JP3699234B2 - Magnetic resonance imaging device - Google Patents

Description

【００５４】
ここで、
Ｔrmp ：台形状波形の立ち上がり（立ち下がり）時間
Ｔsmp ：エコーデータのサンプリング時間
台形状波形の立ち上がり（立ち下がり）時間Ｔrmp は、台形状波形の瞬時値が０から最大値まで（または最大値から０まで）変化する時間である。エコーデータのサンプリング時間Ｔsmp は、アナログ・ディジタル変換部ＡＤによるエコーデータの１個当たりのサンプリング時間である。
【００５５】
すなわち、（１）式により、台形状波形の立ち上がり（立ち下がり）時間Ｔrmp 内にアナログ・ディジタル変換部ＡＤによってサンプリング可能なデータ数がプリフィル数ＰＦＬとされる。
【００５６】
このため、このプリフィル数ＰＦＬに相当するデータ数をサンプリングする期間をデータウィンドウＤＷＤの前後に付加しても、プロセシングウィンドウＰＷＤは台形状波形の底辺の範囲に収まるようになる。したがって、従来のように、正の台形状波形と負の台形状波形の間に待ち合わせ部を設ける必要がなくなり、与えられたリードアウト勾配について可能な最短値までエコースペースが短縮される。
【００５７】
ディジタルフィルターＤＦＬのタップ数については、プリフィル数に対して一定の関係が決められている。そこで、そのような関係に基づいて、コンピュータＣＯＭにより、例えば次式によってタップ数ＴＡＰが設定される。
【００５８】
【数２】

【００５９】
以上のように、リードアウト勾配の立ち上がり（立ち下がり）時間Ｔrmp を基準とし、その時間内に許容し得るデータサンプリング数をプリフィル数ＰＦＬとし、このプリフィル数ＰＦＬに基づいてタップ数ＴＡＰを決定する。
【００６０】
ところで、このようにした場合、必ずしも十分大きなプリフィル数ＰＦＬおよびタップ数ＴＡＰが得られないことがある。その場合、ディジタルフィルターＤＦＬは、特にカットオフ周波数が比較的低いものであるとき、カットオフ特性がやや悪いものとなる。この影響は、例えば再構成画像上では折り返しの発生等となって現れるが、画像の表示範囲を限定すること等により、画面上に出さないようにすることができるので、実用上問題になることはない。
【００６１】
【発明の効果】
以上詳細に説明したように、本発明では、プリフィル数設定手段により、読み出し勾配磁場強度の瞬時値が０と最大値との間を変化する時間内に収まるデータサンプリング数に基づいてプリフィル数を設定し、このプリフィル数に基づいてタップ数設定手段によりタップ数を設定するようにしたので、読み出し勾配磁場に待ち合わせ部を設ける必要がなくなる。すなわち、短いエコースペースでデータ収集を行う磁気共鳴撮像装置を実現することができる。
【図面の簡単な説明】
【図１】 本発明の実施の形態の一例の装置のブロック図である。
【図２】 本発明の実施の形態の一例の装置が実行するパルスシーケンスを示す図である。
【図３】 スピンエコー法を実行したときのスピンの挙動の概念図である。
【図４】 本発明の実施の形態の一例の装置によって読み出されるエコー信号の概念図である。
【図５】 本発明の実施の形態の一例の装置による２次元フーリエ空間におけるデータ収集の概念図である。
【図６】 本発明の実施の形態の一例の装置が実行するパルスシーケンスを示す図である。
【図７】 本発明の実施の形態の一例の装置におけるリードアウト勾配とディジタルフィルター定数との関係を示す図である。
【図８】 ＥＰＩのパルスシーケンスの一例を示す図である。
【図９】 ディジタルフィルターにおける各定数の関係を示す図である。
【図１０】 従来のＥＰＩにおけるリードアウト勾配とディジタルフィルター定数との関係を示す図である。
【符号の説明】
Ｏ 被検体
Ｍ 静磁場発生部
Ｇ 勾配コイル部
Ｂ ボデイコイル部
ＧＲ 勾配駆動部
ＴＲ 送信部
ＲＶ 受信部
ＡＤ アナログ・ディジタル変換部
ＤＦＬ ディジタルフィルター
ＣＮＴ 制御部
ＣＯＭ コンピュータ
ＤＩＳ 表示部
ＯＰ 操作部
ＤＷＤ データウィンドウ
ＰＷＤ プロセシングウィンドウ
ＰＦＬ プリフィル数
ＴＡＰ タップ数[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a magnetic resonance imaging apparatus, and more particularly to an improvement of a magnetic resonance imaging apparatus that executes an echo planer method.
[0002]
[Prior art]
Magnetic resonance imaging (echo planer imaging: EPI) by the echo planar method is performed by a 90 ° pulse as shown in (a) of FIG. 8, for example, by a pulse sequence as shown in FIG. After the spin in the subject is excited and the spin is reversed by a 180 ° pulse, the readout gradient magnetic field (readout gradient) is exchanged at high speed as shown in FIGS. In addition, a phase encode gradient magnetic field (phase encode gradient) is applied and an echo signal as shown in (a) of the figure is repeatedly generated many times.
[0003]
The echo signals are collected in a memory as digital data, and an image is reconstructed by two-dimensional inverse Fourier (Fourie) transformation of the echo data. Since such EPI collects echo data for one screen by one excitation (one shot), magnetic resonance imaging can be performed at extremely high speed.
[0004]
In collecting the echo data, filtering for removing unnecessary high frequency components is performed using a digital filter such as a FIR (finite impulse response) filter.
[0005]
In this type of digital filter, for example, as shown in FIG. 9 (a), data is divided by a superposition hold method or the like in a data window DWD for sampling a required number of input data. When the sampling L1 × 3 and L2 are performed, it is necessary to overlap the data segment N in the processing window PWD as shown in FIG. The number of data corresponding to this overlap is called the number of taps. The tap number TAP is one of the constants of the digital filter.
[0006]
In addition, before and after the data window DWD, it is necessary to add the number of data determined according to the tap number TAP. This number of data is called the prefil number. The prefill number PFL is also one of the constants of the digital filter. A processing window PWD is obtained by adding a prefill number PFL before and after the data window DWD.
[0007]
The tap number TAP and the prefill number PFL are counted by data sampling number. When the cutoff frequency of the digital filter is relatively low, a large value is required for the tap number TAP and the prefill number PFL in order to realize a good cutoff characteristic.
[0008]
[Problems to be solved by the invention]
The size of the data window DWD corresponds to the length of the continuous period (flat portion) FLT of the maximum value of the read gradient magnetic field as shown in FIG. Then, the processing window PWD in which the prefill number PFL is added before and after that must not be overlapped between subsequent echo data. For this reason, when the number of prefills is increased in order to improve the cutoff characteristic, the waveform of the read gradient magnetic field must have a waiting unit WTE in the middle part of polarity switching as shown in FIG. 10, for example. .
[0009]
The interval of the data window DWD for reading echoes is increased by the amount of the waiting unit WTE. As a result, the interval between the peaks of adjacent echoes, that is, the echo space becomes longer. That is, although the readout gradient magnetic field can be linearly shifted from positive to negative and from negative to positive, it is necessary to provide a queuing unit WTE each time the polarity is switched for the convenience of filtering. It must be longer than the expected echo space.
[0010]
A long echo space in EPI is not preferable because it is easily affected by magnetic field inhomogeneity and distortion or the like appears in the reconstructed image.
The present invention has been made to solve the above problems, and an object thereof is to realize a magnetic resonance imaging apparatus that collects data in a short echo space.
[0011]
[Means for Solving the Problems]
The present invention for solving the above problems is a magnetic resonance imaging apparatus having a digital filter and executing an echo planar method using a readout gradient magnetic field provided by a trapezoidal waveform alternating magnetic field, wherein the readout gradient magnetic field intensity is Prefill number setting means for setting the number of prefills of the digital filter based on the number of data samplings that falls within a time in which the instantaneous value of the signal changes between 0 and the maximum value, and a tap for the digital filter based on the number of prefills And tap number setting means for setting the number.
[0012]
In the present invention, the prefill number setting means sets the prefill number based on the data sampling number that falls within the time when the instantaneous value of the read gradient magnetic field strength changes between 0 and the maximum value, and based on this prefill number. The tap number is set by the tap number setting means. For this reason, it is not necessary to provide a waiting unit for the readout gradient magnetic field. That is, a magnetic resonance imaging apparatus that collects data in a short echo space can be realized.
[0013]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. The present invention is not limited to the embodiment.
[0014]
FIG. 1 shows a block diagram of the magnetic resonance imaging apparatus. This apparatus is an example of an embodiment of the present invention. An example of an embodiment relating to the apparatus of the present invention is shown by the configuration of the apparatus.
[0015]
〔Constitution〕
The configuration of this apparatus will be described. As shown in FIG. 1, in this apparatus, a substantially cylindrical static magnetic field generator M forms a uniform static magnetic field in its internal space. A gradient coil (coil) part G and a body coil (body coil) part B having a substantially cylindrical shape are arranged in the static magnetic field generating part M in common with the central axis. The subject O is carried into a substantially cylindrical space formed inside the body coil part B by a carrying means (not shown).
[0016]
A gradient driving unit GR is connected to the gradient coil unit G. The gradient drive part GR gives a drive signal to the gradient coil part G to generate a gradient magnetic field. Three types of gradient magnetic fields are generated: a slice gradient magnetic field, a read (read out) gradient magnetic field, and a phase encode (phase encode) gradient magnetic field.
[0017]
A transmitter TR is connected to the body coil B. The transmission unit TR supplies a drive signal (RF (radio frequency) signal) to the body coil unit B to generate an RF magnetic field, thereby exciting the spin in the body of the subject O. The magnetic resonance signal generated by the excited spin is detected by the body coil section B. The body coil part B is connected to a receiving part RV. The receiving unit RV is configured to receive the signal detected by the body coil unit B.
[0018]
An analog-to-digital converter AD is connected to the receiver RV. The analog / digital converter AD converts the output signal of the receiver RV into a digital signal. The analog / digital converter AD is connected to the computer COM via the digital filter DFL. The digital filter DFL is an example of an embodiment of a digital filter in the present invention. Specifically, the digital filter DFL is realized as one function of the computer COM, for example.
[0019]
The computer COM inputs a digital signal from the analog / digital conversion unit AD via the digital filter DFL and stores it in a memory (not shown). A data space is formed in the memory. This data space constitutes a two-dimensional Fourier space. The computer COM reconstructs an image of the subject O by performing two-dimensional inverse Fourier transform on the data in the two-dimensional Fourier space.
[0020]
A control unit CNT is connected to the computer COM. A gradient driver GR, a transmitter TR, a receiver RV, and an analog / digital converter AD are connected to the controller CNT. The control unit CNT controls the gradient driving unit GR, the transmission unit TR, the reception unit RV, and the analog / digital conversion unit AD based on a command given from the computer COM.
[0021]
The computer COM sets the number of prefills and the number of taps of the digital filter DFL. The details will be described later. The computer COM is an example of an embodiment of the prefill number setting means and the tap number setting means in the present invention.
[0022]
A display unit DIS and an operation unit OP are connected to the computer COM. The display unit DIS displays various information including a reconstructed image output from the computer COM. The operation unit OP is operated by an operator and inputs various commands and information to the computer COM.
[0023]
[Operation]
The operation of this apparatus will be described. First, EPI will be described, and then the constant setting of the digital filter will be described.
[0024]
(EPI)
FIG. 2 shows an example of a pulse sequence by the echo planar method. This pulse sequence is for collecting spin echoes.
[0025]
In the figure, the horizontal axis represents time and the vertical axis represents signal intensity. (A) RF pulse and echo signal, (b) readout gradient magnetic field (leadout gradient) and dephase gradient magnetic field (dephase gradient), (c) phase encode gradient magnetic field (phase encode gradient) Indicates. Note that the illustration of the slice gradient magnetic field is omitted. The echo signal is much weaker than the RF pulse, but is shown with the same amplitude for convenience. The same applies to other pulse sequences to be described later.
[0026]
The sequence of RF pulses indicates the operation of the transmitter TR. The sequence of the readout gradient and the phase encoding gradient indicates the operation of the gradient driver GR. The same applies to other pulse sequences to be described later.
[0027]
FIG. 3 conceptually shows the spin behavior. In the figure, x ′, y ′ and z ′ indicate three coordinate axes perpendicular to each other in the rotating coordinate system. The operation will be described below with reference to FIGS.
[0028]
As shown in FIG. 2, spin excitation is performed by a 90 ° pulse at time t1. As a result, as shown in FIG. 3A, the spin directed in the z ′ direction is tilted by 90 ° and directed in the y ′ direction.
[0029]
Next, a dephase gradient and a phase encode gradient are applied for a predetermined time at time t2. As a result, the spin phase is dispersed (dephased) as shown in FIG.
[0030]
Next, at time t3, spin inversion is performed by a 180 ° pulse. As a result, as shown in FIG. 3C, the direction of the spin in the y ′ direction is reversed.
Next, a readout gradient is applied at time t4. The readout gradient remains constant until the polarity is reversed at time t6. The spin phase change continues during the readout gradient application period, and the dispersed phase converges as shown in FIG.
[0031]
At time t5 in the middle, the integrated value of the readout gradient becomes equal to the integrated value of the dephase gradient applied at time t2, and the spin phases are aligned as shown in FIG. At this time, the peak of the first magnetic resonance signal (spin echo signal) is generated.
[0032]
After time t5, as shown in (f) of FIG. 3, the phase is dispersed by the continuation of the phase change of the spin, and the echo signal is attenuated.
The echo signal is read out during the readout gradient application period from time t4 to time t6. Reading of the echo signal is performed by the system of body coil unit B-receiving unit RV-analog / digital converting unit AD-digital filter DFL-computer COM. The same applies hereinafter.
[0033]
FIG. 4 shows an echo signal in this period when the time axis is enlarged. However, the echo signal is shown as a conceptual diagram rather than an accurate waveform diagram. As shown in the figure, the echo signal gradually increases in amplitude from time t4 to t5 and reaches a peak, and the amplitude attenuates from time t5 to time t6.
[0034]
Returning to FIG. 2, the phase encode gradient is applied for a short time in accordance with the polarity reversal of the readout gradient at time t6, and thereby the phase encode is advanced by one step. This phase encode gradient is called a blip pulse.
[0035]
The second echo signal is read by the negative readout gradient from time t6 to time t8. The absolute value of the amplitude of the negative readout gradient is the same as that of the positive polarity. The time from time t6 to t8 is equal to the time from time t4 to t6.
[0036]
With this lead-out gradient, the phase of the dispersed spin is pulled back as shown in FIG. As a result, the phase changes in the direction opposite to the arrow in FIG. 3 (f), and the state shown in FIG. 3 (e) is reached, and after that, the state becomes (d). However, the direction of spin phase change is opposite to that of the arrow.
[0037]
For this reason, the echo signal reaches a peak at time t7 and attenuates from that point to time t8. The integrated value of the readout gradient from time t6 to t7 has a relationship that cancels out the integrated value of the readout gradient from time t5 to t6. As a result, an echo signal similar to that shown in FIG. 4 is read out.
[0038]
Similarly, the polarity inversion of the readout gradient and the application of the blip pulse are repeated, and by repeating (d) → (e) → (f) → (e) → (f) → (d) in FIG. A plurality of echo signals are sequentially read out. The echo peak changes along an envelope env as indicated by a broken line, for example, according to the phase encoding amount.
[0039]
Accompanying such readout of the echo signal, collection of echo data proceeds along a predetermined trajectory (trajectory) in the two-dimensional Fourier space. This is shown in FIG. In FIG. 5, k is a two-dimensional Fourier space. This is also called k-space. kx and ky are two coordinate axes orthogonal to each other in the two-dimensional Fourier space k, where kx is a frequency axis (lead-out axis) and ky is a phase axis (phase encode axis).
[0040]
The trajectory trj for collecting echo data starts from, for example, the points kx = −100 and ky = 100. The unit of coordinates is%. kx = -100 corresponds to the left end of the echo signal shown in FIG. ky = 100 is the phase encoding amount of the echo signal shown in FIG. This is determined by the phase encoding gradient at time t2.
[0041]
As the echo signal is read out (collection of echo data) during the readout period from time t4 to t6, the trajectory trj reaches kx = 100 along the arrow. The point of kx = 0 in the middle corresponds to the time t5, and the peak value is collected here.
[0042]
Due to the phase encoding at time t6, the trajectory is lowered by one step. Next, during the readout period from time t6 to t8, data collection is performed for the next echo, and trajectory trj reaches kx = −100 along the arrow. To reach. The point of kx = 0 in the middle corresponds to the time t7, and the peak value is collected here.
[0043]
Similarly, data is collected in the two-dimensional Fourier space k along the kx axis while sequentially transitioning downward along the ky axis at each phase encoding. As a result, echo data for the entire two-dimensional Fourier space k is collected.
[0044]
Based on such echo data, the image is reconstructed by the computer COM. Image reconstruction is performed by two-dimensional inverse Fourier transform.
The above is an example of EPI using spin echo, but EPI using gradient echo can also be performed. Next, the operation will be described. For example, a pulse sequence as shown in FIG. 6 is used.
[0045]
As shown in FIG. 6, spin is excited by an α ° pulse at time t1, a phase encode gradient is applied for a predetermined time at time t2, and a negative dephase gradient is applied at time t3. A lead-out gradient is set at time t4.
[0046]
The readout gradient remains constant until the polarity is reversed at time t6. At time t5, the integrated value of the readout gradient cancels with the integrated value of the dephase gradient from time t3 to t4, and at this time, the peak of the first echo signal (gradient echo signal) occurs. The echo signal attenuates after time t5.
[0047]
The echo signal is read during the period from time t4 to time t6. If the echo signal in this period is shown by enlarging the time axis, it becomes the same as that shown in FIG. That is, as shown in the figure, the echo signal gradually increases in amplitude from time t4 to t5 and reaches a peak, and the amplitude attenuates from time t5 to time t6.
[0048]
Returning to FIG. 6, a blip pulse is applied in accordance with the polarity inversion of the readout gradient at time t6, and the phase encoding is advanced by one step.
The second echo signal is read by the negative readout gradient from time t6 to time t8. The absolute value of the amplitude of the readout gradient is the same as the positive polarity amplitude. The time from time t6 to t8 is equal to the time from time t4 to t6. The echo signal reaches a peak at time t7 and attenuates from that point to time t8. The integrated value of the readout gradient from time t6 to time t7 cancels with the integrated value of the readout gradient from time t5 to time t6. As a result, an echo signal similar to that shown in FIG. 4 is read out.
[0049]
Similarly, the polarity inversion of the readout gradient and the application of the blip pulse are repeated, and a plurality of echo signals are sequentially read out. As a result, an echo signal sequence similar to that shown in FIG. 2 is read out. Therefore, data collection in the two-dimensional Fourier space is performed in the same manner as shown in FIG. 5, and an image is reconstructed by performing two-dimensional inverse Fourier transform on the data collection.
[0050]
(Digital filter constant setting)
Next, constant setting of the digital filter will be described. A readout gradient in which positive and negative alternates at high speed actually has rise and fall times determined by the operating speed of a power amplifier or the like that outputs it. Therefore, the waveform is, for example, an alternating trapezoidal waveform as shown in FIG.
[0051]
As shown in the figure, the continuation period (flat part) FLT of the positive or negative maximum value of this trapezoidal waveform is the echo data collection period, that is, the data window DWD of the digital filter DFL. Then, a processing window PWD of the digital filter DFL is formed by adding a period for sampling the number of data corresponding to the prefill number PFL before and after that.
[0052]
Here, as the prefill number PFL, a value obtained by the following equation is set by the computer COM.
[0053]
[Expression 1]

[0054]
here,
Trmp: Trapezoidal waveform rising (falling) time Tsmp: Echo data sampling time Trapezoidal waveform rising (falling) time Trmp is the trapezoidal waveform instantaneous value from 0 to the maximum value (or from the maximum value to 0) It is time to change. The sampling time Tsmp of the echo data is a sampling time per echo data by the analog / digital conversion unit AD.
[0055]
That is, according to the equation (1), the number of data that can be sampled by the analog / digital conversion unit AD within the rise (fall) time Trmp of the trapezoidal waveform is set as the prefill number PFL.
[0056]
For this reason, even if a period for sampling the number of data corresponding to the prefill number PFL is added before and after the data window DWD, the processing window PWD falls within the range of the bottom of the trapezoidal waveform. Therefore, unlike the prior art, there is no need to provide a waiting portion between the positive trapezoidal waveform and the negative trapezoidal waveform, and the echo space is shortened to the shortest possible value for a given readout gradient.
[0057]
As for the number of taps of the digital filter DFL, a certain relationship is determined with respect to the number of prefills. Therefore, based on such a relationship, the tap number TAP is set by the computer COM, for example, by the following equation.
[0058]
[Expression 2]

[0059]
As described above, based on the rise (falling) time Trmp of the lead-out gradient, the data sampling number allowable within that time is set as the prefill number PFL, and the tap number TAP is determined based on the prefill number PFL.
[0060]
By the way, in this case, a sufficiently large prefill number PFL and tap number TAP may not be obtained. In that case, the cutoff characteristics of the digital filter DFL are somewhat poor, especially when the cutoff frequency is relatively low. This effect appears as, for example, the occurrence of wrapping on the reconstructed image, but it can be prevented from appearing on the screen by limiting the display range of the image and the like, which becomes a practical problem. There is no.
[0061]
【The invention's effect】
As described above in detail, in the present invention, the prefill number setting means sets the prefill number based on the number of data samplings that fall within the time when the instantaneous value of the read gradient magnetic field strength changes between 0 and the maximum value. In addition, since the tap number is set by the tap number setting means based on the prefill number, there is no need to provide a waiting unit for the read gradient magnetic field. That is, a magnetic resonance imaging apparatus that collects data in a short echo space can be realized.
[Brief description of the drawings]
FIG. 1 is a block diagram of an exemplary apparatus according to an embodiment of the present invention.
FIG. 2 is a diagram showing a pulse sequence executed by an apparatus according to an example of an embodiment of the present invention.
FIG. 3 is a conceptual diagram of spin behavior when a spin echo method is executed.
FIG. 4 is a conceptual diagram of an echo signal read out by an apparatus according to an example of an embodiment of the present invention.
FIG. 5 is a conceptual diagram of data collection in a two-dimensional Fourier space by an apparatus according to an example of an embodiment of the present invention.
FIG. 6 is a diagram showing a pulse sequence executed by an apparatus according to an example of an embodiment of the present invention.
FIG. 7 is a diagram showing a relationship between a readout gradient and a digital filter constant in an apparatus according to an example of an embodiment of the present invention.
FIG. 8 is a diagram illustrating an example of an EPI pulse sequence;
FIG. 9 is a diagram illustrating a relationship between constants in the digital filter.
FIG. 10 is a diagram showing a relationship between a readout gradient and a digital filter constant in conventional EPI.
[Explanation of symbols]
O Subject M Static magnetic field generation unit G Gradient coil unit B Body coil unit GR Gradient drive unit TR Transmission unit RV Reception unit AD Analog / digital conversion unit DFL Digital filter CNT Control unit COM Computer DIS Display unit OP Operation unit DWD Data window PWD Processing Window PFL Prefill count TAP Tap count

Claims (1)

A magnetic resonance imaging apparatus having a digital filter and executing an echo planar method using a readout gradient magnetic field given by an alternating magnetic field having a trapezoidal waveform,
Prefill number setting means for setting the number of prefills of the digital filter based on the number of data samplings that falls within a time when the instantaneous value of the readout gradient magnetic field strength changes between 0 and the maximum value;
Tap number setting means for setting the number of taps of the digital filter based on the number of prefills;
A magnetic resonance imaging apparatus comprising:

Images