Further, according to the MRI apparatus 100, an instruction to stop the additional acquisition is input, the execution of the additional acquisition is stopped upon input of the stop instruction, and the first MR signal and the stop are triggered by the input of the stop instruction. An MR image can be reconstructed based on the second MR signals acquired until the input of the instruction.

For these reasons, according to the MRI apparatus 100, the image quality of the MR image desired by the operator can be selected in the largest class among the plurality of classes classified based on the body motion state of the imaging target due to the body motion of the subject P. If the data amount of the MR signals required for ensuring is insufficient, additional imaging can be performed to compensate for the lack of data amount, and the image quality of the MR image can be ensured. In addition, according to the MRI apparatus 100, information about additional acquisition can be provided to the operator, so that the operator can arbitrarily stop the additional acquisition and reconstruct the MR image. As a result, the increase in imaging time due to execution of additional acquisition can be appropriately adjusted according to the operator's instruction.

(First modification)
The processing circuit 131 according to the first modification determines whether or not additional acquisition is necessary for each of the plurality of classes, and uses the first MR signal and the second MR signal belonging to each of the plurality of classes to obtain MR data corresponding to each of the plurality of classes. Reconstruct the image.

FIG. 7 is a diagram showing an example of the first trajectory P1 and the second trajectory P2 in the k-space 11E for the expiratory phase EA and the k-space 11I for the inspiratory phase IA. As shown in FIG. 7, in this modified example, the first MR signal in the inspiratory phase IA is used for reconstruction of the MR image corresponding to the inspiratory phase IA without being discarded.

The additional collection execution process in this modified example will be described below. 8 and 9 are flowcharts showing an example of the procedure of the additional collection execution process regarding this modified example. Steps Sb1 to Sb5 and step Sb8 in FIG. 8 are the same processes as steps Sa1 to Sa5 and step Sa9 in FIG. 2, respectively, so description thereof will be omitted.

(Step Sb6)
The processing circuit 131 uses the determination function 1317 to compare the data amount of the first MR signal belonging to each of the plurality of classes with the reference value. Specifically, with respect to the data amount of the first MR signals classified into a plurality of classes, the processing circuit 131 requires additional collection when the data amount is less than a reference value in at least one of the plurality of classes. It is determined that More specifically, when the data amount of the expiratory phase EA is less than the reference value (Yes in step Sb6), the processing circuit 131 executes the process of step Sb7. Further, when the data amount of the intake phase IA is less than the reference value (Yes in step Sb6), the processing circuit 131 executes the process of step Sb7. When both the data amount of the expiratory phase EA and the data amount of the inspiratory phase IA are equal to or greater than the reference value (No in step Sb6), the processing circuit 131 executes the process of step Sb9.

(Step Sb7)
The processing circuit 131 uses the additional function 1319 to select a trajectory different from the first trajectory regarding the first MR signal belonging to a class having a data amount less than the reference value (hereinafter referred to as an additional class) as a second trajectory regarding additional acquisition. , is added to the imaging protocol with a corresponding additional class. Specifically, processing circuitry 131 identifies a first acquisition location for a first MR signal belonging to an additional class; A second trajectory indicating is added to the imaging protocol. These processes allow the processing circuitry 131 to add to the imaging protocol a second trajectory different from the first trajectory for the first MR signal belonging to the additional class for additional acquisitions. In addition, since the processing described in step Sa8 is executed for each additional class, the detailed description of the processing contents in this step is omitted.

(Step Sb9)
The processing circuit 131 uses the determination function 1317 to determine whether or not the imaging time set by the imaging conditions has elapsed. If the imaging time has not elapsed (No in step Sb9), the imaging control circuit 121 executes the process of step Sb2. Next, the processing circuit 131 repeats the processing of steps Sb3 to Sb9. At this time, the data amount of each of the plurality of classes in step Sb6 and the second trajectory in step Sb7 are updated. If the imaging time has elapsed (Yes in step Sb9), the imaging control circuit 121 executes the process of step Sb10.

(Step Sb10)
The imaging control circuit 121 performs additional acquisition according to the second trajectory to acquire a second MR signal belonging to the additional class. The imaging control circuit 121 stores the acquired second MR signal in the storage device 129 by attaching it to the additional class. At this time, the MR signals belonging to the additional class have the first MR signal and the second MR signal. After the process of this step, the processing circuit 131 may classify the second MR signals into a plurality of classes by the determination function 1317, and perform the determination described in step Sb6 again. Also, this step may be executed before the process of step Sb9.

(Step Sb11)
The processing circuit 131 uses the reconstruction function 1321 to reconstruct a plurality of MR images respectively corresponding to a plurality of classes based on the first MR signals and the second MR signals belonging to each of the plurality of classes. Specifically, the processing circuit 131 reconstructs an MR image corresponding to the expiratory phase based on the first MR signal and the second MR signal arranged in k-space for the expiratory phase. The processing circuit 131 also reconstructs an MR image corresponding to the inspiratory phase based on the first MR signal and the second MR signal arranged in the k-space related to the inspiratory phase. Processing circuitry 131 outputs the reconstructed MR image to display 127 and storage device 129 . A display 127 displays the reconstructed MR image.

According to the MRI apparatus 100 according to the first modification, regarding the data amount of the first MR signal classified into a plurality of classes using a threshold value for discriminating the body movement state into a plurality of classes, at least one of the plurality of classes In one class, if the amount of data is less than the reference value, determine that additional acquisition is required, add a second trajectory different from the first trajectory for the first MR signal belonging to the class for additional acquisition to the imaging protocol, Using the first MR signals and the second MR signals classified into multiple classes, MR images corresponding to each of the multiple classes can be reconstructed.

As a result, according to the MRI apparatus 100, for each of a plurality of classes classified according to the body motion state of the subject P, the amount of MR signal data required to ensure the image quality of the MR image desired by the operator is If there is a shortage, additional imaging can be performed so as to compensate for the lack of the data amount of the MR signal, and the image quality of the MR image can be ensured.

(Second modification)
The processing circuit 131 according to the second modification acquires the body motion state using MR images without using external devices such as a breathing belt and an optical camera. The process according to the second modification is executed in step Sa3 of FIG. 2 and step Sb3 of FIG. In the additional collection execution process according to the second modification, step Sa3 shown in FIG. 2 is executed after step Sa4 in FIG. 2, and step Sb3 shown in FIG. 8 is executed after step Sb4 in FIG.

The processing circuit 131 uses the reconstruction function 1321 to reconstruct an MR image (hereinafter referred to as a reference MR signal) based on a first MR signal (hereinafter referred to as a reference MR signal) acquired in a reference period shorter than the total period during which the first MR signals are successively acquired. , called the reference image). The reference period corresponds to the predetermined time of steps Sa4 and Sb4. Note that the reference period may be shorter than the predetermined time. Next, the processing circuit 131 generates partial images corresponding to each of the plurality of partial periods based on the first MR signals (hereinafter referred to as partial MR signals) acquired in each of the plurality of partial periods shorter than the predetermined time or reference period. to reconfigure. The reference periods and partial periods are stored in storage device 129 .

FIG. 10 is a diagram showing an example of temporal changes in the position U1 of the diaphragm, the reference period RP, and the partial period PP. As shown in FIG. 10, the diaphragm in the reference image reconstructed using the reference MR signal belonging to the reference period RP is imaged at a position obtained by averaging the positions U1 of the diaphragm over the reference period RP. On the other hand, the diaphragm in each of the plurality of partial images reconstructed using the partial MR signals belonging to the partial period PP is imaged at a position obtained by averaging the positions U1 of the diaphragm in each partial period PP. Although the plurality of partial periods PP shown in FIG. 10 do not overlap, each of the plurality of partial periods PP can be extended to the adjacent partial period by extending the partial period with respect to the center of each of the plurality of partial periods PP. may overlap.

Processing circuitry 131 performs registration between the reference image and each of the plurality of partial images through acquisition function 1315 . Processing circuitry 131 determines the relative position of the diaphragm position in each partial image with respect to the position of the diaphragm in the reference image through the registration. A relative position corresponds to, for example, a displacement of the diaphragm along the z-direction. The processing circuit 131 acquires the body motion state based on the relative position and the threshold. When the plurality of classes are two classes of the expiratory phase and the inspiratory phase, the threshold for discriminating the body motion state of the subject P into the plurality of classes corresponds to the position of the diaphragm in the reference image. If the number of classes is 3 or more, the threshold values may be determined by multiplying the position of the diaphragm by a predetermined coefficient. In these cases, threshold storage is not required in the second modification.

It should be noted that the processing circuit 131 may acquire the body motion state by using the acquisition function 1315 using a trained model that has been trained to input the reference image and each of the plurality of partial images and output the body motion state. good. A trained model is realized by, for example, a deep neural network, a convolutional neural network (CNN: Convolutional Neural Network), or the like.

According to the MRI apparatus 100 according to the second modification, the reference image is reconstructed based on the reference MR signals acquired in the reference period shorter than the total period during which the sequential acquisition of the first MR signals is performed, and A partial image corresponding to each partial period is reconstructed based on the partial MR signals acquired in each of a plurality of short partial periods, and the body motion state is acquired by performing registration between the reference image and the partial image. can do.

As a result, the body motion state of the subject P can be acquired without having the subject P wear a breathing belt or the like, so the burden on the subject P in MR imaging can be reduced. Moreover, since the body motion state of the subject P can be acquired without using an external device such as an optical camera, the cost for acquiring the body motion state can be reduced.

(Third modification)
The processing circuit 131 according to the third modification acquires the body motion state using non-reconstructed images without using external devices such as a breathing belt and an optical camera. The process according to the third modification is performed in step Sa3 of FIG. 2 and step Sb3 of FIG. In the additional collection execution process according to the third modification, step Sa3 shown in FIG. 2 is executed after step Sa4 in FIG. 2, and step Sb3 shown in FIG. 8 is executed after step Sb4 in FIG.

Processing circuitry 131 generates first non-reconstructed image data based on the reference MR signal via acquisition function 1315 . The processing circuit 131 generates second non-reconstructed image data corresponding to each of the plurality of partial periods based on the partial MR signal. To make the description concrete, MR data corresponding to the reference MR signal (hereinafter referred to as reference MR data) and MR data corresponding to the partial MR signal (hereinafter referred to as partial MR data) are, for example, Three-dimensional data defined by the kx, ky, and kz directions will be described. Processing circuitry 131 obtains first non-reconstructed image data by performing a Fourier transform on the reference MR data along the kz direction.

Similarly, processing circuitry 131 acquires second non-reconstructed image data by performing a Fourier transform on the partial MR data along the kz direction, via acquisition function 1315 . At this time, the first non-reconstruction image data and the second non-reconstruction image data correspond to projection data generated by projecting an MR image in the z direction from a direction perpendicular to the z direction. Note that the first non-reconstruction image data and the second non-reconstruction image data are not limited to projection data, and may be MR data.

FIG. 11 is a diagram showing an example of the projection image I1. The projection image I1 is generated by arranging the second non-reconstructed image data at different acquisition times along the time axis. As shown in FIG. 11, the position of the diaphragm corresponds to the boundary DL where the contrast changes. Note that the second non-reconstructed image data corresponding to the projection data may be acquired using, for example, the navigation echo method. FIG. 12 is a diagram showing the acquisition range NEA of the navigation echo method shown in the positioning image I2.

Processing circuitry 131 performs registration of the first non-reconstructed image data and the second non-reconstructed image data through acquisition function 1315 . Processing circuitry 131 determines the relative position of the diaphragm position in the second non-reconstructed image data with respect to the position of the diaphragm in the first non-reconstructed image data by the alignment. The processing circuit 131 acquires the body motion state based on the relative position and the threshold. When the plurality of classes are two classes of the expiratory phase and the inspiratory phase, the threshold for discriminating the body motion state of the subject P into the plurality of classes corresponds to the position of the diaphragm in the first non-reconstructed image data. . If the number of classes is 3 or more, the threshold values may be determined by multiplying the position of the diaphragm by a predetermined coefficient. In these cases, it is not necessary to store threshold values in this modified example.

In addition, the processing circuit 131 uses a trained model that has been trained to input the second non-reconstructed image data and the MR data corresponding to the first MR signal and output the body movement state by the acquisition function 1315, You may acquire a body movement state. In addition, the processing circuit 131 uses a learned model trained to input the first MR signal and the second non-reconstructed image data and output the classified first MR signal by the determination function 1317, and the first MR Signals may be classified.

According to the MRI apparatus 100 according to the third modification, the first non-reconstructed image data based on the reference MR signals acquired in the reference period shorter than the total period in which the sequential acquisition of the first MR signals is performed, and the reference period By performing registration based on the partial MR signals acquired in each of the shorter partial periods with the second non-reconstructed image data corresponding to each partial period, the motion state can be obtained.

(Fourth modification)
The processing circuit 131 according to the fourth modification reconstructs an MR image (hereinafter referred to as a state image) corresponding to the body motion state based on the first MR signal for each body motion state, and adjusts the noise intensity in the state image. The purpose is to determine whether or not additional collection is necessary based on the results. The process according to the fourth modification is used in place of the process of step Sa7 in FIG. 2 and step Sb6 in FIG. 8 in the additional collection execution process.

The processing circuit 131 uses the reconstruction function 1321 to reconstruct a state image corresponding to the body motion state based on the first MR signal for each body motion state. Specifically, the processing circuit 131 reconstructs state images corresponding to each of the plurality of classes based on the first MR signals belonging to each of the plurality of classes. At step Sa7 in FIG. 2, the processing circuit 131 reconstructs a state image corresponding to the maximum class based on the first MR signals belonging to the maximum class.

The processing circuit 131 determines whether or not additional acquisition is necessary based on the noise intensity in the state image by the determination function 1317 . Specifically, the processing circuit 131 generates a denoised image by applying a denoising filter to the state image. The processing circuit 131 calculates the sum of the squares of the pixel value differences between the state image and the denoised image as the noise intensity. Processing circuitry 131 compares the noise threshold stored in memory 129 with the noise intensity and determines that additional acquisitions are required if the noise intensity is greater than the noise threshold.

Note that the processing circuit 131 does not necessarily need to reconstruct the state image to identify the noise intensity. For example, the processing circuitry 131 may identify the edge data in k-space where the first MR signal is located as the noise intensity. In this case, the noise threshold corresponds to, for example, the signal collected by the receiving circuit 119 without applying a transmitted RF wave to the subject P. FIG.

In addition, the processing circuit 131 uses a learned model trained to input the state image or the first MR signal belonging to each of a plurality of classes and output whether or not additional acquisition is necessary. may be determined. A trained model is implemented by a machine learning model such as a deep neural network, a support vector machine, or a random forest.

According to the MRI apparatus 100 according to the fourth modification, a state image corresponding to the body motion state is reconstructed based on the first MR signal for each body motion state, and additional acquisition is required based on the noise intensity of the state image. No can be determined. Since the necessity of additional acquisition is determined based on the noise intensity of the MR image, a high-quality MR image with reduced noise can be reconstructed.

(Fifth Modification)
The processing circuit 131 according to the fifth modification reconstructs a state image for each body motion state based on the first MR signal for each body motion state, and additionally acquires based on the detected amount of artifacts included in the state image. Determine the necessity of The processing according to the fifth modification is used in place of the processing of step Sa7 of FIG. 2 and step Sb6 of FIG. 8 in the additional collection execution processing.

The processing circuit 131 uses the reconstruction function 1321 to reconstruct a state image corresponding to the body motion state based on the first MR signal for each body motion state. Specifically, the processing circuit 131 reconstructs state images corresponding to each of the plurality of classes based on the first MR signals belonging to each of the plurality of classes. At step Sa7 in FIG. 2, the processing circuit 131 reconstructs a state image corresponding to the maximum class based on the first MR signals belonging to the maximum class.

The processing circuit 131 uses the determination function 1317 to determine whether or not additional acquisition is necessary based on the detected amount of artifacts included in the state image. Specifically, the processing circuitry 131 detects artifacts in the state image using filters for detecting artifacts (hereinafter referred to as detection filters). Detection filters may be stored in advance in the storage device 129 for each scanning method, for example. It should be noted that the processing circuit 131 may detect artifacts by performing template matching on the state image using a template representing the shape of a typical artifact instead of the detection filter.

FIG. 13 is a diagram showing an example of body motion artifact MAR rendered in a partial image when the scanning method is radial scanning. FIG. 14 is a diagram showing an example of body motion artifact MAC in a partial image when the scanning method is Cartesian scanning. As shown in FIGS. 13 and 14, the lack of data amount in the largest class can appear as artifacts in the state image.

Processing circuitry 131 uses the detected artifacts to calculate the amount of artifact detection in the state image. The artifact detection amount is, for example, the number of pixels included in the artifact region in the state image. The processing circuit 131 compares the threshold stored in the storage device 129 with the amount of detected artifacts, and determines that additional acquisition is necessary if the amount of detected artifacts is greater than the threshold.

Note that the processing circuit 131 inputs the state image or the first MR signal belonging to each of the plurality of classes, and uses a learned model trained to output whether or not additional acquisition is necessary. may be determined. A trained model is realized by a machine learning model such as a deep neural network, a support vector machine, or a random forest, for example.

According to the MRI apparatus 100 according to the fifth modification, a state image corresponding to the body motion state is reconstructed based on the first MR signal for each body motion state, and based on the detected amount of artifact in the state image, additional It is possible to determine whether collection is necessary. Whether or not to perform additional acquisition is determined based on the amount of artifact detected in the MR image, regardless of the amount of data in the largest class, so that a high-quality MR image with reduced artifacts can be reconstructed.

(Sixth modification)
The imaging control circuit 121 according to the sixth modification performs additional acquisition subsequent to the sequential acquisition of the first MR signal. It should be noted that whether or not to perform additional collection following sequential collection may be selected according to the sequence type in sequential collection. For example, when sequential acquisition is performed in a gradient echo sequence, processing according to the sixth modification is performed.

The processing circuit 131 uses the determination function 1317 to determine the time required to determine whether additional acquisition is necessary and to add the second trajectory to the imaging protocol after a predetermined period of time (hereinafter referred to as determination additional time) as the first MR. Whether or not additional collection is necessary is determined during a time (hereinafter referred to as determination start limit time) subtracted from the scheduled end time of signal sequential collection. The determination additional time is stored in the storage device 129 in advance. Specifically, the processing circuit 131 calculates the time required for sequential acquisition (hereinafter referred to as sequential acquisition time) based on the imaging conditions prior to executing the sequential acquisition of the first MR signals. The processing circuit 131 calculates the scheduled end time by adding the successive collection time to the collection start time corresponding to the start time of step Sa2. The processing circuit 131 calculates the determination start limit time by subtracting the determination additional time from the scheduled end time. The processing circuit 131 determines whether or not additional collection is required after a predetermined period of time has elapsed and before the determination start limit time is reached.

FIG. 15 shows, in the additional collection execution process, a collection start time Ts1, a predetermined time PT, a predetermined time end time Tppt, a judgment start limit time Tbdt, an additional judgment time DT, and an additional collection start time Ts2. It is a figure which shows an example with. As shown in FIG. 15, the processing circuit 131 determines whether or not additional collection is necessary between the end time Tppt of a predetermined period and the determination start limit time Tbdt.

According to the MRI apparatus 100 according to the sixth modification, in order to perform additional acquisition subsequent to the sequential acquisition of the first MR signal, after a predetermined time has elapsed, it is determined whether or not additional acquisition is necessary and the imaging protocol is changed. It is possible to determine whether or not additional acquisition is necessary until the time obtained by subtracting the time required to add the second trajectory from the scheduled end time of sequential acquisition. As a result, additional acquisition can be performed in the same state as the magnetization during transmission of the transmission RF wave during acquisition of the first MR signal. At the start of the additional acquisition, there is no need to apply a dummy RF pulse that does not acquire MR signals to the subject P in order to stabilize the magnetization after T1 recovery, so the total imaging time including the additional acquisition can be shortened. can.

If the sequential acquisition and the additional acquisition are not performed continuously, the subject P may mistakenly recognize that the examination has ended due to the stop or reduction of the collected sound during the transition period from the sequential acquisition to the additional acquisition. There is Since the MRI apparatus 100 according to the sixth modification successively acquires the first MR signal and additionally acquires the first MR signal, the sound caused by the acquisition of the MR signal (hereinafter referred to as the acquired sound) is continuously generated. can be made This prevents the subject P from erroneously recognizing the end of the examination due to the stoppage of the collected sound, so that the psychological burden on the subject P can be reduced.

(Seventh Modification)
When the operator inputs the total imaging time including the sequential acquisition and the additional acquisition of the first MR signal via the interface 125, the processing circuit 131 according to the seventh modification changes the total imaging time and the first trajectory. determine a second trajectory based on the

Prior to the processing of step Sa1 in the additional collection execution processing, or during a period until a predetermined time elapses from the collection start time, the operator instructs via the interface 125 to input the total imaging time. The total imaging time corresponds to the upper limit of the imaging time for the subject P. It should be noted that an imaging extension time may be input instead of the total imaging time.

Processing circuitry 131, via additional function 1319, determines a second trajectory based on the total imaging time (or imaging extension time) and the first trajectory. Specifically, the processing circuitry 131 determines the number of second trajectories based on the total imaging time (or the extended imaging time). Processing circuitry 131 then determines imaging conditions for the second trajectory such that the first trajectory and the second trajectory are evenly distributed in k-space. For example, as shown in FIG. 4, a first trajectory is arranged in k-space 11, two regions CP2 in which a plurality of first trajectories do not exist exist in k-space 11 at a ratio of 1:2, and the total imaging time is If the imaging extension time based on is a time during which additional acquisition can be performed along 3n (n is a natural number) second trajectories, the processing circuit 131 performs n The imaging conditions are determined so that the second trajectories are evenly arranged, and the imaging conditions are determined so that the 2n second trajectories are evenly arranged in the wider area.

According to the MRI apparatus 100 according to the seventh modification, the total imaging time including the sequential acquisition and the additional acquisition of the first MR signal is input, and based on the input total imaging time and the first acquisition position, the 2 collection locations can be determined. Since the additional amount of the second trajectory can be controlled based on the upper limit of the imaging time, the first trajectory and the second trajectory can be arranged in the k-space as evenly as possible within the total imaging time desired by the operator. . As a result, the MRI apparatus 100 can generate an MR image with improved image quality within the total imaging time limited by the operator.

(Eighth modification)
The processing circuit 131 according to the eighth modification is to apply the additional acquisition execution processing when there are multiple factors of body motion of the subject P. FIG. In the following, to make the explanation concrete, the body motion factors are assumed to be heartbeat and respiration.

FIG. 16 is a diagram showing an example of a respiratory waveform RWF and an electrocardiographic waveform EWF. The storage device 129 stores, for example, four states (expiratory phase EA and systolic state, expiratory phase EA and diastolic , inspiratory phase IA and systole state, and inspiratory phase IA and diastole state) are stored according to the body motion state.

The additional collection execution process is the same as described in the embodiment, the first modified example, and the like, so the description is omitted. Note that the processing circuit 131 may classify the first MR signal according to the state of body movement by a gating method using the electrocardiographic waveform EWF or the respiratory waveform RWF. The processing circuitry 131 may discard the first MR signals related to body motion states that are not related to the imaging subject by a synchronous method.

(Application example)
The processing circuit 131 according to the application example sets the upper limit of the sum of the data amount of the first MR signal and the data amount of the second MR signal (hereinafter referred to as the acquired data amount) based on the Specific Absorption Rate (SAR). is calculated, and when the amount of acquired data reaches the upper limit during execution of additional acquisition, additional acquisition is stopped and an MR image is reconstructed.

FIG. 17 is a diagram showing an example of the configuration of the processing circuit 131 according to the application. As shown in FIG. 17, the processing circuit 131 further has a calculation function 1323 in addition to the system control function 1311, setting function 1313, acquisition function 1315, determination function 1317, addition function 1319 and reconstruction function 1321 described above. Calculation function 1323 is stored in storage device 129 in the form of a computer-executable program. The processing circuit 131 reads a program corresponding to the calculation function 1323 from the storage device 129 and executes it, thereby realizing the calculation function 1323 corresponding to this program. The calculation function 1323 of the processing circuit 131 is an example of a calculation unit.

The processing circuit 131 calculates the data amount of the first MR signal (hereinafter referred to as The upper limit (hereinafter referred to as the upper limit data amount) of the sum (hereinafter referred to as the sum data amount) of the sum (hereinafter referred to as the sum data amount) of the data amount of the second MR signal (hereinafter referred to as the second data amount) is calculated. . For example, the processing circuit 131 sets the collectible second data amount (hereinafter referred to as collectable data amount) so that the SAR due to the sequential collection of the first MR signal and the additional collection of the second MR signal is equal to or less than the upper limit SAR. decide. If the amount of data that can be collected is zero, processing circuitry 131 may indicate on display 127 that no additional collections are possible. Processing circuitry 131 calculates the first data amount based on the imaging conditions. The processing circuit 131 calculates the sum of the first data amount and the collectible data amount as the upper limit data amount. The processing circuit 131 uses the calculation function 1323 to calculate the sum data amount when the second MR signal is acquired in the additional acquisition execution process.

The processing circuit 131 further determines, by the determination function 1317, whether or not the sum data amount has reached the upper limit data amount when executing the additional collection. Triggered by the sum data amount reaching the upper limit data amount, the processing circuit 131 outputs an instruction to stop the additional collection (hereinafter referred to as a collection stop instruction) to the imaging control circuit 121 .

The imaging control circuit 121 stops the additional collection when the sum data amount reaches the upper limit data amount, that is, in response to the input of the collection stop instruction from the processing circuit 131 . At this time, display 127 may indicate that additional collection has been stopped.

Triggered by the sum data amount reaching the upper limit data amount, the processing circuit 131 causes the reconstruction function 1321 to reconstruct an MR image using the first MR signal and the second MR signal.

According to the MRI apparatus 100 according to the application example, the data amount of the first MR signal and the data amount of the second MR signal are determined based on the specific absorption rate and the imaging conditions regarding the subject P determined before acquisition of the first MR signal. Calculate the upper limit of the sum of , further determine whether the sum has reached the upper limit when executing additional collection, stop additional collection when the sum reaches the upper limit, and stop the additional collection when the sum reaches the upper limit Using this as a trigger, an MR image can be reconstructed.

As a result, according to the MRI apparatus 100, additional acquisition can be performed without exceeding the upper limit SAR determined before imaging the subject P, so both the safety of the subject P and the image quality of the MR image are ensured. be able to.

(others)
According to some embodiments described above, the MRI apparatus 100 comprises imaging control circuitry 121 and processing circuitry 131 . The imaging control circuit 121 acquires MR signals according to the first pulse sequence set in the imaging protocol. The processing circuit 131 determines whether additional MR signals need to be acquired by determining the image quality based on the already acquired MR signals during acquisition by the first pulse sequence. Processing circuitry 131 adds a second pulse sequence to the imaging protocol for additional acquisitions if it is determined that additional acquisitions are required. The imaging control circuit 121 acquires MR signals according to the added second pulse sequence. A processing circuit 131 reconstructs an MR image based on the MR signals from the first pulse sequence and the second pulse sequence.

In some of the embodiments described above, the processing circuit 131 estimates a shortage of data in the MR signal during execution of the first pulse sequence, and if a shortage is estimated, the imaging protocol being executed takes the first step. Add 2 pulse sequences. The second pulse sequence is executed following the first pulse sequence or after a certain period of time. This process allows data acquisition to proceed automatically until the amount of MR signals required for reconstruction has been acquired.

In some of the embodiments described above, the processing circuit 131 determines whether or not additional imaging is performed by comparing the data amount of the MR signal with a reference value. However, this embodiment is not limited to this. The processing circuitry 131 may use the trained model to determine the presence or absence of additional imaging.

FIG. 18 is a diagram schematically showing an example of inputs and outputs of trained models according to another embodiment. As shown in FIG. 18, the trained model is trained to input MR images and output image quality tolerances. A trained model is, for example, a neural network with two or more layers. A neural network structure includes, for example, one or more CNNs, one or more fully connected layers, and a classifier. The structure of the neural network according to this embodiment is not particularly limited, and any structure may be used as long as it is possible to input an MR image and output an image quality tolerance.

The image quality tolerance is a numerical value or code indicating the degree to which the image quality of the input MR image is acceptable or unacceptable. For example, as shown in FIG. 18, the image quality tolerance is a code indicating that the image quality of the input MR image is acceptable (hereinafter referred to as an acceptable code) and a code indicating that the image quality is not acceptable (hereinafter referred to as an unacceptable code). (referred to as ).

A neural network may be trained, for example, by supervised learning, based on a large number of training samples. Learning of a neural network, that is, generation of a trained model may be performed by a computer such as a model learning device. The model learning device may be incorporated in the MRI apparatus 100, or may be a separate computer.

For example, supervised learning includes MR images as inputs and image quality tolerances as teachers. The image quality tolerance is determined in advance by an annotator by observing the input MR image and determining whether or not it is acceptable. Note that the image quality tolerance is not limited to manual determination, and may be determined by image processing or the like. Criteria for acceptability are determined according to whether or not the diagnosis is tolerable. For example, if noise or artifacts caused by missing data or the like are superimposed on an image and the image is determined to be unsuitable for diagnosis, it is determined to be "unacceptable." It should be noted that it is not always necessary to clarify the criteria for acceptance or non-acceptance. The model learning device applies a neural network to an MR image as an input, performs forward propagation processing, and outputs an image quality tolerance (hereinafter referred to as an estimated image quality tolerance). Next, the model learning device applies the difference (error) between the estimated image quality tolerance and the image quality tolerance as a teacher to the neural network, performs backpropagation processing, and calculates a gradient vector. Next, the model learning device updates parameters such as the weighting matrix and bias of the neural network based on the gradient vector. A learned model is generated by repeating the forward propagation process, the back propagation process, and the parameter update process while changing the learning sample.

Next, a brief description will be given of the operation of the trained model during operation. Processing circuitry 131 performs image reconstruction on the acquired first MR signals to reconstruct an MR image. As the acquired first MR signal, for example, an MR signal acquired after a predetermined time has elapsed from the start of execution of the first pulse sequence is used. The reconstruction method at this stage does not require the reconstructed image to have an image quality that can withstand diagnosis, so a simple and rapid method may be used. For radial scanning, for example, the Jackson method or the gridding method may be used.

When the MR image is generated, processing circuitry 131 applies the learned model to the generated MR image and outputs image quality tolerance. If a non-tolerant code is output as an image quality tolerance, processing circuitry 131 determines that additional acquisition is required. In this case, a second pulse sequence is added to the imaging protocol. If the tolerance code is output as the image quality tolerance, processing circuitry 131 determines that no additional acquisition is required.

The added second pulse sequence is typically a trajectory with the same or different acquisition position as compared to the first pulse sequence. However, this embodiment is not limited to this. The second pulse sequence may be changed in flip angle, transmission frequency of various RF pulses, and echo order compared to the first pulse sequence.

For example, when the first pulse sequence and the second pulse sequence are the FSE (Fast Spin Echo) method, a trajectory having the same acquisition position as the first pulse sequence and a different echo order is added as the second pulse sequence. be done. Any order such as a sequential order, a centric order, or an arbitrary order may be used as the echo order. As a result, it is possible to reduce factors of image quality deterioration due to echo order dependence.

For example, when the first pulse sequence and the second pulse sequence are of the multi-segment EPI (Echo Planar Imaging) method, trajectories with different segment positions in k-space are added in units of segments as the second pulse sequence. For example, if the first pulse sequence was acquired according to an even phase-encoding trajectory, then the odd phase-encoding trajectory may be added as a second pulse sequence. Also, if data acquisition is performed according to the even and odd phase-encoding order of the first pulse sequence, the central phase-encoding trajectory may be added as the second pulse sequence. As a result, it is possible to reduce image quality degradation factors due to segment position dependency.

For example, when the first pulse sequence and the second pulse sequence include fat saturation pulses, fat saturation pulses having a different transmission frequency and flip angle from the first pulse sequence may be added as the second pulse sequence. good. It should be noted that the present invention is applicable not only to fat saturation pulses, but also to arbitrary component suppression pulses such as water suppression pulses. As a result, the second pulse sequence can enhance the suppressing effect more than the first pulse sequence.

The changes in the transmission frequency and flip angle described above are not limited to suppression pulses, and can be applied to arbitrary RF pulses such as 90-degree pulses and 180-degree pulses.

Note that the image quality tolerance, which is the output of the trained model, is not limited to the two-class classification as described above, and may be other-class classification. For multi-class classification, a continuous or discrete value is output that indicates the acceptable degree of image quality.

The trained model may be trained to output the necessity of additional imaging instead of the image quality tolerance. Not allowing image quality corresponds to requiring additional imaging, and allowing image quality corresponds to not requiring additional imaging. In the case of two-class classification, learning may be performed so that a code indicating that additional imaging is required is output instead of the non-permissible code, and a code indicating that additional imaging is not required is output instead of the permissible code. In the case of other class classification, a final output layer for inputting image quality tolerance and outputting necessity of additional imaging may be provided in the output layer of the trained model.

As in the various embodiments described above, the need for additional collection may be determined without using the trained model. In these embodiments, the processing circuit 131 calculates reliability information for evaluating image quality based on the acquired first MR signal and determines whether additional acquisition is necessary based on the reliability information. Reliability information includes the amount of acquired MR signal data, the degree of difference with respect to the subject model, the amount of detected artifacts associated with body motion of the subject, and the degree of difference in pixel values associated with contrast agent delay.

An example in which the reliability information is the amount of collected MR signal data is described in the above embodiment, the first modification, and the like. In this case, for example, as shown in step Sa7 of FIG. 3, the necessity of additional collection is determined by comparing the amount of data with a reference value.

Next, a case where the reliability information is the degree of dissimilarity with respect to the object model will be described. A subject model is prior knowledge of the subject's anatomy that is obtained by analyzing MR images of various subjects. The anatomical structure of the subject depicted in the MR image has a complex shape with respect to the subject model.

The processing circuitry 131 calculates the degree of difference between the MR image and the subject model based on the MR image based on the acquired MR signals and the subject model. The dissimilarity is calculated for the entire MR image. Next, the processing circuit 131 compares the degree of difference with a preset reference value. If the degree of dissimilarity is greater than the reference value, processing circuitry 131 determines that the MR image contains noise or artifacts due to missing data. Even if the amount of data is insufficient even after all trajectories have been collected, the presence of noise or artifacts can be determined with high accuracy. In this case, processing circuitry 131 determines that additional acquisitions are required. If the dissimilarity is less than the reference value, processing circuitry 131 determines that the MR image does not contain noise or artifacts associated with missing data. In this case, processing circuitry 131 determines that no additional acquisition is required.

Specifically, the subject model is determined based on a total variation (TV) model, wavelet model, neural network, or the like. The case of the total variation model will be explained. For example, total variation is calculated for each of a number of MR images free of noise or artifacts associated with missing data. These multiple total variation statistics are used as subject models. As another example, total variation may be calculated for each of a number of MR images in which noise or artifacts associated with missing data are present, and statistics of these multiple total variations may be used as subject models. The processing circuit 131 calculates the total variation of the input MR image based on the acquired first MR signal, and calculates the difference between the calculated total variation and the statistical value as the dissimilarity. The degree of difference represents the difference in image structure between the input MR image and the object model. As described above, the processing circuit 131 determines the necessity of additional imaging based on the comparison between the difference (dissimilarity) and a predetermined reference value.

Note that the total variation may be calculated for the entire MR image, or may be calculated for each local region or pixel. Wavelet models can also be used in a similar manner.

Next, the case of a neural network will be explained. This neural network is a neural network for image recognition that is trained to receive an MR image as an input and output the presence or absence of noise or artifacts due to missing data. Such neural networks are trained based on a plurality of training samples using, for example, the L1 error or the L2 error. This neural network can be learned, for example, by supervised learning. For example, in the case of supervised learning, an MR image is used as an input, and the presence or absence of noise or artifacts due to missing data is used as a teacher.

Next, a case where the reliability information is the amount of artifacts detected due to the subject's body motion will be described. This embodiment corresponds to the fifth modification. By setting the reliability information to the artifact detection amount, it is possible to compensate for the lack of data amount due to data corruption caused by the subject's conscious body movement or unconscious or physiological body movement by additional imaging. can.

Next, a case where the reliability information is the degree of difference between pixel values due to the delay of the contrast agent will be described. Due to the contrast agent delay, the state in which the contrast agent does not reach the ROI continues during all or part of the contrast imaging period. Therefore, the delay of the contrast medium makes it impossible to obtain a desired MR image that renders the contrast tissue. The contrast agent delay can be evaluated by the difference between the pixel value of a pixel when the contrast agent has arrived and the pixel value of the same pixel when it has not. For example, the pixel values of the pixels when the contrast medium reaches them are collected from a plurality of MR images, and statistical values such as average and median values of the pixel values (hereinafter referred to as contrast reference values) are recorded in advance. The contrast enhancement reference value is preferably recorded for each type of tissue such as heart and liver, and for each type of contrast enhancement method such as T1-weighting and T2-weighting.

The processing circuit 131 generates an MR image based on the acquired MR signals, and the difference between the pixel values of the pixels of the tissue to be contrast-enhanced in the generated MR image and the contrast enhancement reference value corresponding to the tissue to be contrast-enhanced. Calculate The difference corresponds to reliability information, that is, the degree of difference in pixel values associated with contrast agent delay. The tissue to be contrast-enhanced may be specified by subjecting the MR image to arbitrary image recognition processing, or may be designated by the user via the interface 125 .

Next, the processing circuit 131 compares the degree of difference with a preset reference value. If the dissimilarity is greater than the reference value, processing circuitry 131 determines that the contrast agent has not arrived. In this case, processing circuitry 131 determines that additional acquisitions are required. Additional acquisitions are made after the subject is injected with contrast again. If the difference is less than the reference value, processing circuitry 131 determines that the contrast agent has arrived. In this case, processing circuitry 131 determines that no additional acquisition is required.

According to at least one of the embodiments described above, additional acquisition can be performed by adaptively estimating the amount of MR signal data necessary for reconstructing an MR image according to the object to be imaged. The additional acquisition can ensure the quality of the MR image.

While several embodiments of the invention have been described, these embodiments have been presented by way of example and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, replacements, and modifications can be made without departing from the scope of the invention. These embodiments and their modifications are included in the scope and spirit of the invention, as well as the scope of the invention described in the claims and equivalents thereof.

Reference Signs List 100 MRI apparatus 101 static magnetic field magnet 103 gradient magnetic field coil 105 gradient magnetic field power supply 107 bed 109 bed control circuit 111 bore 113 transmission circuit 115 transmission coil 117 reception coil 119 reception circuit 121 imaging control Circuit 125 Interface 127 Display 129 Storage device 131 Processing circuit 1071 Top plate 1311 System control function 1313 Setting function 1315 Acquisition function 1317 Determination function 1319 Addition function 1321 Reconfiguration function 1323 Calculation function

Claims (21)

a first acquisition unit that acquires MR signals according to a first pulse sequence set in an imaging protocol;
a determination unit that determines whether or not additional acquisition of MR signals is necessary by determining image quality based on already acquired MR signals during acquisition by the first pulse sequence;
an adding unit that adds a second pulse sequence for the additional acquisition to the imaging protocol if it is determined that the additional acquisition is required;
a second acquisition unit that acquires MR signals according to the added second pulse sequence;
a reconstruction unit that reconstructs an MR image based on the MR signals generated by the first pulse sequence and the second pulse sequence;
A magnetic resonance imaging apparatus comprising
further comprising an acquisition unit that acquires a body motion state of the subject during acquisition of MR signals by the first pulse sequence;
The determination unit determines whether or not the additional acquisition is necessary based on the body motion state and the amount of data of the acquired MR signals.
Magnetic resonance imaging equipment. 2. The magnetic resonance imaging apparatus according to claim 1, wherein said determination unit determines whether or not said additional acquisition is necessary after a predetermined time has elapsed from the start of acquisition by said first pulse sequence. The determination unit discriminates the MR signals by the first pulse sequence into a plurality of classes according to the body motion state, and specifies the data amount of the MR signals by the first pulse sequence for each of the plurality of classes. , identifying the class with the largest amount of data among the plurality of classes, and determining that the additional collection is necessary when the amount of data in the largest class is less than a reference value;
The adding unit adds, as the second pulse sequence, a trajectory different from the MR signal by the first pulse sequence belonging to the largest class,
The reconstruction unit reconstructs the MR image based on the MR signals from the first pulse sequence and the second pulse sequence belonging to the largest class.
The magnetic resonance imaging apparatus according to claim 1 . The determination unit discriminates the MR signals by the first pulse sequence into a plurality of classes according to the body motion state, and specifies the data amount of the MR signals by the first pulse sequence for each of the plurality of classes. , if the amount of data in at least one class among the plurality of classes is less than a reference value, determining that the additional collection is necessary;
The adding unit adds, as the second pulse sequence, a trajectory different from the MR signal by the first pulse sequence belonging to the at least one class,
The reconstruction unit reconstructs the MR image corresponding to each of the plurality of classes, based on the MR signals from the first pulse sequence and the second pulse sequence classified into the plurality of classes.
The magnetic resonance imaging apparatus according to claim 1 . The reconstruction unit reconstructs an MR image for each body motion state based on the MR signal from the first pulse sequence for each body motion state,
The determination unit determines whether or not the additional acquisition is necessary based on the noise intensity in the MR image for each body movement state.
The magnetic resonance imaging apparatus according to claim 1 . The reconstruction unit reconstructs an MR image for each body motion state based on the MR signal from the first pulse sequence for each body motion state,
The determination unit determines whether or not the additional acquisition is necessary based on the amount of artifact detected in the MR image for each body motion state.
The magnetic resonance imaging apparatus according to claim 1 . 2. The magnetic resonance imaging apparatus according to claim 1 , wherein said acquisition unit acquires the respiration level of said subject as said body movement state. The acquisition unit acquires first non-reconstructed image data based on MR signals acquired in a reference period shorter than the total period of the first pulse sequence, and in each of a plurality of partial periods shorter than the reference period. 2. A magnetic resonance imaging apparatus according to claim 1 , wherein said body motion state is acquired based on registration with second non-reconstructed image data corresponding to each of said partial periods based on MR signals. The reconstruction unit reconstructs a reference image based on MR signals acquired in a reference period shorter than the period of the first pulse sequence, and MR signals acquired in each of a plurality of partial periods shorter than the reference period. reconstructing partial images corresponding to each of the partial periods based on
The acquisition unit acquires the body movement state based on alignment between the reference image and the partial image.
The magnetic resonance imaging apparatus according to claim 1 . 2. The magnetic resonance imaging apparatus according to claim 1, wherein said first pulse sequence and said second pulse sequence include, as a parameter, a trajectory indicating acquisition positions of MR signals in k-space. 11. Magnetic resonance according to claim 10 , further comprising a setting unit that sets a trajectory of said second pulse sequence so as not to overlap a trajectory of said first pulse sequence before acquisition by said first pulse sequence is started. imaging device. a calculation unit that calculates an upper limit of a sum of data amounts of MR signals obtained by the first pulse sequence and the second pulse sequence based on the specific absorption rate determined before the start of acquisition by the first pulse sequence; further prepared,
The second collection unit stops collection by the second pulse sequence when the sum reaches the upper limit,
The reconstructing unit reconstructs the MR image based on the MR signals from the first pulse sequence and the MR signals from the second pulse sequence that have been acquired before stopping.
The magnetic resonance imaging apparatus according to claim 1. When it is determined that the additional acquisition is necessary, at least one of the execution of the additional acquisition, the increased imaging time due to the execution of the additional acquisition, and the scheduled end time of imaging including the additional acquisition. 2. The magnetic resonance imaging apparatus according to claim 1, further comprising a display for displaying one. 2. The magnetic resonance imaging apparatus according to claim 1, wherein said second acquisition unit performs acquisition by said second pulse sequence following acquisition by said first pulse sequence. The determination unit determines, after a predetermined time has elapsed from the start of acquisition by the first pulse sequence, the time required to determine whether the additional acquisition is necessary and to add the trajectory of the second pulse sequence. 15. The magnetic resonance imaging apparatus according to claim 14 , wherein whether or not said additional acquisition is necessary is determined during a period up to a time subtracted from the scheduled acquisition end time by the pulse sequence. further comprising an input unit for inputting the total imaging time,
The adding unit determines the second pulse sequence based on the total imaging time and the first pulse sequence.
The magnetic resonance imaging apparatus according to claim 1. further comprising an input unit for inputting an instruction to stop acquisition by the second pulse sequence;
The second acquisition unit stops acquisition by the second pulse sequence triggered by the input of the instruction to stop,
The reconstruction unit reconstructs the MR image based on the MR signals from the first pulse sequence and the MR signals from the second pulse sequence acquired until the stop instruction is input.
The magnetic resonance imaging apparatus according to claim 1. 2. The magnetic resonance imaging apparatus according to claim 1, wherein said reconstruction unit reconstructs said MR image using parallel imaging, compressed sensing, or a trained neural network. 2. The magnetic resonance imaging apparatus according to claim 1, wherein said determination unit calculates reliability information for evaluating said image quality based on said acquired MR signals, and determines necessity of said additional acquisition based on said reliability information. imaging device. The reliability information is the amount of data of the acquired MR signals, the degree of difference with respect to the subject model, the detected amount of artifacts associated with body movement of the subject, and the degree of difference in pixel values associated with contrast agent delay. 20. A magnetic resonance imaging apparatus according to Item 19 . The determination unit determines whether or not the additional acquisition is necessary based on a trained model that inputs an MR image and outputs an image quality tolerance of the MR image and an MR image based on the acquired MR signal. 2. The magnetic resonance imaging apparatus according to claim 1, wherein