KR101330945B1 - Epi image correction method and epi image correction system thereof - Google Patents

Abstract

Applying RF pulses with varying bow angles, applying a gradient magnetic field pulse to the target in time, performing ciEPI to obtain segments of the echo train in time in time in k-space, and collecting the echo train Obtaining a navigator of all segments and correcting the size of the echo train using the average energy of the navigator, and reconstructing the missing segment of the k-space using parallel imaging to form an EPI image of the target. An EPI image correction method is disclosed.

Description

EPI IMAGE CORRECTION METHOD AND EPI IMAGE CORRECTION SYSTEM THEREOF

The present invention relates to an EPI image correction method, an EPI image correction system and a recording medium for executing the same.

Echo Planar Imaging (EPI) is one of the most effective Magnetic Resonance Imaging (MRI) technologies. EPI technology enables ultra fast imaging technology by receiving MR images in a single free induction decay (FID) signal for 40 to 100 ms. Thanks to EPI's fast scan technology, defects caused by target movement or problems with MRI images can be reduced. However, since EPI technology generates a large field gradient field and switches at a rate of about 1 kHz, imaging on hardware is not easy and severe distortion of the image is likely to occur due to the magnetic susceptibility of the target. .

Conventional EPI technology includes ss Single Shot Echo Planar Imaging (EPI), which realizes a single scan at a single repetition time, but geometric distortion of an image is easily caused by field unevenness compared to a short scan time. To improve this, an interleaved mutli shot EPI technique has been developed that realizes multiple scans during a single TR. However, the technique has a disadvantage in that the scan time is increased due to the multi-shot scan instead of the geometric distortion of the image.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-described problems of the prior art, and has been developed by ci (consecutive interleaved) EPI technology for realizing continuous scanning to reduce scan time and parallel imaging for improving temporal resolution. By applying parallel imaging technology together and performing RF pulse and amplitude correction with variable flip angle in MRI devices equipped with multi-channel coils, image distortion can be reduced to obtain robust, high-resolution images. We propose an EPI image correction method.

In addition, the present invention provides an EPI image correction system for realizing the EPI image correction method and a recording medium recording the same.

According to an embodiment of the present invention, an EPI image correction method of a magnetic resonance apparatus including a multi-channel coil, an RF pulse having a varying bow angle, which is set in consideration of the total number and reduction factors of segments constituting k-space. And continuously applying the gradient magnetic field pulse to the target; Receiving an echo train occurring at the target with the multi-channel coil and then executing ciEPI to acquire segments of the echo train in time k-space continuously; Acquiring a navigator of each segment before collecting the echo train and correcting the size of the echo train using the average energy value of the navigator; And reconstructing the missing segment of the k-space by using a parallel imaging method to form an EPI image of the target.

The leaning angle a _i is given by Equation 1 for the i-th segment when the total number of segments constituting the k-space is n.

n> 0, i = 0, 1, 2, ..., n-1

The reduction factor is a divisor of the total number of segments, and the number of leaning angles may correspond to the number of quotients divided by the total number of segments.

The reduction factor is a divisor of the total number of segments, and the lean angle may be selected by the number of shares divided by the reduction factor from the largest angle.

In the step of executing the ciEPI, the set including the number of segments equal to the quotient divided by the total number of the segments by the reduction factor, and the number of the set may be the same number as the reduction factor.

Here, a combination of lines may be implemented to form at least two sets that contain the segments discontinuously, and to align the segments of the set to achieve a substantial reduction factor in the k-space.

The set and at least one additionally acquired set may be combined to form a reference k-space.

In the step of correcting the size of the echo train, the correction coefficient λ _i obtained by dividing the average energy value E _na of the navigator by the energy value E _ni of each navigator is an echo train of each segment according to equation (2). It can be applied to (K _i ) to obtain an echo train K'i whose size is corrected.

In forming an EPI image of the target, the parallel imaging method may reconstruct the missing segment using the reference k-space.

In forming the EPI image of the target, the parallel imaging method may use Generalized Auto-calibrating Partially Parallel Acquisitions (GRAPPA) or Sensitivity Encoding (SENSE).

According to another embodiment of the present invention, an RF pulse generator for continuously applying an RF pulse having a varying bow angle, which is set in consideration of the total number and a reduction factor of segments constituting k-space to a target; A gradient magnetic pulse amplifier for continuously applying gradient magnetic pulses to the target in time; A multi-channel coil for continuously receiving an echo train and a navigator received from the target; A processor for performing an EPI correction method including ciEPI and amplitude correction and parallel imaging using the echo train and the navigator and obtaining an image of the target; And a receiver for receiving the echo train and the navigator from the multi-channel coils and delivering them to the processor, wherein the processor acquires the navigator of each segment before collecting the echo train in the k-space; It is possible to provide an EPI image correction system for correcting the size of the echo train by using the average energy value of the navigator, and reconstructing the missing segment of the k-space using a parallel imaging method to form an EPI image of the target. have.

According to another embodiment of the present invention, there is provided a computer-readable recording medium on which a program for implementing the EPI image correction method is recorded.

According to an embodiment of the present invention, the EPI image correction method and the EPI image correction system for performing the same, and the recording medium recording the same, improve the resolution and improve the SNR by reducing the scan time and the geometric distortion of the image. MRI images that are robust to flow or target movement can be obtained.

1 is a pulse sequence diagram (PSD) showing RF pulses and gradient magnetic field pulses, which are used in an EPI image correction method according to an embodiment of the present invention.
2 is a view showing a two-dimensional k-space having various segments in the EPI image correction method according to an embodiment of the present invention.
3 is a view showing an EPI image correction method according to an embodiment of the present invention.
4 is a diagram illustrating a navigator, an echo train, and a method of correcting an amplitude applied to an EPI image correction method according to an embodiment of the present invention.
5 is a block diagram illustrating an amplitude correction algorithm used in an EPI image correction method according to an embodiment of the present invention.
6 is a diagram showing the principle of configuring a reference k-space for parallel imaging used in an EPI image correction method according to an embodiment of the present invention.
7 is a flow chart showing each step of the EPI image correction method according to an embodiment of the present invention.
8 is a schematic view of an EPI image correction system for realizing an EPI image correction method according to an embodiment of the present invention.
9 is a schematic diagram of an example of a kernel of GRAPPA used in the EPI image correction method according to an embodiment of the present invention.

Hereinafter, with reference to the drawings of the present invention will be described in detail a preferred embodiment of the present invention.

1 is a pulse sequence diagram (PSD) showing RF pulses and gradient magnetic field pulses, which are used in an EPI image correction method according to an embodiment of the present invention.

PSD is an RF pulse, magnitude and timing of gradient magnetic field pulses (X, Y, Z Gradient), and data acquisition (Analog Digital Converter) (ADC) in order to effectively obtain MR data. Refer to the diagram showing the timing in order.

The PSD shown in FIG. 1 represents an EPI technique that fills a k-space with Fourier transform data by RF pulses and gradient magnetic field pulses.

In FIG. 1, a continuous interleaved EPI (ciEPI) is shown in which three segments of an interleaved multi-shot EPI are executed continuously in time. In the ciEPI technology, the RF pulse and the gradient magnetic field pulse shown in FIG. 1 are repeatedly executed according to the number of segments, and the time delay between the segments is set to be close to zero to obtain time continuous data. This can reduce the scan time.

Referring to FIG. 1, the RF pulse, also referred to as a slice select pulse, selectively excites protons that process with respect to the main magnetic field B0 in the plane of one of the targets. RF pulses used in the EPI image correction method according to an embodiment of the present invention are applied to the target at varying bow angles (VFA). The VFA will be described later.

Z gradient magnetic field, also known as slice select gradient magnetic field, is used together with the RF pulse to determine the scan position of the target. The phase encoding gradients (Y Gradient) and read out (X Gradient) are given in a direction perpendicular to the Z gradient field and are used to analyze the spatial distribution of magnetic momentum spins. The initial readout gradient field dephases the spin of the magnetic resonance signal excited by the RF pulse. Then, the rephase process of applying the lead out gradient magnetic field of opposite polarity generates the gradient magnetic field echo by inducing an in-phase of the phased spindle.

By repeating the above process, an echo train of magnetic resonance signals received from the target can be generated. The intensity of the magnetic resonance signal is attenuated to zero as a result of spin-spin relaxation (T2) relaxation time or field inhomogeneity and magnetic susceptibility (T2 *) relaxation time effects. The loss of magnetic resonance signals due to T2 and T2 * can be minimized by reducing the data acquisition time.

EPI image correction according to an embodiment of the present invention can minimize the filtering effect of the image due to T2 * attenuation by adjusting the length of the echo train based on the interleaved multi-shot EPI technique.

The signal received from the Analog Digital Converter (ADC) consists of a navigator and an echo train that fills the k-space. The navigator does not directly construct an image, but is usually used to perform amplitude correction to remove artifacts.

2 is a view showing a two-dimensional k-space having various segments in the EPI image correction method according to an embodiment of the present invention. In the case of the three-dimensional k-space, the three-dimensional space (x, y, z) corresponding to the spatial coordinates is a frequency space (kx, ky, kz) obtained by Fourier transform (FT). Fourier transform the k-space to obtain a spatial distribution of the spin, or image. The NRM signal moves the k-space in proportion to the integral of the gradient field over time.

Referring to the two-dimensional k-space of FIG. 2, (a) shows a conventional single shot EPI and shows that the k-space is filled while one arrow trajectory is zigzag.

2 (b)-(f) show the multi-shot EPI technique, showing that the k-spaces are filled with 2, 3, 4, 6 and 8 segments in turn. Multi-shot EPI technology fills one k-space by multiple attempts to apply RF and gradient magnetic pulses to the target in the same pattern.

In FIG. 2, each segment is represented by an arrow trajectory. Whether to select a k-space filled with several segments may be determined in consideration of factors such as scan time, image acquisition time, SNR, and resolution.

The main purpose of the conventional ssEPI is to realize fast 2D imaging using single RF excitation. The gradient magnetic field pulse should be designed to include all points of kx, ky in k-space. However, ssEPI has the disadvantage of being sensitive to echo time (TE) and artifacts. In addition, ssEPI technology is difficult to achieve in a strong magnetic field (eg 5T or more) because the magnetic susceptibility effect seriously distorts the image or causes loss of the magnetic resonance signal.

In the ciEPI technique, the multi-shot EPI technique shown in FIGS. 1 and 2 is executed continuously in time. As shown in FIG. 2, the complete k-space employs ciEPI technology using several trajectories, and the remaining trajectories (segments) are estimated and filled using parallel imaging to complement the ciEPI technology. Can form spaces

3 is a view showing an EPI image correction method according to an embodiment of the present invention. In the embodiment shown in FIG. 3, in the magnetic resonance apparatus equipped with four multi-channel coils, when the total number of segments of k-space is 6 and the reduction factor is set to 2, the EPI image correction according to the embodiment of the present invention. It shows the algorithm of law.

First, in the magnetic resonance apparatus applied with the main magnetic field, the changing bow angle (a) is set in consideration of the total number of segments k and the reduction factor r in the k-space, and the RF pulse is adjusted according to the bow angle (a). To the target. RF pulses with a constant bow angle cannot achieve steady-state magnetization and exhibit low SNR. Therefore, a variable magnetization angle (VFA) is employed to provide a stable magnetization state for continuous RF pulses. SNR can be maximized.

The leaning angle ai is given by Equation 3 for the i th segment when the total number of segments of the k-space is n.

n> 0, i = 1, 2, ... n-1

Therefore, as shown in FIG. 3, when n = 6, the leaning angle FA is given as shown in the following table.

Segment total 1st FA 2nd FA 3rd FA 4th FA 5th FA 6th FA 6 24.1 ° 26.6 ° 30.0 ° 35.3 ° 45.0 ° 90.0 °

The changing bowing angle may be given according to the total number of segments and the order (i) of the segments. In the EPI image correction method according to an embodiment of the present invention, only a partial bowing angle is selectively considered in consideration of the decreasing factor of the changing bowing angles. use. The reduction factor r is a divisor of the segment total number n, and the number of leaning angles selected corresponds to the number of quotients s divided by the total number n of segments by the reduction factor r. Therefore, in the example shown in FIG. 3, when the total number of segments n = 6 and the reduction factor r = 2 are set, the number of bow angles applied to the RF pulse is quotient s = 6/2 = 3.

Here, the third to six bow angles select the fourth to six bow angles, namely 35.3 °, 45.0 ° and 90.0 °. This takes into account SNR, and the bow angle ai selects the total number n of the segments from the largest angle by the number of quotients s divided by the reduction factor r.

Therefore, when the total number of segments is n = 6, the bow angle that can be selected according to the reduction factor (r = 1, 2, 3) is shown in the following table.

Segment total 1 ^st FA 2 ^nd FA 3 ^rd FA 4 ^th FA 5 ^th FA 6 ^th FA 6 (r = 1) 24.1 ° 26.6 ° 30.0 ° 35.3 ° 45.0 ° 90.0 ° 6 (r = 2) 35.3 ° 45.0 ° 90.0 ° 6 (r = 3) 45.0 ° 90.0 °

Referring to FIG. 3, the k-space of a segment obtained from an RF pulse to which a bow angle of 35.3 ° is applied is respectively determined according to each of four multichannel coils of I11, I12, and I13 when four multichannel coils are mounted on a magnetic resonance scanner. May be named I14. The k-space of the segment obtained from the RF pulse with 45.0 ° of lean angle can be given by I21, I22, I23 and I24, respectively, according to each of the four multi-channel coils. The k-space of the segment obtained from the RF pulse to which the tilt angle of 90.0 ° is applied may be set to I31, I32, I33, and I34 according to each of the four multi-channel coils. The scan direction of the k-space progresses from bottom to top in the bottom up direction as shown by the arrow trajectory shown in FIG. 3.

Once RF pulses and gradient magnetic field pulses are applied to the target, an echo train generated at the target is received by the multi-channel coil, and then ciEPI is executed to obtain segments corresponding to each bow angle in time in k-space.

In the EPI image correction method according to an embodiment of the present invention, the ciEPI technique aligns sets so that parallel imaging techniques can be applied together. The set is provided with the same number r of reduction factors as each set comprising the number of segments corresponding to the quotient s, depending on the reduction factor r. The segments included in each set are selected discontinuously instead of continuously selecting the paths formed side by side.

Referring to FIG. 3, the number of sets of segments in the k-space of the EPI image correction method according to an embodiment of the present invention is two. The scheme of segments may be set to set 1 comprising 1, 3, 5 segments and set 2 comprising 2, 4, 6 segments. Here, each segment is selected discontinuously such as 1, 3, 5 segments and 2, 4, 6 segments to achieve a constant reduction factor.

Thus, in FIG. 3, the three segments obtained first in time succession by applying the bow angles (35.3 °, 45.0 ° and 90.0 °) become 2, 4 and 6 segments of set 2, and again three additionally obtained for reference. The segment is 1, 3, 5 segments of set 1. Initially, a set is selected that includes segments that cross the centerline (Gy = 0) of Set 1 and Set 2.

As each set is prepared, it executes a combination of lines in k-space to form a complete k-space. In FIG. 3, each of the echo trains received in the multi-channel coil is composed of three k-spaces separately in the first k-space, with segments having a reduction factor of 6, but with a line combination of k-spaces with each segment. This results in the formation of a k-space with a substantial reduction factor of two. That is, the segments included in each set are selected discontinuously, and segments of k-space can be aligned by line combinations between sets, thereby realizing a substantial reduction factor.

However, varying tilt angles, applied to reduce motion-related artifacts, cause amplitude distortion between segments and ghost artifacts due to their imperfections and uneven magnetic fields. Therefore, in order to solve this problem, the EPI image correction method according to an embodiment of the present invention performs a step of correcting the magnitude of the amplitude of the magnetic resonance signal. To this end, the navigator of each segment is obtained before the echo train is obtained, and the size of the echo train is corrected using the average energy value of the navigator.

4 is a diagram illustrating a navigator, an echo train, and a method of correcting an amplitude applied to an EPI image correction method according to an embodiment of the present invention. FIG. 4 (a) shows a navigator collecting segments by 1 ^st seg., 2 ^nd seg., 3 ^rd seg. Corresponding to each bow angle (35.3 °, 45 °, 90 °) before each echo train. Seems to 4 (b) shows an echo train whose amplitude is corrected by deriving a correction coefficient using the average energy value of the navigator and then applying it to the echo train.

5 is a block diagram illustrating an amplitude correction algorithm used in an EPI image correction method according to an embodiment of the present invention.

Referring to FIG. 5, the energy average value E _na of the energy E _ni of the navigator N _i of the i-th segment is derived using Equation 4.

The correction coefficient λ _i is given by the value of equation (5).

The echo train K'i whose amplitude is obtained by applying the correction coefficient λ _i to the echo train Ki is given by Equation 6.

This amplitude correction can reduce image aliasing and significantly reduce artifacts.

6 is a diagram showing the principle of configuring a reference k-space for parallel imaging used in an EPI image correction method according to an embodiment of the present invention. Parallel imaging technology is introduced to reduce the long imaging time of ciEPI technology.

FIG. 6A shows a k-space having a segment of 6 and a decreasing factor r of 1. FIG. 6 (b) and 6 (c) show examples of k-spaces having two types of sets consisting of three substantial segments when the substantial reduction factor r is two.

Comparing (a) and (b) of FIG. 6, it can be seen that a missing segment occurs in FIG. 6 (b). If the k-space acquired with ciEPI technology becomes (b) of FIG. 6, the missing segment causes aliasing to the image made with the k-space of (b). In order to estimate missing segments with parallel imaging or to eliminate aliasing of the image, a reference k-space such as (a) must be collected at least once. To do this, at least one additionally obtain a k-space as in (c), and combine the previously obtained (b) with the additionally obtained (c) so that the segments are aligned continuously so that the reference with a reduction factor of (a) is 1 Construct a k-space.

Since k-space has the property of spatial harmonics, data of some segments can be used to derive data of other missing segments. Thus, parallel imaging techniques can reduce the time for data acquisition and at the same time estimate the data of missing segments through the reference k-space to obtain a complete k-space.

The method of estimating missing data using the GRAPPA method in the parallel imaging method is as follows. The value of any missing point on the k-space is estimated by linear combination of the values of the points around it actually obtained. Since the k-space has a reduction factor in the phase encoding direction, in order to estimate each point of the missing phase encoding line, a linear combination of the values of the points in the phase encoding direction of the corresponding points and the points in the readout direction is used. .

In the GRAPPA technique, 20 points included in four rectangles in the phase encoding direction and five rectangles in the readout direction and points at the same position of k-space obtained from each of the multi-channels at the same time are used. If there are four multi-channels as shown in Fig. 3, a calculation of a linear combination of a total of 80 (= 4x5x4) acquired points is used to estimate one missing point. The relationship between the missing point and the point for estimating it is called the kernel of the GRAPPA technique. The size of the kernel affects performance, data processing time, and so on.

The kernel of GRAPPA is given by Equation 7 in a linear combination. Sj (x, y) is the value of the point located at (x, y) of the k-space obtained from the jth coil, R is the reduction factor, and r is the location of the missing data. 1 indicates a coil, b indicates a point in the phase encoding direction, and h indicates a point in the readout direction. ky and kx are the minimum distances in the phase encoding direction or the readout direction between a point and the next point in k-space.

9 is a schematic diagram of an example of a kernel of GRAPPA used in the EPI image correction method according to an embodiment of the present invention. Referring to FIG. 9, linear coefficients of a kernel or an interpolation net are calculated through an auto-calibration signal (ACS) using information of an acquired line obtained from each coil 1, 2, and 3. You can see the extraction process. The missing line of another part may be obtained using the linear coefficient of the kernel obtained through the above process.

The linear coefficients (Wj, r) shown in Equation 7 are linear coefficients for linear combinations in the GRAPPA technique, and can be obtained through reference k-spaces. The reference k-space will serve as the ACS mentioned in the paragraph above. Here, several linear equations can be derived using samples of reference k-space as a sample according to a predetermined kernel. As in the above example, if 80 points are needed to estimate one missing point, 80 linear coefficients are also required. Since we know the values for all points in the reference k-space, we can create an equation for 80 unknowns (linear coefficients). At least 80 data samples can be generated in the reference k-space, and 80 linear coefficients of the linear equation can be set by the least squares method through a pseudo inverse matrix. The missing point can be estimated using the linear coefficient.

Referring again to FIG. 3, a complete k-space is formed using parallel imaging techniques and transformed using an inverse Fourier Transform to obtain an image of a desired target. Here, k-space-based GRAPPA is selected as the parallel imaging technology, but image-based technology such as SENSE (SENSitivity Encoding) can be used. That is, the parallel imaging technique is not limited to a specific technique, and of course, various techniques may be used.

7 is a flowchart showing each step of the EPI image correction method according to an embodiment of the present invention described above.

Referring to FIG. 7, an EPI image correction method of a magnetic resonance apparatus including a multi-channel coil may first include an RF pulse having a varying bow angle set in consideration of the total number of k-space segments and a reduction factor, and a gradient magnetic field pulse. To the target (step 101).

Then, the echo train generated at the target is received by the multi-channel coil, and then ciEPI is performed to obtain a segment corresponding to each bow angle in time in the k-space (step 103). In step 102, a navigator of each segment is acquired before collecting the echo train. The navigator is obtained without the application of the phase encoding gradient magnetic field.

The navigator may correct the size of the echo train using the average energy value (step 103). In the k-space after the size of the echo train is corrected, the number of missing segments reduced by the reduction factor is reconstructed using parallel imaging to form a complete k-space (step 105). If the k-space is inverse Fourier transformed, an EPI image of the target may be obtained (step 107).

8 is a schematic view of an EPI image correction system for realizing an EPI image correction method according to an embodiment of the present invention.

Referring to FIG. 8, an EPI image correction system 10 for implementing an EPI image correction method according to an embodiment of the present invention may include a multi-channel coil 11 installed in a bore 12 to which a main magnetic field is applied. In addition, an RF pulse generator 13 for applying an RF pulse to a target (not shown) inside the bore 12 and an gradient magnetic field pulse generator 15 for applying an gradient magnetic pulse are connected to the bore 12. In addition, a receiver 17 for receiving a navigator and an echo train emitted from a target and an ADC 19 for analog-to-digital converting a signal collected by the receiver 17 are connected to the receiver 17.

The processor 21 connected to the receiver 17 executes the EPI image correction method for realizing the EPI image correction method according to an embodiment of the present invention. Here, since the EPI image correction method is the same as the above-described content, a detailed description thereof will be omitted. The processor 21 also sends an instruction to the RF pulse generator 13 to generate an RF pulse having a varying bow angle.

An EPI image correction system for realizing an EPI image correction method according to an embodiment of the present invention can achieve a short scan time, an imaging time, a high SNR, and provide a high resolution image.

Since the above-described embodiments of the present invention can be implemented as a program that can be executed in a computer, it is possible to provide a computer-readable recording medium in which a program for implementing an EPI correction method according to an embodiment of the present invention is recorded.

The computer-readable recording media may be read-only memory (ROM), floppy disks, hard disks, magnetic storage media such as magnetic tape, optical read media such as CD-ROM, DVD, optical data storage devices, and transmission through the Internet. It includes a storage medium, such as a carrier wave, such as.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, but, on the contrary, Modification is possible. Accordingly, the spirit of the present invention should be understood only in accordance with the following claims, and all equivalents or equivalent variations thereof are included in the scope of the present invention.

11 multichannel coils
12 bore
13 RF pulse generator
15 gradient magnetic field pulse generator
17 receiver
19 ADC

Claims (23)

An EPI image correction method of a magnetic resonance device including a multi-channel coil,
continuously applying RF pulses having varying bow angles and gradient magnetic field pulses to the target, which are set in consideration of the total number of segments constituting the k-space and the reduction factor, wherein the reduction factor is a divisor of the total number of segments And the number of bow angles corresponds to the number of shares divided by the total number of segments by the reduction factor;
Receiving an echo train occurring at the target with the multi-channel coil and then executing ciEPI to acquire segments of the echo train in time k-space continuously;
Acquiring a navigator of each segment before collecting the echo train and correcting the size of the echo train using the average energy value of the navigator; And
Reconstructing the missing segment of the k-space using parallel imaging to form an EPI image of the target;
EPI image correction method comprising a. The method of claim 1,
The bow angle (a _i ) is given by Equation 8 for the i-th segment when the total number of segments constituting the k-space is n.
(8)

(n> 0, i = 0, 1, 2, ..., n-1) delete The method of claim 1,
And the bow angle is selected by the number of shares divided by the reduction factor from the largest angle to the total number of segments. The method of claim 1,
In the step of executing the ciEPI,
Align a set comprising the same number of segments as the quotient of the total number of segments divided by the reduction factor, and the number of sets is the same number as the reduction factor. The method of claim 5,
Forming at least two sets that contain the segments discontinuously, and performing a line combination to align the segments of the set to achieve a substantial reduction factor in the k-space. The method according to claim 6,
And combining said set with at least one additionally acquired set to form a reference k-space. The method of claim 1,
In the step of correcting the size of the echo train,
The size is corrected by applying the correction coefficient (λ _i ) obtained by dividing the average energy value (E _na ) of the navigator by the energy value (E _ni ) of each navigator to the echo train (K _i ) of each segment according to Equation (9). EPI image correction to obtain an echo train (K'i).
(9)

The method of claim 7, wherein
In the step of correcting the size of the echo train,
The size is corrected by applying the correction coefficient λ _i obtained by dividing the average energy value E _{na of the} navigator by the energy value E _ni of each navigator to the echo train K _i of each segment according to Equation 10. EPI image correction to obtain an echo train (K'i).
(Equation 10)

10. The method of claim 9,
In forming the EPI image of the target,
Wherein the parallel imaging method reconstructs the missing segment using the reference k-space. The method of claim 1,
In forming the EPI image of the target,
The parallel imaging method is an EPI image correction method, which is a GRAPPA technique or a SENSE technique. RF pulse generator for continuously applying an RF pulse having a varying bow angle, which is set in consideration of the total number of segments constituting the k-space and a reduction factor to the target, wherein the reduction factor is a divisor of the total number of segments. The number of bow angles corresponds to the number of shares divided by the total number of segments by the reduction factor;
A gradient magnetic pulse amplifier for continuously applying gradient magnetic pulses to the target in time;
A multi-channel coil for continuously receiving an echo train and a navigator received from the target;
A processor for performing an EPI correction method including ciEPI and amplitude correction and parallel imaging using the echo train and the navigator and obtaining an image of the target; And
A receiver for receiving said echo train and said navigator from said multi-channel coil and delivering it to said processor,
The processor acquires the navigator of each segment before collecting the echo train, corrects the size of the echo train using the average energy value of the navigator, and uses the parallel imaging method for the missing segments of the k-space. EPI image correction system for restoring to form an EPI image of the target. The method of claim 12,
The bow angle (a _i ) is given by Equation 11 for the i-th segment when the total number of segments constituting the k-space is n.
(Equation 11)

(n> 0, i = 0, 1, 2, ..., n-1) delete The method of claim 12,
The bow angle is selected by the number of shares divided by the reduction factor from the largest angle to the total number of segments. The method of claim 12,
The processor sorts a set comprising a number of segments equal to the quotient of the total number of segments divided by the reduction factor in the k-space, wherein the number of sets is the same number as the reduction factor. 17. The method of claim 16,
EPI image correction system for forming at least two sets that contain the segments discontinuously, and performing line combinations to align the segments of the set to achieve a substantial reduction factor in the k-space. 18. The method of claim 17,
And combining said set with at least one additionally acquired set to form a reference k-space. The method of claim 12,
The processor applies the correction coefficient λ _i obtained by dividing the average energy value E _{na of the} navigator by the energy value E _ni of each navigator to the echo train K _i of each segment according to Equation 12. EPI image correction system to obtain a scaled echo train (K'i).
(Equation 12)

19. The method of claim 18,
The processor applies a correction coefficient λ _i obtained by dividing the average energy value E _{na of the} navigator by the energy value E _ni of each navigator to the echo train K _i of each segment according to Equation 13. EPI image correction system to obtain a scaled echo train (K'i).
(Equation 13)

21. The method of claim 20,
And the processor recovers missing segments using the reference k-space in the parallel imaging method. The method of claim 12,
The parallel imaging method is an EPI image correction system, which is a GRAPPA technique or a SENSE technique. A computer-readable recording medium having recorded thereon a program for implementing the EPI image correction method according to any one of claims 1, 2 and 4.

Images