JPH05207989A - Mr imaging method - Google Patents

Abstract

PURPOSE:To improve the time resolution of an image without extending an inspection time of the whole. CONSTITUTION:Repeated data collection performed while varying the phase encoding amount of a high speed field echo sequence of a short TR is repeated continuously at each combination set of its phase encoding amount, an image is reconstituted by a data set corresponding to each phase encoding amount set, and also, the image is reconstituted by fetching the data of one set from the adjacent data of two sets.

Description

Detailed Description of the Invention

[0001]

This invention relates to nuclear magnetic resonance (NM).
The present invention relates to an MR imaging method for performing imaging using R), and particularly to a cine imaging method using a high speed field echo method.

[0002]

2. Description of the Related Art Conventionally, a cine imaging method using a segmented high-speed field echo method has been reported as a method for speeding up the cine imaging method in MR imaging (Dennis J. Atkinson, MSRobert, R. Edelman,
MD Radiology 1991; 178: 357-360). In this method, a very short TR high-speed field echo sequence is repeated while changing the amount of phase encoding, thereby collecting data for a plurality of lines within the time of one temporal phase. When performing cine imaging of the heart region, data for a plurality of lines is obtained within each time phase of one heartbeat period, and data for another plurality of lines is obtained for each time phase of the next one heartbeat period. , It is repeated for each heartbeat period, and if the data for a given line for each time phase is available, the image of each time phase can be reconstructed. It can be reduced to one-half, and imaging can be completed within the patient's breath-hold period.

[0003]

However, the conventional cine imaging method using the high-speed field echo sequence has a problem that the time resolution of the image is insufficient.

For example, when performing cine imaging of a heart region, there is a restriction that the imaging must be completed within a time period when the patient can hold his / her breath, and there is a circumstance that the number of heartbeat periods for imaging cannot be increased. .. Under such circumstances, it is not possible to shorten the time for collecting the data of each time phase in relation to the shortest possible TR of the MR imaging apparatus. That is, many time phase images cannot be obtained within one heartbeat period, and the time resolution of the image becomes insufficient.

Generally, the breath-holding time is said to be about 16 heartbeats, and when an image of a 128 × 128 matrix is obtained, 16 times are obtained for each time phase during one heartbeat.
It is necessary to collect line data. This means that 16 times the shortest TR is required for imaging in each time phase, and if so, the time resolution cannot be improved so much.

In order to improve this, the same cine imaging is performed after a predetermined delay time is provided from the ECG trigger,
It is conceivable to collect data at time phases located between the time phases in cine imaging without delay. According to this, since it is possible to obtain an image at an intermediate time phase of images of each time phase obtained by cine imaging without delay, time resolution can be doubled, but cine imaging without delay and delay can be performed. Since the cine imaging provided must be performed in each of the two breath-holding periods, there is a problem that the entire examination time becomes long.

In view of the above, the present invention has improved the conventional cine imaging method using the high-speed field echo method to improve the time resolution of an image without extending the entire examination time. The purpose is to provide.

[0008]

To achieve the above object, the MR imaging method according to the present invention uses a short TR.
The high-speed field echo sequence of (1) is repeatedly acquired while changing the phase encoding amount, and the data acquisition is continuously repeated for each combination set of the phase encoding amounts. Then, a data set necessary for image reconstruction is obtained for each phase encoding amount set, and one image is reconstructed using the data set. At the same time, for each adjacent two-phase encoding amount set, a part of the data is extracted from both of them, and a data set corresponding to one set of the combination of the above-mentioned phase encoding amounts, which is located in the middle, is constructed. One image is reconstructed from each of these data sets. Since each of the latter images has a time phase located in the middle of the time phase of the former image, the time resolution of the image is improved. Further, since the cine imaging need only be performed once, and it is not necessary to perform the cine imaging twice, there is no extension of the entire inspection time.

[0009]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described in detail below with reference to the drawings. FIG. 1 shows the entire sequence of an embodiment of the present invention. In this embodiment, each time phase (first time phase, second time phase, third time phase, fourth time phase, ...) Between the R wave and the R wave of the electrocardiographic waveform shown in FIG. 1A. Collect data for each. The field echo sequence as shown in FIG. 2 is repeated eight times for each time phase.

The field echo sequence shown in FIG. 2 is of a high-speed type having a short TR, and when an RF pulse is applied as shown in A of FIG. 2, Gz for slice selective excitation is simultaneously shown as shown in B of FIG. A pulse (gradient magnetic field whose magnetic field strength is inclined in the Z direction) is applied to selectively excite only a slice existing at a certain position in the Z direction. Next, a Gy (gradient magnetic field whose magnetic field strength is inclined in the Y direction) pulse for phase encoding is added, and a Gx (gradient magnetic field whose magnetic field strength is inclined in the X direction) pulse for reading and frequency encoding is added. Then, the Gx pulse is inverted to generate an echo signal.

The matrix of the reconstructed image is 128 × 128.
Then, in order to reconstruct this image, the echo signal generated by one excitation is sampled at 128 sampling points to obtain one line of data, and this is changed 128 times by changing the phase encode amount in 128 ways. It is necessary to repeatedly collect data for 128 lines. Phase encode amount is 1 from minus side to plus side
Since it can be changed in 28 ways, numbers # 1 to # 128 are given to each of the phase encoding amounts. By arranging the 128 lines of data thus obtained as the raw data array P of FIG. 3 and performing a two-dimensional Fourier transform, the image of the slice selectively excited can be reconstructed.

Here, the field echo sequence is repeated 8 times for each time phase in FIG. That is, as shown in FIGS. 1B and 3, the phase encoding amounts are # 1, # 17, # 33, # 49, # 65, # 81, and # 9 in each time phase after the first R wave.
7 and # 113. Then, in each time phase after the second R wave, the phase encoding amount at the time of the first R wave is shifted by 1 each, and then # 2, # 18, # 34, # 50, # 66, # 82, # 9.
8 and # 114. In this way, if the phase encode amount is shifted by 1 for each R wave, the phase encode amount for each time phase in the 16th R wave is # 16, # 32, # 48, #.
64, # 80, # 96, # 112, # 128,
For each time phase from the 1st R wave to the 16th R wave,
128 with 128 different phase encoding amounts of # 128
This means that the field echo sequence has been performed twice.

As a result, 128 lines of data can be collected for each time phase. By using the 128 lines of data collected for each time phase, the images in each of the first time phase, the second time phase, the third time phase, the fourth time phase, ... Can be reconstructed. ..

In addition to such image reconstruction, the following image reconstruction is performed. First, from the data obtained in the fifth field echo sequence in the latter half of the first time phase to the data obtained in the eighth field echo sequence, the first field echo sequence in the first half of the second time phase is extracted. From the obtained data to the data obtained by the fourth field echo sequence are taken out. Second
Similarly, data is extracted from the time phase and the third time phase. Similarly, data is similarly extracted for all two adjacent time phases.

The combination of the phase encoding amounts in the extracted data is the same as the data for each time phase, although the order is different. Therefore, when the first R wave to the 16th R wave are collected, the phase encode amounts of # 1 to # 128 are all aligned. Therefore, if they are rearranged and arranged in the order of the phase encode amount, raw data similar to P of FIG. It becomes an array and the image can be reconstructed. Each of the images reconstructed in this way has a first 'time phase, a second' time phase, a third 'time phase, ... Which are located in the middle of the respective first, second, third, ... Time phases. It becomes an image. That is, the intermediate images of the first, second, third, ... Time phases can be interpolated, and the time resolution is doubled by this interpolation.

In the data used for reconstructing the image of the first 'time phase, the second' time phase, the third 'time phase, ..., The phase encoding amount increases from near zero (DC) to the plus side. Since it becomes the maximum on the minus side in the middle, it changes in the order of going to near zero again, so
The data acquisition times near DC are temporally separated. Since the data in which the amount of phase encoding is near zero has the largest signal component, if these are temporally separated, there is a possibility that image blurring may occur due to movement of organs or the like between them.

Therefore, in consideration of this point, the order of changing the amount of phase encoding can be set as shown in FIG. FIG. 4 shows an MR imaging method according to another embodiment. In terms of each time phase of the first R wave, the phase encoding amount is #
1, # 33, # 65, # 97, # 17, # 49, # 8
Two cycles are performed in which the phase encoding amount changes from 1 to # 113, and the phase encoding amount changes from the minus side to the plus side through near DC. Therefore, the first time phase, the second time phase,
Third time phase, ... Intermediate 1'time phase, 2'time phase, 3 '
For the time phases, ..., the phase encoding amount near DC appears twice without being relatively separated in time, and the first 'time phase, the second' time phase, the third 'time phase ,. Each image of is less affected by the movement of organs. For the first time phase, the second time phase, the third time phase, ... The phase encoding amounts near DC are temporally different from those in the case of FIG. 1, so some deterioration of the image cannot be avoided. But,
The image quality of the first 'time phase, the second' time phase, the third 'time phase, etc. is the same, and it can be evaluated that a large number of images with uniform image quality and high time resolution can be obtained as a whole. ..

In each of the above embodiments, the first
The first 'time phase, the second' time phase, the third 'time phase, which are located exactly in the middle of the respective time phases of the time phase, the second time phase, the third time phase, ...
The image of ... Is reconstructed, but if the data set shifted by an arbitrary number is fetched as the data for the interpolated image instead of fetching the data set shifted by four, the first
It is possible to obtain up to seven interpolated images whose time phases are deviated between the respective time phases of the time phase, the second time phase, the third time phase, .... That is, the first data of the first time phase to the first data of the second time phase, the second data of the second time phase to the first data of the third time phase, and so on. By taking out the shifted data set, it is possible to create an image of the time phase which is deviated from the first time phase by 1/8 of the time phase interval. Only 3 data of the 1st time phase to the 2nd data of the 2nd time phase, 3rd data of the 2nd time phase to 2nd data of the 3rd time phase, and so on. By extracting the data set, it is possible to create an image of the time phase that is shifted from the first time phase by 2/8 of the time phase interval.

[0019]

As described above, according to the MR imaging method of the present invention, the data set for each time phase is extracted from the data set for each time phase collected by one cine imaging. By reconstructing the image,
Images located between each time phase can be interpolated without extending the entire inspection time, and the time resolution can be improved.

[Brief description of drawings]

FIG. 1 is a time chart for explaining an MR imaging method according to an embodiment of the present invention.

FIG. 2 is a time chart showing a pulse sequence of the field echo method.

FIG. 3 is a diagram for explaining a raw data array and its acquisition order.

FIG. 4 is a time chart for explaining an MR imaging method according to another embodiment.

[Explanation of symbols]

 RF high-frequency excitation pulse Gz Z-direction gradient magnetic field Gy Y-direction gradient magnetic field Gx X-direction gradient magnetic field P Raw data array

Claims (1)

[Claims] 1. A high-speed field echo sequence of short TR is repeatedly acquired for each phase encoding amount combination by repeatedly collecting data while changing the phase encoding amount, and obtained in each phase encoding amount set. Each image is reconstructed using the data set, and data corresponding to one set of the above-mentioned combination of the phase encoding amounts, which is located in the middle of both adjacent two phase encoding amount sets, is extracted. An MR imaging method characterized by reconstructing one image from each of these data sets.