CN103809140A - Small-view-field magnetic resonance imaging method based on single-sweep super-speed orthogonal space-time coding - Google Patents

Abstract

The invention discloses a small-view-field magnetic resonance imaging method based on single-sweep super-speed orthogonal space-time coding. The method comprises the following steps of: enabling protons in a space to automatically rotate in an excitation phase by virtue of the organic combination of an orthogonally-distributed space coding gradient and a linear frequency sweep pulse to acquire a secondary phase related to a space position, thus carrying out two-dimensional space-time coding on the automatic rotation of the protons in an imaging plane; for the automatic rotation of the protons in an orthogonal space-time coding space, only the automatic rotation of the protons with static phase distribution can be detected during a decoding sampling period, and according to the characteristic of orthogonal space-time coding, decoding sampling can be carried out on a plurality of randomly-distributed areas in the space by designing a decoding sampling gradient, thus acquiring the magnetic resonance data of a plurality of areas of interest finally. High-resolution reconstruction is sequentially carried out on acquired magnetic resonance data of the plurality of areas, and then the high-resolution small-view-field magnetic resonance images of the plurality of areas can be obtained finally.

Description

Based on the low coverage MR imaging method of the supper-fast orthogonal space-time coding of single sweep

Technical field

The present invention relates to the method for magnetic resonance imaging, especially relate to the method for the High-resolution MRI of the multiple discrete distributed areas of single sweep acquisition.

Background technology

Echo planar imaging imaging (echo planar imaging, EPI)) carry out quick sampling by the gtadient echo of a series of quick switchings, once excite and just can obtain a width magnetic resonance picture.EPI relies on its high s/n ratio and ultrafast image taking speed to become gradually the first-selected formation method of functional neurosurgery imaging (functional neuroimaging) and the contour dimensional imaging of diffusion tensor imaging (diffusion tensor imaging, DTI).But EPI formation method for example, because intrinsic problem itself has had a strong impact on its effect, ghost and annular artifact, the problems such as the pattern distortion causing because magnetic susceptibility is inhomogeneous.Picture quality that these problems are badly damaged and limited spatial resolution.These artifacts are due to sampled signal, in the phase error of phase-encoding direction, through long excessive the causing of echo train sampling period accumulation.Although there are now a lot of phase alignments can alleviate the impact of artifact, but the calibration steps of the overwhelming majority all needs extra prescan, this will affect imaging efficiency greatly, and for some dynamic imagings, occurrence positions variation in imaging position when prescan, to make extra prescanned data occur error, thus ineffective.Another method is exactly to reduce the length of sampling echo train, to shorten the integration time of error, thereby alleviates the impact of artifact.But simply reduce the length of echo train, can sacrifice spatial resolution.In order still to keep good spatial resolution in reducing echo train, low coverage imaging technique is applied to phase-encoding direction.But, for EPI method, because the sampling bandwidth of phase encoding dimension is too low, in the time that phase-encoding direction is implemented low coverage imaging, there will be folding artifact.In order to solve this folding artifact, some low coverage technology are arisen at the historic moment.More existing low coverage methods, comprise space saturation method, quadrature excitation method, Two-dimensional Pulsed advocate approach, but in the imaging sequence of single sweep, performance is all barely satisfactory, is not in last data, to still have the residual of other signals, is easy to exactly be subject to the impact of nonuniform field.And these low coverage formation methods, once excite the low coverage magnetic resonance picture that can only obtain unique region in space, want to obtain N discrete distributed areas low coverage image in space, just must carry out being excited into for N time picture and just can complete, like this just reduced imaging efficiency.

Summary of the invention

Technical problem underlying to be solved by this invention is to provide a kind of in the situation that once exciting, just can obtain multiple discrete distributed areas, have resist very by force nonuniform field ability, avoid the low coverage formation method high with spatial resolution that artifact perplexs

In order to solve above-mentioned technical matters, the invention provides the low coverage MR imaging method based on the supper-fast orthogonal space-time coding of single sweep, comprise following steps:

(1) first imaging object is carried out to area-of-interest location, then carry out tuning, shimming, power and frequency correction;

(2) import compiled single sweep orthogonal space-time coded sequence in advance; According to experiment situation, the parameters of pulse train is set;

The structure of described single sweep orthogonal space-time coded sequence is followed successively by: 90 degree linear frequency sweep pulses of high frequency sweep rate, TE ₁/ 2,180 of low frequency sweep rate degree linear frequency sweep pulses, TE ₁/ 2, TE ₂sinc pulse, the TE of/2,180 degree ₂/ 2, sampling echo train;

90 degree linear frequency sweep pulses of described high frequency sweep rate are in conjunction with the space encoding gradient G of 90 degree ₉₀low bandwidth dimension (y direction) is carried out to space encoding, reunion gradient G _erand then act on described space encoding gradient G ₉₀afterwards, its gradient area is described space encoding gradient G ₉₀the half of area, positive and negative and described space encoding gradient G ₉₀on the contrary; 180 degree linear frequency sweep pulses of described low frequency sweep rate are in conjunction with the space encoding gradient G of 180 degree ₁₈₀high bandwidth dimension (x direction) is carried out to space encoding; The sinc pulse of described 180 degree and layer select gradient G _sscarry out layer choosing; Before and after 180 degree linear frequency sweep pulses of described low frequency sweep rate, execute and added respectively echo time delay TE ₁/ 2, before and after the sinc pulse of described 180 degree, apply respectively echo time delay TE ₂/ 2, before and after 180 degree linear frequency sweep pulses of described low frequency sweep rate and the sinc pulse of 180 degree, there is x, y, the destruction gradient effect of tri-directions of z;

Described sampling echo train is by acting on respectively x, the gradient chain composition of y direction; The gradient chain of x direction is to be made up of the gradient of a series of positive and negative switchings, and the area of each gradient is described space encoding gradient G ₁₈₀two times of area; The gradient chain of y direction is to be made up of a series of equal-sized " blips " gradient, and the total area of described " blips " gradient and equal described space encoding gradient G ₉₀area, positive and negative and described space encoding gradient G ₉₀unanimously;

Before described sampling echo train, x and y direction have applied respectively reunion gradient G _rorand G _ar, described G _rorarea be the half of first gradient area of x direction, direction is in contrast; Described G _ararea be the half of the total area of all described " blips " gradients, direction is contrary with described " blips " gradient;

(3) the described single sweep orthogonal space-time coded sequence that execution step (2) sets, carries out data sampling; After data sampling completes, sampled data is carried out to super-resolution reconstruction, obtain high-resolution orthogonal space-time coding magnetic resonance image;

The step of described super-resolution reconstruction is by advanced the two-dimentional single sweep orthogonal space-time coded data obtaining line phase smoothing processing, then carries out two-dimentional interpolation, finally carries out super-resolution reconstruction by dimension;

(4) described high-resolution orthogonal space-time coding magnetic resonance image step (3) being obtained carries out region of interest field mark, then writes down coordinate and the visual field size in each region;

(5) the each area coordinate obtaining according to step (4) and visual field size, according to the principle of orthogonal space-time coding, design corresponding low coverage decoding gradient;

Low coverage decoding gradient concrete steps corresponding to described design are: first, according to coordinate and the visual field size of the area-of-interest of record in step (4), obtain corresponding gradient file by following relation:

$\left\{\begin{array}{c}y=(-\frac{1}{2}+\frac{{\∫}_0^{t_{\mathrm{ay}}}{G}_{\mathrm{acq}}^y\left(t\right)\mathrm{dt}}{G_{90}{T}_{90}})\·L_y\\ x=(\frac{1}{2}-\frac{{\∫}_0^{t_{\mathrm{ax}}}{G}_{\mathrm{acq}}^x\left(t\right)\mathrm{dt}}{G_{180}{T}_{180}})\·L_x\end{array}\right.$

X in formula, y is locus coordinate, L _x, L _ybe respectively x, the visual field size of y direction; T ₉₀for 90 of described high frequency sweep rate is spent the time of linear frequency sweep pulses; T ₁₈₀for 180 of described low frequency sweep rate is spent the time of linear frequency sweep pulses; Just can obtain low coverage decoding gradient according to this relational expression;

(6) the flexible visual field imaging gradient of step (5) design is imported in described single sweep orthogonal space-time coded sequence, obtain low coverage imaging sequence operation based on single sweep orthogonal space-time coding, obtain low coverage magnetic resonance and look like into data;

(7) described low coverage magnetic resonance imaging data step (6) being obtained, by the sequencing of decoding sampling, thereby carry out successively High resolution reconstruction and obtain high-resolution low coverage magnetic resonance imaging data, and coordinate and the visual field size write down according to step (4), the described high-resolution low coverage magnetic resonance imaging data in the multiple regions that obtain is combined in piece image, finally obtains complete magnetic resonance image (MRI);

The step of described low coverage magnetic resonance imaging data being carried out to super-resolution reconstruction is: by the sequencing of decoding sampling, respectively each described low coverage magnetic resonance imaging data is filled into independently in grid matrix, form multiple two-dimensional data matrix, then these 2-D datas are carried out to High resolution reconstruction according to the process of reconstruction of step (3).

Low coverage MR imaging method based on the supper-fast orthogonal space-time coding of single sweep provided by the present invention, in the situation that once exciting, just can carry out low coverage imaging to the region of multiple discrete distributions in space, and improve the spatial resolution of image.

Accompanying drawing explanation

Fig. 1 is single sweep orthogonal space-time structural coding sequence figure in the present invention;

Fig. 2 is the low coverage imaging sequence structural drawing based on single sweep orthogonal space-time coding in the present invention;

Fig. 3 has shown the magnetic resonance image (MRI) that different sequences obtain, wherein:

A is the gtadient echo image of many scannings;

B is the single sweep orthogonal space-time coded image in the full visual field;

C is the single sweep orthogonal space-time coded image of low coverage;

Embodiment

Below by reference to the accompanying drawings and embodiment, the present invention will be further described:

Each step in specific implementation process of the present invention is as follows:

(1) first imaging object is carried out to area-of-interest location, then carry out tuning, shimming, power and frequency correction;

(2) import compiled single sweep orthogonal space-time coded sequence in advance; According to experiment situation, the parameters of pulse train is set;

The structure of described single sweep orthogonal space-time coded sequence is followed successively by: 90 degree linear frequency sweep pulses of high frequency sweep rate, TE ₁/ 2,180 of low frequency sweep rate degree linear frequency sweep pulses, TE ₁/ 2, TE ₂sinc pulse, the TE of/2,180 degree ₂/ 2, sampling echo train;

90 degree linear frequency sweep pulses of described high frequency sweep rate are in conjunction with the space encoding gradient G of 90 degree ₉₀low bandwidth dimension (y direction) is carried out to space encoding, reunion gradient G _erand then act on described space encoding gradient G ₉₀afterwards, its gradient area is described space encoding gradient G ₉₀the half of area, positive and negative and described space encoding gradient G ₉₀on the contrary; 180 degree linear frequency sweep pulses of described low frequency sweep rate are in conjunction with the space encoding gradient G of 180 degree ₁₈₀high bandwidth dimension (x direction) is carried out to space encoding; The sinc pulse of described 180 degree and layer select gradient G _sscarry out layer choosing; Before and after 180 degree linear frequency sweep pulses of described low frequency sweep rate, execute and added respectively echo time delay TE ₁/ 2, before and after the sinc pulse of described 180 degree, apply respectively echo time delay TE ₂/ 2, before and after 180 degree linear frequency sweep pulses of described low frequency sweep rate and the sinc pulse of 180 degree, there is x, y, the destruction gradient effect of tri-directions of z;

Described sampling echo train is by acting on respectively x, the gradient chain composition of y direction; The gradient chain of x direction is to be made up of the gradient of a series of positive and negative switchings, and the area of each gradient is described space encoding gradient G ₁₈₀two times of area; The gradient chain of y direction is to be made up of a series of equal-sized " blips " gradient, and the total area of described " blips " gradient and equal described space encoding gradient G ₉₀area, positive and negative and described space encoding gradient G ₉₀unanimously;

Before described sampling echo train, x and y direction have applied respectively reunion gradient G _rorand G _ar, described G _rorarea be the half of first gradient area of x direction, direction is in contrast; Described G _ararea be the half of the total area of all described " blips " gradients, direction is contrary with described " blips " gradient;

(3) the described single sweep orthogonal space-time coded sequence that execution step (2) sets, carries out data sampling; After data sampling completes, sampled data is carried out to super-resolution reconstruction, obtain high-resolution orthogonal space-time coding magnetic resonance image;

The step of described super-resolution reconstruction is by advanced the two-dimentional single sweep orthogonal space-time coded data obtaining line phase smoothing processing, then carries out two-dimentional interpolation, finally carries out super-resolution reconstruction by dimension;

(4) described high-resolution orthogonal space-time coding magnetic resonance image step (3) being obtained carries out region of interest field mark, then writes down coordinate and the visual field size in each region;

(5) the each area coordinate obtaining according to step (4) and visual field size, according to the principle of orthogonal space-time coding, design corresponding low coverage decoding gradient;

Low coverage decoding gradient concrete steps corresponding to described design are: first, according to coordinate and the visual field size of the area-of-interest of record in step (4), obtain corresponding gradient file by following relation:

$\left\{\begin{array}{c}y=(-\frac{1}{2}+\frac{{\∫}_0^{t_{\mathrm{ay}}}{G}_{\mathrm{acq}}^y\left(t\right)\mathrm{dt}}{G_{90}{T}_{90}})\·L_y\\ x=(\frac{1}{2}-\frac{{\∫}_0^{t_{\mathrm{ax}}}{G}_{\mathrm{acq}}^x\left(t\right)\mathrm{dt}}{G_{180}{T}_{180}})\·L_x\end{array}\right.$

X in formula, y is locus coordinate, L _x, L _ybe respectively x, the visual field size of y direction; T ₉₀for 90 of described high frequency sweep rate is spent the time of linear frequency sweep pulses; T ₁₈₀for 180 of described low frequency sweep rate is spent the time of linear frequency sweep pulses; Just can obtain low coverage decoding gradient according to this relational expression;

(6) the flexible visual field imaging gradient of step (5) design is imported in described single sweep orthogonal space-time coded sequence, obtain low coverage imaging sequence operation based on single sweep orthogonal space-time coding, obtain low coverage magnetic resonance and look like into data;

(7) described low coverage magnetic resonance imaging data step (6) being obtained, by the sequencing of decoding sampling, thereby carry out successively High resolution reconstruction and obtain high-resolution low coverage magnetic resonance imaging data, and coordinate and the visual field size write down according to step (4), the described high-resolution low coverage magnetic resonance imaging data in the multiple regions that obtain is combined in piece image, finally obtains complete magnetic resonance image (MRI);

The step of described low coverage magnetic resonance imaging data being carried out to super-resolution reconstruction is: by the sequencing of decoding sampling, respectively each described low coverage magnetic resonance imaging data is filled into independently in grid matrix, form multiple two-dimensional data matrix, then these 2-D datas are carried out to High resolution reconstruction according to the process of reconstruction of step (3).

Embodiment:

We,, by the low coverage MR imaging method based on the supper-fast orthogonal space-time coding of single sweep, carry out embodiment displaying on live body SD rat, are used for verifying feasibility of the present invention.Experiment test is to carry out under a Varian7T imager (Agilent Technologies, Santa Clara, CA, USA).The imaging sample that adopts of experiment is the live body SD rat of approximately 250 grams, first uses 10% chloral hydrate solution before experiment, in 0.4mg/100g ratio, rat is carried out to injecting anesthetic, tests accordingly after waiting mouse to enter dormant state.Before testing, first rat is fixed on experimental bed, then imports in imager.On magnetic resonance imager operator's console, open the corresponding function software of imager, interested rat position is positioned, we select the brain of rat to carry out coronal surface imaging herein.Then we carry out tuning, shimming, frequency and capability correction.According to the operating process of the above-mentioned flexible visual field formation method based on single sweep orthogonal space-time coding, before carrying out the imaging of the flexible visual field, first carry out the orthogonal space-time coding method in the full visual field, obtain the magnetic resonance image (MRI) of the full visual field of a width rat brain, be used for coordinate setting and mark.

First measure respectively the power of 90 degree used and 180 degree linear frequency sweep pulses by the space-time code sequence of one dimension.Then import compiled full visual field single sweep orthogonal space-time coded sequence as shown in Figure 1, open each correlation module of pulse train, comprise orthogonal space-time coding module, echo time delay module TE ₁and TE ₂, decoding sampling module, experiment parameter is set, the sample specifically adopting for the present embodiment, its test parameters arranges as follows: 90 spend the stimulating frequency width Delta Ο of linear frequency sweep pulses ₉₀for 64kHz, firing time T ₉₀be 3 milliseconds, the stimulating frequency width Delta Ο of 180 degree linear frequency sweep pulses ₁₈₀for 8kHz, firing time T ₁₈₀be 4 milliseconds, x direction sampling number N _xbe 64, y direction sampling number N _ybe 64, the sampling bandwidth sw of high bandwidth dimension (x direction) is 250kHz, echo delay time TE ₁be 3.2 milliseconds, TE ₂be 31.97 milliseconds, the visual field FOV of x direction _xbe 4.5 centimetres, the visual field FOV of y direction _ybe 4.5 centimetres, imaging layer thickness t hk is 2 millimeters, and the intensity of destroying gradient is every centimetre of 3.0 Gauss, and the time is 1 millisecond.After above experiment parameter is set, the sampling time of directly moving whole sequence is about 60 milliseconds.After sampling finishes, just obtain the rat brain MR data in the full visual field, after High resolution reconstruction, obtained high-resolution rat brain image (shown in Fig. 3 B); Under the environment of identical field, jump to other test blocks, call gradin-echo, arrange behind the identical visual field and bed thickness, test, obtain many scan images as with reference to figure (shown in Fig. 3 A).

The full visual field rat brain image obtaining with the coding method of single sweep orthogonal space-time, as egative film, selects interested region to carry out flexible field-of-view image; In the present embodiment, chosen respectively the eyes of rat and mouse brain the latter half as interested region, in Fig. 3 B, irised out with green and red square frame, the sampling number in two regions is all mutually: x direction sampling number is that 64, y direction sampling number is 30.Then obtain the gradient file for flexible visual field imaging according to step (4), (5).

The flexible gradient file copy producing is before deposited to the proprietary file underedge of gradient file to imager.In order to keep and the complete the same test parameters of visual field imaging, the imaging experiment district, the full visual field of sampling before being transferred to, then call in flexible visual field gradient file, sequence now as shown in Figure 2, check errorless after, the pulse train of execution graph 2, then carries out High resolution reconstruction processing by the data of being adopted according to step (7), obtains flexible visual field magnetic resonance image (MRI) (shown in Fig. 3 C).Can find out from Fig. 3 C, the eyes of rat are compared Fig. 3 B with olfactory bulb region, have better spatial resolution and integrity degree; Compared with Fig. 2 B, the mouse brain bottom half image in Fig. 3 C and the matching degree of Fig. 3 A are higher.Can prove thus, based on the flexible visual field of the single sweep formation method of orthogonal space-time coding, can realize and once excite the low coverage magnetic resonance picture that obtains multiple discrete distributed areas, and spatial resolution also get a promotion.

The above, only for preferred embodiment of the present invention, therefore can not limit according to this scope of the invention process, the equivalence of doing according to the scope of the claims of the present invention and description changes and modifies, and all should still belong in the scope that the present invention contains.

Claims (1)

1. the low coverage MR imaging method based on the supper-fast orthogonal space-time coding of single sweep, comprises following steps:

(1) first imaging object is carried out to area-of-interest location, then carry out tuning, shimming, power and frequency correction;

(2) import compiled single sweep orthogonal space-time coded sequence in advance; According to experiment situation, the parameters of pulse train is set;

The structure of described single sweep orthogonal space-time coded sequence is followed successively by: 90 degree linear frequency sweep pulses of high frequency sweep rate, TE ₁/ 2,180 of low frequency sweep rate degree linear frequency sweep pulses, TE ₁/ 2, TE ₂sinc pulse, the TE of/2,180 degree ₂/ 2, sampling echo train;

90 degree linear frequency sweep pulses of described high frequency sweep rate are in conjunction with the space encoding gradient G of 90 degree ₉₀low bandwidth dimension (y direction) is carried out to space encoding, reunion gradient G _erand then act on described space encoding gradient G ₉₀afterwards, its gradient area is described space encoding gradient G ₉₀the half of area, positive and negative and described space encoding gradient G ₉₀on the contrary; 180 degree linear frequency sweep pulses of described low frequency sweep rate are in conjunction with the space encoding gradient G of 180 degree ₁₈₀high bandwidth dimension (x direction) is carried out to space encoding; The sinc pulse of described 180 degree and layer select gradient G _sscarry out layer choosing; Before and after 180 degree linear frequency sweep pulses of described low frequency sweep rate, execute and added respectively echo time delay TE ₁/ 2, before and after the sinc pulse of described 180 degree, apply respectively echo time delay TE ₂/ 2, before and after 180 degree linear frequency sweep pulses of described low frequency sweep rate and the sinc pulse of 180 degree, there is x, y, the destruction gradient effect of tri-directions of z;

Described sampling echo train is by acting on respectively x, the gradient chain composition of y direction; The gradient chain of x direction is to be made up of the gradient of a series of positive and negative switchings, and the area of each gradient is described space encoding gradient G ₁₈₀two times of area; The gradient chain of y direction is to be made up of a series of equal-sized " blips " gradient, and the total area of described " blips " gradient and equal described space encoding gradient G ₉₀area, positive and negative and described space encoding gradient G ₉₀unanimously;

Before described sampling echo train, x and y direction have applied respectively reunion gradient G _rorand G _ar, described G _rorarea be the half of first gradient area of x direction, direction is in contrast; Described G _ararea be the half of the total area of all described " blips " gradients, direction is contrary with described " blips " gradient;

(3) the described single sweep orthogonal space-time coded sequence that execution step (2) sets, carries out data sampling; After data sampling completes, sampled data is carried out to super-resolution reconstruction, obtain high-resolution orthogonal space-time coding magnetic resonance image;

The step of described super-resolution reconstruction is by advanced the two-dimentional single sweep orthogonal space-time coded data obtaining line phase smoothing processing, then carries out two-dimentional interpolation, finally carries out super-resolution reconstruction by dimension;

(4) described high-resolution orthogonal space-time coding magnetic resonance image step (3) being obtained carries out region of interest field mark, then writes down coordinate and the visual field size in each region;

(5) the each area coordinate obtaining according to step (4) and visual field size, according to the principle of orthogonal space-time coding, design corresponding low coverage decoding gradient;

Low coverage decoding gradient concrete steps corresponding to described design are: first, according to coordinate and the visual field size of the area-of-interest of record in step (4), obtain corresponding gradient file by following relation:

$\left\{\begin{array}{c}y=(-\frac{1}{2}+\frac{{\∫}_0^{t_{\mathrm{ay}}}{G}_{\mathrm{acq}}^y\left(t\right)\mathrm{dt}}{G_{90}{T}_{90}})\·L_y\\ x=(\frac{1}{2}-\frac{{\∫}_0^{t_{\mathrm{ax}}}{G}_{\mathrm{acq}}^x\left(t\right)\mathrm{dt}}{G_{180}{T}_{180}})\·L_x\end{array}\right.$

X in formula, y is locus coordinate, L _x, L _ybe respectively x, the visual field size of y direction; T ₉₀for 90 of described high frequency sweep rate is spent the time of linear frequency sweep pulses; T ₁₈₀for 180 of described low frequency sweep rate is spent the time of linear frequency sweep pulses; Just can obtain low coverage decoding gradient according to this relational expression;

(6) the flexible visual field imaging gradient of step (5) design is imported in described single sweep orthogonal space-time coded sequence, obtain low coverage imaging sequence operation based on single sweep orthogonal space-time coding, obtain low coverage magnetic resonance and look like into data;

(7) described low coverage magnetic resonance imaging data step (6) being obtained, by the sequencing of decoding sampling, thereby carry out successively High resolution reconstruction and obtain high-resolution low coverage magnetic resonance imaging data, and coordinate and the visual field size write down according to step (4), the described high-resolution low coverage magnetic resonance imaging data in the multiple regions that obtain is combined in piece image, finally obtains complete magnetic resonance image (MRI);

The step of described low coverage magnetic resonance imaging data being carried out to super-resolution reconstruction is: by the sequencing of decoding sampling, respectively each described low coverage magnetic resonance imaging data is filled into independently in grid matrix, form multiple two-dimensional data matrix, then these 2-D datas are carried out to High resolution reconstruction according to the process of reconstruction of step (3).

Images