JP5165791B2 - Magnetic resonance imaging device - Google Patents

Description

  The present invention relates to spectroscopic imaging, and more particularly to a technique for improving the accuracy of eddy current correction for correcting spectral distortion caused by eddy currents.

  In the nuclear magnetic resonance signal measured by MRI, a chemical shift phenomenon having slightly different resonance frequencies occurs due to the difference in molecular structure. Using this phenomenon, MRS (Magnetic Resonance Spectroscopy) that separates nuclear magnetic resonance signals for each molecule (metabolite) and obtains a spectrum, and CSI (Chemical Shift) that images a spatial signal intensity distribution for each metabolite. Imaging (chemical shift imaging) and MRSI (Magnetic Resonance Spectroscopic Imaging) are known.

  The main metabolites of the human body that can be detected by MRS and MRSI include choline (Cho), creatine (Cr), N-acetylaspartic acid (NAA), lactic acid (Lac), and the like. From the amount of these metabolites, it is possible to determine the degree of progression and early diagnosis of metabolic disorders such as cancer. It is also considered possible to perform non-invasive diagnosis of tumor malignancy.

  In MRS and MRSI, an eddy current is generated by a gradient magnetic field applied during measurement. Eddy currents cause static magnetic field inhomogeneities spatially and temporally and distort the shape of the spectrum obtained by measurement. This spectral distortion is generally corrected using the phase value of the reference signal. For example, a water signal having a signal intensity greater than that of the metabolite is used as the reference signal. A phase value is acquired from this water signal and used for correction (for example, see Non-Patent Document 1). In the method disclosed in Non-Patent Document 1, spatial and temporal phase values are calculated from this water signal, phase correction is performed on metabolite image data, and spectral distortion due to eddy current is corrected.

  In CSI or MRSI, the number of matrices (voxels) to be measured is as very small as about 8 × 8 to 32 × 32 from the viewpoint of measurement time and SNR (signal to noise ratio). For this reason, truncation occurs due to the Fourier transform performed in the image reconstruction, and a signal of a distant voxel is mixed. As a result, when the static magnetic field inhomogeneity exists, a water signal having a frequency different from that of the water signal in the target voxel is mixed.

  When water signals of different frequencies are mixed, a so-called phase jump region occurs in which the variation in the phase value per unit time is prominently larger than other locations in the time variation of the phase value used for eddy current correction. To do. The magnitude of the phase change amount in this phase jump region is proportional to the concentration of the mixed water signal. Here, this phase skip will be described with reference to FIG.

  FIG. 13A shows the spectrum of the water signal by computer simulation. Water 1 and Water 2 in FIG. 13A are water signal spectra having frequencies of 2 Hz and 5 Hz, respectively, and a concentration ratio of 1.0 to 0.9. FIG. 13B shows the time change of the phase value in the time domain when the two signals are mixed. As shown in FIGS. 13A and 13B, a phase jump proportional to the concentration ratio occurs at a time interval of 1 / Δf with respect to the frequency difference Δf. Here, static magnetic field fluctuations due to eddy currents are added to cause a gradual phase change.

  Here, the state before and after correction when correcting the spectral distortion due to the eddy current of the metabolite using the phase data indicating the change shown in FIG. 13B in the time direction is shown in FIGS. d). FIG. 13C shows a state before eddy current correction, and FIG. 13D shows a state after eddy current correction. As shown in FIG. 13D, when correction is performed using phase data having a phase jump, ringing artifacts are generated by the phase jump, and the spectrum is deteriorated conversely by the eddy current correction process.

  On the other hand, for example, there is a method of reducing ringing artifacts by applying a low-pass filter to the spectrum of the water signal (see, for example, Non-Patent Document 2). In addition, there is a method of correcting a phase jump appearing in the phase data and reducing ringing artifacts (see, for example, Non-Patent Document 3). Here, the timing at which the absolute value intensity of the water signal in the time domain takes an extreme value is defined as the occurrence of a phase jump. Then, a first derivative with respect to the time t is calculated with respect to the phase data of the water signal in the time domain, and the periphery of the phase jump occurrence point is fitted with a model function. At this time, the fitted region is set as a phase skip correction region, and the phase skip is corrected by linearly interpolating the phase value.

Uwe Klose “In Vivo Proton Spectroscopy in Presence of Eddy Currents” Magnetic Resonance in Medicine, Vol. 14, pp. 26-30 (1990) J. et al. M.M. Wild ”Artifacts Introduced by Zero Order Phase Correction in Proton NMR Spectroscopy and a Method of Elimination by Phase Filtering” A. W. Simonetti, et al. “Automated correction of unphased phase jumps in reference signals, whiscorrupt MRSI spectaor after eddy current correction, Vol. 1, Journal of Man, Vol.

  However, in the technique described in Non-Patent Document 2, ringing artifacts cannot be completely removed because a low-frequency component of phase jump remains. Further, since the high-frequency component of the eddy current is cut, a sufficient eddy current correction effect cannot be obtained.

  In the method described in Non-Patent Document 3, when a phase jump occurs even if there is no extreme value in the absolute value intensity of the water signal in the time domain, the phase jump is not corrected. In addition, when there are a plurality of locations where the phase change is steep, it is difficult to identify and extract the location where the phase jumps. Furthermore, since the fitting area is not specified, the fitting accuracy decreases when there are a plurality of phase jumps having different phase change amounts. That is, in order to increase the fitting accuracy, it is necessary to perform fitting for each phase jump occurrence location, and the process becomes complicated and it is difficult to automatically correct the phase jump.

  The present invention has been made in view of the above circumstances, and an object of the present invention is to prevent artifacts due to correction processing and improve the accuracy of correction results when performing processing for correcting the influence of eddy currents in an MRI apparatus.

  In the eddy current correction processing for correcting the influence of eddy currents using the phase data of a substance having a signal intensity higher than that of the metabolite to be measured, the phase data includes the phase change amount in the time domain. The region where the phase skip occurs is specified, and the phase skip correction is performed only for the specified location. The magnitude of the amount of change is determined in each area divided according to a predetermined area division.

  Specifically, a static magnetic field application unit that applies a static magnetic field to the subject, a gradient magnetic field application unit that applies a gradient magnetic field to the subject, and a high-frequency magnetic field pulse irradiation unit that irradiates the subject with a high-frequency magnetic field pulse; A magnetic resonance imaging apparatus comprising: receiving means for receiving a nuclear magnetic resonance signal from the subject; and eddy current correction means for performing eddy current correction on the nuclear magnetic resonance signal, wherein the eddy current correction means comprises: Phase data calculation means for calculating phase data of a nuclear magnetic resonance signal of a substance having a signal intensity greater than that of a metabolite to be measured; phase data dividing means for dividing the phase data into a plurality of small regions in the time direction; A phase jump generation area extracting means for extracting a phase jump occurrence area where a phase jump exists from the small area; and a phase jump correction means for correcting a phase jump in the phase jump generation area. To provide a magnetic resonance imaging apparatus characterized by.

  According to the present invention, when the process of correcting the influence of eddy current in the MRI apparatus is performed, artifacts due to the correction process can be prevented, and the accuracy of the correction result can be improved.

(A)-(c) is an external view of the magnetic resonance imaging device of embodiment of this invention. (A) is a functional block diagram of the magnetic resonance imaging apparatus of embodiment of this invention, (b) is a functional block diagram of the computer with which the magnetic resonance imaging apparatus of embodiment of this invention is provided. It is a figure which shows an example of the MRSI pulse sequence of embodiment of this invention. (A)-(c) is a figure for demonstrating the area | region excited by the MRSI pulse sequence of embodiment of this invention. It is a figure for demonstrating an example of the phase return connection process of embodiment of this invention. It is a flowchart of the phase jump correction process of embodiment of this invention. (A) is a flowchart of the area dividing process of the embodiment of the present invention, and (b) and (c) are diagrams for explaining the area dividing process of the embodiment of the present invention. It is a figure for demonstrating the phase variation calculation process of embodiment of this invention. It is a figure for demonstrating the phase jump correction area | region determination process and phase jump correction process of embodiment of this invention. (A) is the phase data before performing the phase skip correction processing of the embodiment of the present invention and the phase data after correcting one phase skip region. (B) is a figure which shows the phase data before implementing the phase jump correction process of embodiment of this invention, and the phase data after correct | amending all the phase jump areas. (A)-(c) is a figure for demonstrating the effect by embodiment of this invention using a computer simulation result. (A)-(c) is a figure for demonstrating the effect by embodiment of this invention using a phantom experiment result. (A)-(d) is a figure for demonstrating the cause which a ringing artifact produces when eddy current correction is performed using the phase data of a water signal.

  Hereinafter, embodiments to which the present invention is applied will be described. Hereinafter, in all the drawings for explaining the embodiments, the same reference numerals are given to those having the same function, and the repeated explanation thereof is omitted.

  First, the magnetic resonance imaging apparatus (MRI apparatus) of this embodiment will be described. FIG. 1 is an external view of the MRI apparatus of this embodiment. FIG. 1A shows a horizontal magnetic field type MRI apparatus 100 using a tunnel magnet that generates a static magnetic field with a solenoid coil. FIG. 1B shows a hamburger type (open type) vertical magnetic field type MRI apparatus 110 in which magnets are separated into upper and lower sides in order to enhance the feeling of opening. FIG. 1C shows an MRI apparatus 120 that uses the same tunnel-type magnet as in FIG. 1A and has a feeling of openness by shortening the depth of the magnet and tilting it obliquely. In the present embodiment, any of these MRI apparatuses having these appearances can be used. These are merely examples, and the MRI apparatus of the present embodiment is not limited to these forms. In the present embodiment, various known MRI apparatuses can be used regardless of the form and type of the apparatus. Hereinafter, when there is no need to distinguish between them, the MRI apparatus 100 is representative.

  FIG. 2A is a functional configuration diagram of the MRI apparatus 100 of the present embodiment. As shown in the figure, the MRI apparatus 100 of the present embodiment generates a static magnetic field coil 2 that generates a static magnetic field and a gradient magnetic field in each of the x, y, and z directions in the space where the subject 1 is placed. Generated from the subject 1, the shim coil 4 that adjusts the static magnetic field distribution, the measurement high-frequency coil 5 that irradiates the measurement region of the subject 1 with a high-frequency magnetic field (hereinafter simply referred to as a transmission coil) A receiving high-frequency coil 6 that receives a nuclear magnetic resonance signal (hereinafter simply referred to as a receiving coil).

  Various types of static magnetic field coils 2 are adopted according to the structure of each MRI apparatus 100, 110, 120 shown in FIG. The gradient magnetic field coil 3 and the shim coil 4 are driven by a gradient magnetic field power supply unit 12 and a shim power supply unit 13, respectively. In the present embodiment, a case where separate transmission coils 5 and reception coils 6 are used will be described as an example. However, the transmission coil 5 and the reception coil 6 are configured as a single coil. May be. The high frequency magnetic field irradiated by the transmission coil 5 is generated by the transmitter 7. The nuclear magnetic resonance signal detected by the receiving coil 6 is sent to the computer 9 through the receiver 8.

  The sequence control device 14 controls the operations of the gradient magnetic field power supply unit 12 that is a drive power supply for the gradient magnetic field generating coil 3, the shim power supply unit 13 that is the drive power supply for the shim coil 4, the transmitter 7, and the receiver 8. Control the timing of application of gradient magnetic field and high-frequency magnetic field and reception of nuclear magnetic resonance signals. The control time chart is called a pulse sequence, is preset according to measurement, and is stored in a storage device or the like provided in the computer 9 described later.

  The computer 9 performs various arithmetic processes on the received nuclear magnetic resonance signal, generates image information and spectrum information, and controls the overall operation of the MRI apparatus 100. The computer 9 is an information processing device including a CPU, a memory, a storage device, and the like, and a display 10, an external storage device 11, an input device 15, and the like are connected to the computer 9. The display 10 is an interface that displays results obtained by the arithmetic processing to the operator. The input device 15 is an interface for an operator to input conditions, parameters, and the like necessary for arithmetic processing performed in the present embodiment. The external storage device 11 holds data used for various types of arithmetic processing executed by the computer 9, data obtained by the arithmetic processing, input conditions, parameters, and the like together with the storage device included in the computer 9.

  Hereinafter, functions realized by the computer 9 of this embodiment will be described. FIG. 2B is a functional block diagram of the computer 9 of this embodiment. The computer 9 of this embodiment includes a measurement control unit 910, a display information generation unit 920, and an eddy current correction unit 930. The measurement control unit 910 operates the sequence control device 14 according to the pulse sequence and controls each unit to perform measurement. The display information generation unit 920 performs various arithmetic processes on the nuclear magnetic resonance signal obtained by measurement to generate image information and spectrum information.

  The eddy current correction unit 930 performs eddy current correction processing for correcting spectral distortion due to eddy current. In the present embodiment, spectral distortion due to eddy current is corrected by phase data of a water signal having a signal intensity greater than that of a metabolite. In order to realize this, the eddy current correction unit 930 corrects all phase jumps in the phase data by repeating calculation with a phase data calculation unit 940 that calculates the phase data of the water signal from the obtained nuclear magnetic resonance signal. And a phase data correction unit 950. Furthermore, the phase data correction unit 950 of the present embodiment extracts an area where phase jump occurs from the divided area 953 that divides the phase data into a plurality of small areas in the time direction. A phase jump generation area extraction unit 951 and a phase jump correction unit 952 that corrects the phase jump in the small area extracted by the phase jump generation area extraction unit 951.

  The various functions realized by the computer 9 are realized by the CPU loading a program held in the storage device into the memory and executing it. Of the various functions realized by the computer 9, at least one function is realized by an information processing apparatus that is independent of the MRI apparatus 100 and capable of transmitting and receiving data to and from the MRI apparatus 100. It may be.

Hereinafter, the overall flow of measurement of the present embodiment by the above functions will be briefly described. In this embodiment, a water signal is used as the eddy current correction signal. Therefore, first, the measurement control unit 910 controls the sequence control device 14 according to a predetermined pulse sequence, performs non-water suppression measurement, and obtains a water signal F (t). Here, t is a sampling time and is a variable taking a discrete value. From the obtained water signal F (t), the phase data calculation unit 940 calculates phase data φ (t). Then, the phase data correction unit 950 obtains phase data φc (t) in which all phase skips existing in the phase data φ (t) are corrected. Next, the measurement control unit 910 controls the sequence control device 14 according to a predetermined pulse sequence, performs water suppression measurement, and obtains a metabolite signal S (t). The eddy current correction unit 930 corrects the obtained metabolite signal S (t) with the phase data _φc (t) to obtain the metabolite signal S _ecc (t) after the eddy current correction. The display information generation unit 920 performs a Fourier transform on the metabolite signal S _ecc (t) to obtain a spectrum or distribution image of the metabolite.

  Here, an example of a pulse sequence used by the measurement control unit 910 for the measurement will be described. Here, a description will be given by taking as an example a region selective MRSI pulse sequence (hereinafter referred to as an MRSI pulse sequence) for imaging a metabolite.

  FIG. 3 is an example of an MRSI pulse sequence 300. In FIG. 3, RF indicates the application timing of the high frequency magnetic field pulse. Gx, Gy, and Gz indicate application timings of gradient magnetic field pulses in the x, y, and z directions, respectively. A / D indicates a signal measurement period Tp1. The MRSI pulse sequence 300 shown in FIG. 3 is the same as a known MRSI pulse sequence. That is, a predetermined time region of interest is selectively excited using one excitation pulse RF1 and two inversion pulses RF2 and RF3, and an FID signal (free induction decay) FID1 is obtained from this region of interest. Repeat at TR interval.

  The region excited according to this MRSI pulse sequence 300 is shown in FIG. 4A and 4B are scout images for positioning obtained by measurement performed prior to the main measurement. FIG. 4A is a trans image, FIG. 4B is a sagittal image, and FIG. 4C is a coronal image. It is. Hereinafter, the relationship between the operation of each part and the excited region will be described with reference to FIGS.

  First, a high-frequency magnetic field RF1 and gradient magnetic field pulses Gs1 and Gs1 'in the z direction are applied to excite the cross section 401 in the z direction. A high frequency magnetic field RF2 and a gradient magnetic field pulse Gs2 in the y direction are applied after TE / 4 (where TE is an echo time). As a result, only the phase of the nuclear magnetization in the region where the cross section 401 in the z direction intersects the cross section 402 in the y direction is reversed. Subsequently, the high frequency magnetic field RF3 and the gradient magnetic field pulse Gs3 in the x direction are applied after TE / 2 from the application of the high frequency magnetic field RF2. As a result, only the phase of nuclear magnetization in the region of interest 404 where the cross section 401 in the z direction, the cross section 402 in the y direction, and the cross section 403 in the x direction intersect is inverted, and a free induction decay signal FID1 is generated therefrom. This free induction decay signal FID1 is measured. The gradient magnetic field pulses Gd1 to Gd3 and Gd1 ′ to Gd3 ′ in each direction do not disturb the phase of the nuclear magnetization excited by the high-frequency magnetic field RF1, but dephase the phase of the nuclear magnetization excited by RF2 and RF3. It is a gradient magnetic field. Further, the phase encode gradient magnetic fields Gp1 and Gp2 are applied after the high-frequency magnetic field RF3. Thus, the nuclear magnetic resonance signal of the region of interest 404 is obtained.

Next, eddy current correction by the eddy current correction unit 930 will be described. The eddy current correction unit 930 uses the phase signal _φc (t) of the water signal after the phase jump correction from the metabolite signal S (t), and the metabolite signal S _ecc (t) after the eddy current correction, It calculates according to the following formula | equation (1).
S _ecc (t) = S (t) · exp (−i · φc (t)) (1)
Here, i is an imaginary unit.
In the future, exp (−i · φc (t)) will be referred to as a phase correction amount.

First, phase data calculation processing by the phase data calculation unit 940 will be described. The phase data calculation unit 940 calculates the phase data φ (t) of the water signal from the measured water signal F (t) according to the following equation (2).
φ (t) = tan ⁻¹ (Im (F (t))) / (Re (F (t))) (2)
Here, tan ⁻¹ represents an arctangent function, Im (F (t)) represents an imaginary part of the complex number F (t), and Re (F (t)) represents a real part of the complex number F (t).

  Next, a phase data correction process in which all phase skips existing in the phase data φ (t) by the phase data correction unit 950 are corrected by repeated calculation will be described. The phase data correction unit 950 performs a phase return connection process on the phase data φ (t) obtained by the above equation (2) prior to the execution of the phase data correction process. First, the phase loopback connection process will be described with reference to FIG. In FIG. 5, the horizontal axis represents the time (ms) from the start of water signal measurement, and the vertical axis represents the phase value (rad). Further, the broken line indicates the phase data φ (t) calculated from the measurement result, and the solid line indicates the phase data φz (t) after the phase return connection process.

  The phase data φ (t) is calculated as a value between −π and + π. However, a value exceeding the range of −π to + π occurs in the phase of the water signal acquired by the MRSI pulse sequence 300. At the timing when such a value occurs in the phase, it is folded back to obtain a value between −π and + π. As shown in the figure, a discontinuous state occurs in the folded portion. Therefore, phase loop connection processing for removing such a temporal change in the discontinuous phase value is performed to obtain phase data φz (t) that is the original phase change state. In addition, the phase return connection process can use various existing phase return connection processes. At this time, smoothing may be appropriately performed in order to prevent phase variations due to noise.

  Next, the phase data correction process for the phase data φz (t) after the phase loop back connection process by the phase data correction unit 950 will be described. In general, in a portion where a phase jump occurs, a sudden change occurs in the phase value in the vicinity thereof, and an inflection point occurs there. That is, the region where the phase skip occurs includes a point where the primary time differential value of the phase data is convex upward or downward and the secondary time differential value is zero. Hereinafter, the change amount of the primary time differential value of the phase data is simply referred to as “phase change amount”, and the point at which the secondary time differential value is 0 is simply referred to as “inflection point”. Here, even if the phase value is irregularly changed, a location where the amount of phase change is equal to or less than a predetermined value is not caused by the phase jump but for other reasons such as the effect of eddy current. In the present embodiment, the phase data correction unit 950 identifies a region in the phase data φz (t) where a phase jump occurs based on the magnitude of the phase change amount, and is identified as a phase jump occurring. Phase jump correction is performed only within the specified area.

  Specifically, the phase data correction unit 950 divides the phase data φz (t) into a plurality of small areas in the time direction, and calculates a phase change amount within the small areas. Then, the phase jump is corrected in order from the region where the amount of change is large. At this time, every time correction is performed, the influence of the phase correction amount on the value obtained by Fourier transform is evaluated. When the influence on this value is almost eliminated, it is determined that the correction of the portion where the phase jump has occurred is completed.

  In the present embodiment, the absolute value of the difference between the maximum value and the minimum value of the primary differential φz ′ (t) of the phase data φz (t) in the divided small region is calculated as the phase change amount. And

  FIG. 6 is a processing flow of phase jump correction processing of the phase data φz (t) of the present embodiment by the phase data correction unit 950. First, the area dividing unit 953 performs area dividing processing for dividing the phase data φz (t) into a plurality of small areas in the time direction (step S601).

  Next, the phase jump generation area extraction unit 951 calculates a phase change amount for each small area divided in step S601 (step S602). Then, the phase jump generation area extraction unit 951 associates each small area with the calculated phase change amount, and holds it in a storage device or the like (step S603).

  Next, the phase jump generation area extraction unit 951 extracts, as a phase jump generation area, a small area held in association with the largest phase change quantity from among the phase change quantities held in the storage device (step). S604).

  Next, the phase jump correction unit 952 determines a phase jump correction area in the phase jump generation area extracted in step S604 (step S605). Thereafter, the phase skip correction unit 952 corrects the phase skip in the phase skip correction area (step S606).

  Next, the phase jump generation area extraction unit 951 calculates an evaluation value according to a predetermined evaluation function (step S607). The phase jump generation area extraction unit 951 determines whether or not the calculated evaluation value satisfies a predetermined end condition (step S608). If the end condition is not satisfied, it is assumed that the phase jump generation region still remains, and the phase change amount held in the storage device corresponding to the small region corrected in step S606 is set to 0 (step S609). The process returns to S604. On the other hand, when the end condition is satisfied, the phase data correction unit 950 ends the process, and finally obtains the phase data φc (t) after the phase jump correction. Note that the process in step S609 is not limited to the above, and any process may be used as long as the information stored in step S603 is updated so that the processed small area can be excluded from the subsequent processes.

  Here, the details of the area dividing process in step S601 will be described. As described above, the inflection point is included in the region where the phase jump occurs. Therefore, in this embodiment, the area dividing unit 953 calculates an inflection point of the phase data φz (t), divides the divided area so that one inflection point is included, and obtains each small area. FIG. 7 is a diagram for explaining the region division processing of the present embodiment by the region dividing unit 953, (a) is a processing flow, and (b) and (c) are explanatory diagrams of the division method. .

First, the region dividing unit 953 calculates the second derivative φz ″ (t) of the phase data φz (t) of the water signal, and the point where the value becomes 0 is shown in FIG. 7 is extracted as an inflection point 810 of the phase data φz (t) of the signal (step S801) In the graph of Fig. 7B, the horizontal axis represents time (ms) from the start of measurement of the water signal, and the vertical axis. Is a secondary differential value (au), where a case where five inflection points 810 are extracted during the measurement period (t _{0 to} t _F ) is illustrated.

Next, the region dividing unit 953 _identifies the time t _IPn (n is any one of 1, 2, 3, 4, 5) of each inflection point 810 (step S802), and sequentially _{sets the} boundary points 811 in the time direction. Determine (step S803). Here, the region dividing unit 953 picks up each inflection point 810 in order in the time direction, and the first inflection point 810 and the last inflection point 810 are respectively the nearest data end point (t ₀ or t _F ). an intermediate time _{t IPM1, t IPM6} between, in the case of other inflection point 810, inflection point 810 _{(t IPn} and _{t IPn-1)} the intermediate time _{t IPMN} between, each boundary point 811 Time. Then, the region dividing unit 953 changes the time region of the phase data φz (t) of the water signal by the boundary point 811 as shown in FIG. 7C, for example, small regions 821, 822, 823, 824, 825. , 826, 827. Here, in the graph of FIG.7 (c), a horizontal axis is time (ms) from the measurement start of a water signal, and a vertical axis | shaft is a phase value (rad) of phase data.

In the present embodiment, the area dividing unit 953 associates each area divided by the above procedure, for example, each of the small areas 821, 822, 823, 824, 825, 826, and 827 with each end. time boundary point _{811 (t 0, t IPM1)} , (t IPM1, t IPM2), (t IPM2, t IPM3), (t IPM3, t IPM4), (t IPM4, t IPM5), (t IPM5, t _IPM6 ), (t _IPM6 , t _F ) are stored.

  Next, the phase change amount calculation processing for calculating the phase change amount by the phase jump occurrence region extraction unit 951 in step S602 will be described with reference to FIG. FIG. 8 is a graph 831 of the first derivative φz ′ (t) of the phase data φz (t), the horizontal axis is the time t (ms) from the start of measurement of the water signal, and the vertical axis is the first derivative. (Au). In the phase change amount calculation process, the phase jump generation area extraction unit 951 first obtains the first derivative φz ′ (t) of the phase data φz (t), and for each small area obtained by the division in step S601, the corresponding area The absolute value of the difference between the maximum value and the minimum value of the primary differential φz ′ (t) is calculated. Then, the phase jump generation area extraction unit 951 sets the calculated absolute value as the phase change amount.

  Note that the phase change amount calculation processing for calculating the phase change amount is not limited to this. The first derivative φz ′ (t) of the phase data φz (t) is obtained, and the first derivative φz ′ (t) is fitted with a model function such as Gaussian for each small region obtained by the division in step S601. The height of the fitting function may be the phase change amount.

  Next, the correction area determination process for determining the phase jump correction area in step S605 and the phase jump correction process in step S606 by the phase jump correction unit 952 will be described with reference to FIG. FIG. 9 is a first-order differential graph 831 in the vicinity of the small region determined as the phase jump generation region, where the horizontal axis represents time (ms) and the vertical axis represents the first-order differential value (au). Here, as an example, it is assumed that the small region 822 is determined as the phase jump generation region 841.

  In the correction region determination process, the phase skip correction unit 952 calculates the peak half-value width of the primary differential φz ′ (t) in the phase skip generation region 841, and sets the double width as the phase skip correction region 842. Note that the method of determining the phase skip correction area 1002 is not limited to this. Any region may be used as long as the phase change amount in the small region is minimized. For example, when one point is excluded from both sides of the primary differential peak point corresponding to the inflection point, the phase change amount is minimized. You may decide to become. Further, in the phase change amount calculation process as described above, when the phase skip generation region 841 is determined by fitting with a model function such as Gaussian, the phase skip correction region 842 is set to twice the half-value width of the fitting function. It may be configured.

  Further, in the phase skip correction processing, the phase skip correction unit 952 performs linear interpolation that connects both ends of the phase skip correction region 842 with a straight line on the graph of the primary differential φz ′ (t), for example, as shown in FIG. To determine the corrected primary differential φc ′ (t). Then, the phase jump correction unit 952 calculates the phase value φc (t) after the phase jump correction by integrating the corrected first derivative φc ′ (t). Here, the phase difference φc (t) after phase jump correction may be calculated from the corrected primary differential φc ′ (t), not from the integral, but from the sum of powers.

  Here, FIG. 10A shows an example of a graph of phase data before and after the phase skip correction process. A broken line is a graph of the phase data φz (t) of the water signal before phase jump correction, and a solid line is a graph of the water signal phase data φc (t) after correcting the phase jump of the phase jump generation region 841 (small region 822). is there. In this figure, the horizontal axis represents time (ms) and the vertical axis represents phase (rad). As shown in this figure, it can be seen that the phase jump of the water signal is eliminated by the phase jump correction process.

  Note that the method of phase jump correction is not limited to this. For example, the phase skip correction unit 952 interpolates the phase skip correction region r so that the primary differential φ ′ (t) becomes a smooth curve on both sides of the phase skip correction region r, and the corrected primary differential φc ′ ( t) may be determined.

Next, the evaluation process by the phase jump generation area extraction unit 951 in steps S607 and S608 will be described. In the present embodiment, the phase jump generation area extraction unit 951 calculates the following equation (3) with respect to the phase data φ _n (t) after the n-th phase jump correction.
g _n (t) = exp (i · φ _n (t)) (3)
Next, the phase jump generation region extraction unit 951 calculates G _n (f) obtained by Fourier transform of g _n (t) (formula (4)).
G _n (f) = F [gn (t)] (4)
Here, F [] indicates Fourier transform. Phase jump occurs region extraction unit 951, the absolute value of the calculated _G n (f) _| is calculated a maximum peak _{P n} of | G n (f). Then, the phase jump generation region extraction unit 951 evaluates the absolute value | P _n −P _n−1 | of the difference between the maximum peak P _n and the n−1th phase jump corrected maximum peak P _n−1. A function, and ΔP _n obtained by calculating the evaluation function is an evaluation value.

In addition, as an end condition of the evaluation process of the present embodiment, the obtained evaluation value is less than a threshold value. As a threshold value, a value close to 0 is set. The fact that the change in the maximum peak P _n of | G _n (f) | after phase jump correction approaches 0 means that the effect does not change even if a new area is extracted and phase jump correction is performed. Means. This is because the phase change other than the phase jump due to eddy current or noise is not so large as to significantly change the evaluation value. Therefore, in this embodiment, when this end condition is satisfied, it is determined that the phase skip correction has been completed for all the portions where the phase skip has occurred, and the process is ended.

Incidentally, as the evaluation function, | G n _(f) | not the change of each phase jump correction of the maximum peak value of, | G n _(f) | may be used changes every phase jump correction of the standard deviation of. Further, a change for each phase jump correction of the number Nn of signal values f in which | G _n (f) | takes a value equal to or greater than the threshold value T may be used.

  FIG. 10B shows an example of a phase data graph before the phase skip correction process and a phase data graph after correcting all the phase skip regions. A broken line is a graph of the phase data φz (t) of the water signal before the phase jump correction process of the present embodiment. The solid line is a graph of the phase data φc (t) of the water signal after correcting all the phase jumps. In this figure, the horizontal axis represents time (ms) and the vertical axis represents phase (rad). As shown in this figure, it can be seen that all phase jumps of the water signal have been eliminated by the phase jump correction process based on the repeated calculation.

  As described above, according to the present embodiment, the phase data of the eddy current correction signal is divided into a plurality of regions in the time direction, and the region having the largest phase change amount is extracted. The phase skip of the region is corrected, and the evaluation value is calculated based on the evaluation function. This process is repeated, and when the evaluation value approaches 0, the process ends. With this configuration, a phase jump having a plurality of phase changes is automatically determined and removed. Further, eddy current correction is performed using the phase data after removal.

  According to the present embodiment, the phase jump generation region is extracted until the evaluation value approaches 0, and the phase jump correction is repeated. Therefore, the phase jump is performed at a place where the phase data value is irregularly changed. Only the location can be extracted and corrected. Therefore, it is possible to remove the phase jump of the phase data efficiently and accurately. Thereby, in eddy current correction, phase data having no phase jump can be used, and a good spectrum without ringing artifacts can be acquired.

  Here, the effect of the eddy current correction according to the present embodiment is confirmed by computer simulation and phantom experiment.

  The computer simulation results are shown in FIG. Here, FIG. 11A is a metabolite spectrum when eddy current correction is not performed, and FIG. 11B is a case where eddy current correction is performed using the phase data that is not subjected to phase jump correction of this embodiment. (C) shows the metabolite spectrum when the eddy current correction is performed using the phase data subjected to the phase jump correction of the present embodiment.

  As shown in these drawings, it can be seen that if the eddy current correction is performed after the phase jump correction of the present embodiment, the ringing artifact due to the phase jump can be removed and the spectral distortion due to the eddy current can also be corrected.

  Next, FIG. 12 shows the result of actual measurement using a phantom. The phantom used here is an acetylalanine phantom with a concentration of 10 mM. FIG. 12A shows a metabolite spectrum when eddy current correction is not performed, and FIG. 12B shows a metabolite when eddy current correction is performed using phase data that is not subjected to phase jump correction according to this embodiment. Spectrum (c) shows a metabolite spectrum when eddy current correction is performed using the phase data subjected to the phase skip correction of the present embodiment.

  As shown in these figures, even in the result of actual measurement using a phantom, if eddy current correction is performed after phase jump correction of this embodiment, ringing artifacts due to phase jump can be removed, and a spectrum due to eddy current can be removed. It can be seen that distortion can also be corrected.

  As described above, according to the present embodiment, even when the phase data used for correcting the eddy current distortion includes a plurality of locations where the phase change amount varies greatly, such as a large amount of phase change. Since the phase skip correction can be performed by extracting only the phase skip portion that causes the ringing artifact, the phase skip of the phase data can be corrected efficiently. Therefore, in correcting the spectral distortion due to the eddy current, the generation of ringing artifacts can be effectively prevented, and the spectral distortion due to the eddy current can be corrected well.

  For this reason, according to this embodiment, when the process which correct | amends the influence of an eddy current is performed, the artifact by a correction process can be prevented and the precision of a correction result can be improved. For example, in MRSI, a good metabolite spectrum free from ringing artifacts can be obtained.

  In the present embodiment, the phase data used for eddy current correction is obtained from the water signal as an example, but the present invention is not limited to this. The signal is not particularly limited to a water signal as long as it is a signal of a substance having a signal intensity greater than that of a metabolite to be measured.

  In the phase skip correction process, after the region division unit 953 divides the region, the phase jump generation region extraction unit 951 calculates the phase change amount for all the small regions and holds it in the storage device. I can't. For example, the phase skip correction process may be configured to re-divide the region for each phase skip correction and calculate a small region having the largest phase change amount. That is, after step S602, the result is not retained, the phase jump generation region in step S604 is extracted, and the process proceeds with steps S605, S606, S607, and S608. If the condition is not satisfied, the corrected result is obtained in step S606. The process is repeated from step S601 for the phase data.

  Further, in the above embodiment, the evaluation function is used to evaluate the influence of the phase jump correction on the eddy current correction coefficient, and the phase jump generation region is repeatedly extracted. However, the present invention is not limited to this. For example, a threshold value may be set in advance, a region where the phase change amount is equal to or greater than the threshold value may be extracted as a phase skip generation region, and the phase skip correction process may be performed on all the extracted regions.

  In general, when considered in a local time range, the phase jump is a phase change caused by a vector sum of two vectors having different intensities and frequencies. Therefore, the closer the intensity of the two vectors, the more rapidly the phase changes in a short time width in which the two vectors face each other. On the contrary, as the intensity of the two vectors is different, the influence of the vector having a smaller intensity is reduced, so that the phase gradually changes in the opposing time width. Therefore, the magnitude of the phase change amount and the time width (change width) of the phase skip correction region are in an inversely proportional relationship. By using this, the phase skip correction is not performed in order from the region having the largest phase change amount, but the phase skip correction is performed in order from the region having the smaller change width until the influence of the phase jump correction is eliminated. Good. In this case, after the phase loop-back connection process, a smoothing process is performed to reduce the influence of noise.

  Further, in determining the region for phase jump correction, the phase change amount and the above-described change width may be used in combination. For example, when a threshold value is set in advance and the phase change amount is equal to or greater than a predetermined first threshold value and the change width is equal to or less than a predetermined second threshold value, it is determined as a phase jump occurrence region. The phase jump correction may be performed.

1: Subject, 2: Static coil, 3: Gradient coil, 4: Shim coil, 5: Transmitting coil, 6: Receiving coil, 7: Transmitter, 8: Receiver, 9: Computer, 10: Display, 11 : External storage device, 12: power supply unit for gradient magnetic field, 13: power supply unit for shim, 14: sequence control device, 15: input device, 100: MRI device, 110: MRI device, 120: MRI device, 300: MRSI pulse Sequence: 401: Cross section, 402: Cross section, 403: Cross section, 404: Region of interest, 810: Inflection point, 811: Boundary point, 821: Small region, 822: Small region, 823: Small region, 824: Small region, 825: Small region, 826: Small region, 827: Small region, 831: Graph, 841: Phase skip generation region, 842: Phase skip correction region, 910: Measurement control unit, 920: Display information generation , 930: eddy current correction unit, 940: phase data calculator, 950: phase data correcting unit, 951: phase jump occurs area extracting unit, 952: phase jump correction unit, 953: region dividing unit

Claims (18)

From the subject, a static magnetic field applying unit that applies a static magnetic field to the subject, a gradient magnetic field applying unit that applies a gradient magnetic field to the subject, a high-frequency magnetic field pulse irradiating unit that irradiates the subject with a high-frequency magnetic field pulse, and A magnetic resonance imaging apparatus comprising: reception means for receiving a nuclear magnetic resonance signal; and eddy current correction means for performing eddy current correction on the nuclear magnetic resonance signal,
The eddy current correcting means is
Phase data calculating means for calculating phase data of a nuclear magnetic resonance signal of a substance having a signal intensity greater than a metabolite to be measured;
Phase data dividing means for dividing the phase data into a plurality of small regions in the time direction;
A phase jump occurrence region extracting means for extracting a phase jump occurrence region in which phase jump exists from the small region;
A magnetic resonance imaging apparatus comprising: a phase jump correcting unit that corrects a phase jump in the phase jump generation region. The magnetic resonance imaging apparatus according to claim 1,
The magnetic resonance imaging apparatus, wherein the phase jump generation region extracting unit extracts a region where a phase change amount of the phase data satisfies a predetermined condition as the phase jump generation region. The magnetic resonance imaging apparatus according to claim 1 or 2 ,
The phase jump correction unit performs the phase jump correction every time the phase jump generation region is extracted by the phase jump generation region extraction unit,
The phase jump generation region extraction means includes:
Each time the phase jump is corrected by the phase jump correcting means, an evaluation value calculating means for calculating an evaluation value of the phase data after the correction,
Each time the evaluation value is calculated, it is determined whether or not the evaluation value is equal to or greater than a predetermined threshold value. If the evaluation value is equal to or greater than the threshold value, it is determined to continue extraction of the phase jump occurrence region. And determining means for determining to end the extraction, if any,
Wherein while the determination means determines to continue the extraction, magnetic resonance imaging apparatus characterized by extracting the small region 1 in order from the largest small area phase variation of the phase data as the phase jump occurs region. The magnetic resonance imaging apparatus according to any one of claims 1 to 3 ,
The phase data dividing means includes
An inflection point calculating means for calculating an inflection point of the phase data;
2. The magnetic resonance imaging apparatus according to claim 1, wherein each of the small regions is divided so that the inflection point calculated by one inflection point calculating means is included. The magnetic resonance imaging apparatus according to claim 3,
The phase jump generation region extraction means includes:
Primary differential calculation means for calculating a primary differential value of the phase data;
Phase change amount calculating means for calculating an absolute value of a difference between the maximum value and the minimum value of the primary differential value for each of the small regions;
The magnetic resonance imaging apparatus characterized in that the small region having the maximum absolute value is set to the small region having the largest phase change amount. The magnetic resonance imaging apparatus according to claim 3,
The phase jump generation region extraction means includes:
Change amount calculating means for calculating a first derivative value of the phase data;
Fitting means for fitting the first derivative with a model function for each of the small regions;
Height calculation means for calculating the height of the model function after the fitting for each small region,
The magnetic resonance imaging apparatus characterized in that the small region having the highest height is the small region having the largest amount of phase change. The magnetic resonance imaging apparatus according to claim 5,
The phase skip correction means includes:
Phase jump correction area setting means is provided for setting a half-value width of the peak value of the primary differential value in the small area extracted as the phase jump generation area as a phase jump correction area for correcting the phase jump. Magnetic resonance imaging device. The magnetic resonance imaging apparatus according to claim 6,
The phase skip correction means includes:
A magnetic resonance imaging apparatus, comprising: a phase skip correction region setting unit that sets a phase skip correction region that corrects the phase skip as a region twice the half-value width of the model function after the fitting. The magnetic resonance imaging apparatus according to claim 7 or 8 ,
The magnetic resonance imaging apparatus, wherein the phase skip correction unit connects the primary differential values of the phase skip correction region by interpolation to correct the phase skip. The magnetic resonance imaging apparatus according to claim 9,
The magnetic resonance imaging apparatus, wherein the interpolation is linear interpolation in which both ends of the phase skip correction region of the primary differential value are connected by a straight line. The magnetic resonance imaging apparatus according to claim 9 or 10 ,
The magnetic resonance imaging apparatus characterized in that the phase skip correction unit integrates the first-order differential value after connection by the interpolation to obtain phase data after the phase skip correction. The magnetic resonance imaging apparatus according to claim 9 or 10 ,
The magnetic resonance imaging apparatus characterized in that the phase skip correction means uses the sum of powers of the first differential values after connection by the interpolation as phase data after the phase skip correction. The magnetic resonance imaging apparatus according to any one of claims 3 to 12 ,
The evaluation value calculation means uses the difference between the peak height of the correction spectrum calculated from the phase data after correction and the peak height of the correction spectrum calculated from the phase data before correction as the evaluation value. Magnetic resonance imaging device. The magnetic resonance imaging apparatus according to any one of claims 3 to 12 ,
The evaluation value calculating means uses the difference between the standard deviation of the correction spectrum calculated from the phase data after correction and the standard deviation of the correction spectrum calculated from the phase data before correction as the evaluation value,
A magnetic resonance imaging apparatus. The magnetic resonance imaging apparatus according to any one of claims 3 to 12 ,
The evaluation value calculating means is a signal value calculated from the corrected phase data, the number of signal values having a value equal to or greater than a predetermined value, and a signal value calculated from the phase data before correction. The magnetic resonance imaging apparatus is characterized in that a difference from the number of signal values having a value equal to or greater than a predetermined value is used as the evaluation value. From the subject, a static magnetic field applying unit that applies a static magnetic field to the subject, a gradient magnetic field applying unit that applies a gradient magnetic field to the subject, a high-frequency magnetic field pulse irradiating unit that irradiates the subject with a high-frequency magnetic field pulse, and In a magnetic resonance imaging apparatus comprising: a receiving unit that receives a nuclear magnetic resonance signal; and an eddy current correction calculation unit that performs eddy current correction on the nuclear magnetic resonance signal. A phase data correction method for removing all phase jumps occurring in the phase data of the eddy current correction signal which is a nuclear magnetic resonance signal of a substance whose signal intensity is larger than that of a metabolite which is symmetrical to measurement,
A phase data dividing step of dividing the phase data into a plurality of small regions in the time direction;
A phase jump generation region extracting step for extracting a phase jump generation region in which phase jump exists from the small region;
A phase jump correcting step for correcting a phase jump in the extracted region.
A phase data correction method characterized by the above . The phase data correction method according to claim 16, wherein
An evaluation value calculating step for calculating an evaluation value of phase data after correction by the phase skip correction step;
If the evaluation value is equal to or greater than a predetermined threshold value, the method further comprises: a step of extracting the phase jump generation region, a step of correcting the phase jump, and a step of repeating the step of calculating the evaluation value.
A phase data correction method characterized by the above . From the subject, a static magnetic field applying unit that applies a static magnetic field to the subject, a gradient magnetic field applying unit that applies a gradient magnetic field to the subject, a high-frequency magnetic field pulse irradiating unit that irradiates the subject with a high-frequency magnetic field pulse, and A computer of a magnetic resonance imaging apparatus comprising: a receiving unit that receives a nuclear magnetic resonance signal; and an eddy current correction unit that performs eddy current correction on the nuclear magnetic resonance signal.
Among the nuclear magnetic resonance signals acquired by the receiving means, phase data calculation means for calculating phase data of an eddy current correction signal that is a nuclear magnetic resonance signal of a substance having a signal intensity greater than a measurement-symmetric metabolite;
Phase data dividing means for dividing the phase data into a plurality of small regions in the time direction;
A phase jump occurrence region extracting means for extracting a phase jump occurrence region in which phase jump exists from the small region;
A phase jump correcting means for correcting a phase jump in the phase jump generation region;
A program characterized by making it function .

Images