CN105662413B - A kind of method and apparatus that cardiac muscle is quantitative T1 - Google Patents

Abstract

The present application discloses a method for quantifying myocardial T1, which includes: applying a non-slice-selective inversion pulse after the trigger delay of ECG gating; The above images are interleaved in real time to capture the inversion recovery process of the signal; select the sampling line in the diastolic period as the K-space center line; take the selected sampling line as the center, and select the sampling line symmetrically according to the reconstruction window size for image processing Reconstruction; fit the T1 quantitative map with the reconstructed image. The application also discloses a device based on the method. This application can realize the acquisition of multi-layer T1 quantitative images in one breath-hold, and can cover the whole heart in 2-3 breath-holds, reducing the time waste and patient discomfort caused by breath-holding.

Description

A method and device for quantifying myocardial T1

technical field

The present application relates to the field of cardiovascular imaging, and specifically relates to a method and device for quantifying myocardial T1.

Background technique

Cardiovascular magnetic resonance imaging (CMR) is a non-invasive imaging technique that can evaluate the function, shape and structure of the cardiovascular system. The quantification of myocardial T1 value can be used in the diagnosis of various diseases such as myocardial infarction and myocardial fibrosis. In vivo T1 quantification of myocardium remains challenging due to acquisition time constraints imposed by heart beating and respiratory motion.

The commonly used methods for measuring myocardial T1 value are as follows:

(1) MOLLI (Modified Look-Locker inversion recovery): After the inversion pulse (IR), use bssfp (balance steady state free pregnancy) sequence imaging at different delay (TI) times, and combine the data collected multiple times into A group, to fit the T1 value.

(2) SASHA (Saturation recovery single-shot acquisition): The saturation pulse was used instead of the inversion pulse, and the bssfp sequence was used for imaging with different delay times after the saturation pulse, and the acquisition was completed within 10 heartbeat cycles.

In the above method, only a single-layer image can be collected for each breath-hold, and multiple breath-holds are required to collect multi-layer images. The number of breath-holds is large, the acquisition time is long, and the patient's comfort is low.

Contents of the invention

This application provides a method for quantitative measurement of myocardial T1 value, including:

After the ECG gating trigger delay, apply non-slice-selective inversion pulse, and use the fast spoiled gradient echo small-angle excitation sequence of radial sampling trajectory to perform real-time interleaved acquisition of images of at least two layers, and capture the inversion recovery of the signal Process; select the sampling line in the diastolic period as the K-space center line; take the selected sampling line as the center, and perform image reconstruction according to the selected sampling line symmetrically according to the reconstruction window size; use the reconstructed image to fit T1 Quantitative graph.

The non-slice-selective inversion pulse is applied after the ECG gating trigger delay, and the fast spoiler gradient echo small-angle excitation sequence of the radial sampling trajectory is used to perform real-time interleaved acquisition of images of at least two layers, and the reflection of the signal is captured. The conversion and recovery process includes: applying a non-slice-selected inversion pulse, using the fast spoiler gradient echo small-angle excitation sequence of the radial sampling track for real-time acquisition, and sampling the signal inversion recovery process. The radial sampling The fast spoiler gradient echo small-angle excitation sequence of the trajectory adopts the multi-layer interleaved acquisition mode, and at least two layers of images can be acquired after one inversion pulse.

The above-mentioned radial sampling trajectory adopts the small golden angle sampling mode, and the azimuth angle of the i-th sampling line is (i-1)ψ _N degrees, that is, the azimuth angle of each sampling line is increased by a fixed angle ψ _N , where _N determines the size of ψN, which is determined empirically. The number of sampling lines must be greater than N.

The real-time acquisition of the above-mentioned fast spoiled gradient echo small-angle excitation sequence using the radial sampling trajectory is carried out during the breath-hold process, and the shortest TE and TR are used.

The above selection of the sampling line in diastole as the K-space centerline includes: retrospectively selecting the sampling line whose acquisition time is in diastole as the reconstructed K-space centerline according to the size of the sliding window.

Taking the selected sampling line as the center, and selecting the sampling line symmetrically according to the size of the reconstruction window to perform image reconstruction includes: using the sampling line whose acquisition time is in the diastolic period as the reconstructed K-space center line to determine the position of the reconstruction window , for the reconstruction window, the KWIC method is used to weight the K-space data, and the fast reconstruction algorithm can be used to reconstruct the data.

According to the second aspect of the present application, a device for quantifying myocardial T1 is provided, including: a data acquisition module, which is used to apply a non-layer-selective inversion pulse after the trigger delay of the ECG gating, and adopts fast phase spoiling of the radial sampling trajectory The gradient echo small-angle excitation sequence performs real-time interlaced acquisition of at least two layers of images, and captures the inversion recovery process of the signal; the selection module is used to select the sampling line in the diastolic period as the K-space center line; the image reconstruction module , for performing image reconstruction based on the selected sampling line centered on the selected sampling line with a symmetrical reconstruction window size; a fitting module is used for fitting the T1 quantitative map using the reconstructed image.

The above-mentioned data acquisition module is also used to apply non-layer-selected inversion pulses, and use the fast spoiler gradient echo small-angle excitation sequence of the radial sampling trajectory for real-time acquisition, and sample the inversion recovery process of the signal. The fast spoiled gradient echo small-angle excitation sequence of the sampling track adopts the multi-layer interleaved acquisition mode, and at least two layers of images can be collected after one inversion pulse.

The above-mentioned radial sampling trajectory adopts the small golden angle sampling mode, and the azimuth angle of the i-th sampling line is (i-1)ψ _N degrees, that is, the azimuth angle of each sampling line is increased by a fixed angle ψ _N , where N determines the size of ψN, _N is determined by experience, and the number of sampling lines must be greater than N.

The above-mentioned data collection module is also used for data collection during the breath-holding process, using the shortest TE and TR.

The above selection module is also used to retrospectively select the sampling line whose acquisition time is in diastole as the reconstructed K-space center line according to the size of the sliding window.

The above image reconstruction module is also used to determine the position of the reconstruction window by using the sampling line whose acquisition time is located in the diastolic period as the centerline of the reconstructed K space, and for the reconstruction window, the KWIC method is used to weight the K space data, using The fast reconstruction algorithm reconstructs the data.

Owing to adopting above technical scheme, the beneficial effect that makes this application possess is:

In the specific embodiment of the present application, due to the use of the fast spoiled gradient echo small-angle excitation sequence and the small golden angle sampling mode using the radial sampling trajectory, after the non-slice inversion pulse, the images of at least two layers Real-time staggered acquisition can realize the acquisition of multi-layer T1 quantitative images in one breath-hold, and can cover the whole heart in 2-3 breath-holds, reducing the time waste and patient discomfort caused by breath-holds; using the KWIC method for K-space The weighting of the data can clarify the time of collecting the central point of K space, which can effectively prevent the deviation of the measured value.

Description of drawings

Fig. 1 is a flow chart according to one embodiment of the method of the present application;

Fig. 2 is a schematic diagram of collection according to an embodiment of the method of the present application;

Fig. 3 is a radial sampling gradient waveform diagram according to an embodiment of the method of the present application;

Fig. 4 is a sampling trajectory diagram according to an embodiment of the method of the present application;

Fig. 5 is a diastolic selection diagram according to an embodiment of the method of the present application;

Fig. 6 is a KWIC weighted weight diagram according to an embodiment of the method of the present application;

Fig. 7 is a schematic structural diagram of an embodiment of a device according to the present application.

Detailed ways

The present invention will be further described in detail below through specific embodiments in conjunction with the accompanying drawings.

Embodiment one:

Fig. 1 shows a flowchart according to an embodiment of the method of the present application, including:

Step 102: After the ECG gating trigger delay, apply a non-slice-selective inversion pulse, and use the fast spoiler gradient echo small-angle excitation sequence of the radial sampling trajectory to perform real-time interleaved acquisition of at least two layers of images to capture the signal reverse recovery process.

Apply a non-slice-selected inversion pulse (IR), and then use the fast spoiled gradient echo small-angle excitation sequence (turbo-flash) of the radial sampling trajectory for real-time acquisition to sample the inversion recovery process of the signal, turbo- The flash adopts a multi-layer interleaved acquisition mode (interleaved acquisition), and multi-layer images can be collected after an inversion pulse, as shown in Figure 2.

The radial sampling gradient waveform is shown in Figure 3. In radial sampling, each sampling line passes through the center of K-space. The two-dimensional radial sampling trajectory of the small golden angle adds a fixed angle ψ _N to the orientation angle of each sampling line, Each sampling line in this sampling mode will not overlap, and when the number of sampling lines > N, the distribution of sampling lines is approximately uniform, and any sampling line can be used as the starting position for reconstruction, and the window size >= The size of the reconstruction window is set arbitrarily within the range of N. One implementation mode is N=5, as shown in FIG. 4 , which is the sampling trajectory of the radial small golden angle when N=5.

The trigger delay of ECG gating is set to 60% of the patient's heartbeat cycle (RR) (about 500-650ms). The delay time of ECG gating is to ensure that the initial inversion recovery signal is collected during the diastolic period of the heartbeat, so the delay Timing is determined by the heartbeat cycle of the imaged subject. The entire acquisition process is completed during the breath-hold process. In order to improve the time resolution of imaging, the shortest TE (time of repetition) and TR (time of echo) are used, that is, TR/TE is determined by the hardware limitations of the magnetic resonance scanner (maximum gradient amplitude, gradient climbing rate, etc.) For the shortest time, TR increases with the increase of the number of acquisition layers, specifically, TR = the shortest TR of the acquisition layer × the number of acquisition layers, and the flip angle is 5 degrees.

Step 104: Select a sampling line in diastole as the center line of K-space.

Data acquisition is carried out during the whole heartbeat cycle, but due to the heart motion, direct image reconstruction cannot directly fit the image point by point. The heart runs slowly during diastole, and the sampling line whose acquisition time is in diastole can be retrospectively selected for subsequent image reconstruction and data fitting. The selection method is as follows: when collecting data, the original data header file will take the current ECG trigger time as the starting point to record the current acquisition time, and when the next ECG trigger arrives, set this time to 0, and the maximum time is the heartbeat cycle The time between 65% and 95% of the heartbeat cycle is the diastolic period, and the collection line whose collection time is within this range is selected as the effective collection line, as shown in Figure 5. The first effective acquisition line of each heartbeat cycle is used as the first K-space centerline, and the position of the subsequent K-space centerline is determined according to the size of the sliding window, that is, the distance between adjacent K-space centerlines is the size of the sliding window.

Step 106: Taking the selected sampling line as the center, select the sampling line symmetrically according to the size of the reconstruction window to perform image reconstruction.

Select the sampling line located at the center line-window size/2～center line+window size/2 to reconstruct the current image; a sampling line can be used to reconstruct images at different center positions; the distance between the center lines is the sliding step, and those skilled in the art should Clearly, the reconstruction window size and sliding step are determined empirically. The K-space centerline is selected from the effective acquisition lines selected in step 104 in the diastolic period, and the distance between the centerlines is a sliding step. For each reconstruction window, KWIC (k-space weighted image contrast) method is used to weight the K-space data. , KWIC can clearly collect the time of the K-space center line. Figure 6 is the KWIC weight map when the reconstruction window size = 53, black = 0 and white = 1 in the figure. When the T1 value is measured by radial sampling real-time acquisition after the inversion pulse, if the KWIC weighting method is not used, the measured value will be biased.

Step 108: Fit the T1 quantitative map with the reconstructed image.

The reconstructed images were fitted to obtain T1 value images. The fitting formula is t is the time from the end of the IR module to the acquisition of the k-space centerline of each image, that is, t=TR×centerline position, and three parameters are obtained Then the calculation formula of T1 value is Curve fitting is performed on each point of the spatial coordinates in the image to obtain the T1 value image. In the actual sequence implementation, after the inversion pulse, it is necessary to add a damage gradient (Spoiler) to eliminate the residual transverse magnetization vector introduced by the imperfect inversion pulse, resulting in a certain delay between the inversion pulse and data acquisition , affects the measured T1 value, so the error caused by this delay needs to be corrected, the formula is as follows: T _1true ＝T ₁ +2*Δt, Δt is the delay between IR and data acquisition, and the final measured T1 is obtained value image.

At present, the quantitative measurement of cardiac T1 adopts the method of collecting one layer of images while holding a breath. Breath-holding requires an interactive process between the operator and the subject, resulting in a waste of time, and multiple breath-holds will also make the subject/patient feel tired. The method proposed in this application is a collection sequence and method for acquiring multiple layers of myocardial T1 quantitative images in a single breath-hold, which can cover the whole heart in 2-3 breath-holds, reducing time waste and patient discomfort caused by multiple breath-holds.

Embodiment two:

Fig. 7 is a schematic structural diagram of an embodiment of a device according to the present application, including: a data acquisition module, a selection module, an image reconstruction module and a fitting module.

The data acquisition module is used to apply a non-slice-selective inversion pulse after the trigger delay of the ECG gating, and use the fast spoiler gradient echo small-angle excitation sequence of the radial sampling trajectory to perform real-time interleaved acquisition of images of at least two layers, The inversion recovery process of the captured signal. One embodiment is also used to apply a non-slice-selected inversion pulse, and use the radial sampling trajectory of the fast spoiled gradient echo small-angle excitation sequence to perform real-time acquisition, and sample the inversion recovery process of the signal. The fast spoiler gradient echo small-angle excitation sequence of the radial sampling trajectory adopts a multi-layer interleaved acquisition mode, and at least two layers of images can be acquired after one inversion pulse. In one embodiment, the radial sampling trajectory is a small golden angle sampling mode, and the azimuth of the i sampling line is (i-1)ψ _N degrees, that is, the azimuth of each sampling line is increased by a fixed angle ψ _N , where N determines the size of ψN, _N is determined by experience, and the number of sampling lines must be greater than N. In one embodiment, the data collection module is also used to collect data during the breath-hold process, using the shortest TE and TR.

The selection module is used for selecting the sampling line in diastole as the center line of K-space. In one embodiment, the sampling line whose acquisition time is in the diastolic period is retrospectively selected as the reconstructed K-space centerline.

The image reconstruction module is used for performing image reconstruction by selecting sampling lines symmetrically according to the size of the reconstruction window with the selected sampling line as the center. One embodiment, using the sampling line whose acquisition time is located in diastole as the reconstructed K-space center line to determine the position of the reconstruction window, for the reconstruction window, the K-space data is weighted by using the KWIC method, and the fast reconstruction algorithm is used to The data is reconstructed. In one embodiment, the present application may use the cg-SENSE algorithm for reconstruction, and the present application may also use other methods for reconstruction.

The fitting module is used to fit the T1 quantitative map using the reconstructed image. Those skilled in the art can use some well-established methods for fitting.

The above content is a further detailed description of the present invention in conjunction with specific embodiments, and it cannot be assumed that the specific implementation of the present invention is limited to these descriptions. Those of ordinary skill in the technical field to which the present invention belongs can also make some simple deduction or replacement without departing from the concept of the present invention.

Claims (8)

1. a kind of method that cardiac muscle is quantitative T1, which is characterized in that including：

After ecg-gating trigger delay, apply non-layer choosing inversion pulse, is returned using the phase gradient of quickly disturbing of radial sample track Wave low-angle excitation sequence at least two layers or more of image staggeredly acquire in real time, captures the inversion recovery process of signal；

Select the sampling line in diastole as K space center line；

Centered on selected sampling line, sampling line is symmetrically selected to carry out image reconstruction according to window size is rebuild；

Quantitatively schemed with the image fitting T1 reconstructed；

It is described to apply non-layer choosing inversion pulse after ecg-gating trigger delay, mutually ladder is quickly disturbed using radial sample track The image for spending echo low-angle excitation sequence at least two layers or more staggeredly acquire in real time, captures the inversion recovery mistake of signal Journey, including：

The inversion pulse for applying non-layer choosing is carried out real with the Fast spoiled gradient echo low-angle excitation sequence of radial sample track When acquire, the inversion recovery process of signal is sampled, it is described radial direction sample track Fast spoiled gradient echo low-angle Excitation sequence uses multi-layer intercrossed acquisition mode, and at least two layers or more of image can be acquired after an inversion pulse；

The wherein described radial sample track uses small gold angle sampling configuration, and the azimuth of i-th sampling line is (i-1) ψ_NDegree, The azimuth of i.e. every sampling line increases a fixed angle ψ_N, whereinN determines ψ_N Size, for N by empirically determined, sampling line number mesh must be more than N.

2. the method as described in claim 1, which is characterized in that the small angle of Fast spoiled gradient echo of the radial direction sample track Degree excitation sequence carries out acquisition in real time and is carried out during breathing is held one's breath, the most short TE and TR of use.

3. the method as described in claim 1, which is characterized in that described to select the sampling line in diastole as K skies Between center line include：

According to sliding window size, acquisition time is selected retrospectively and is located at the sampling line of diastole as the K rebuild Space center's line.

4. method as claimed in claim 3, which is characterized in that wherein centered on selected sampling line, according to weight Build window size symmetrically select sampling line carry out image reconstruction include：

Sampling line using the acquisition time positioned at diastole determines reconstruction window as the K space center line rebuild Position is weighted K space data using K- spatial weighting image comparison methods the reconstruction window, uses quick weight Algorithm is built to rebuild data.

5. a kind of device that cardiac muscle is quantitative T1, which is characterized in that including：

Data acquisition module, for after ecg-gating trigger delay, applying non-layer choosing inversion pulse, using radial sample track Fast spoiled gradient echo low-angle excitation sequence at least two layers or more of image staggeredly acquire in real time, capture signal Inversion recovery process；

Selecting module, for selecting the sampling line in diastole as K space center line；

Image reconstruction module, for centered on selected sampling line, symmetrically selection to be adopted according to window size is rebuild Line-transect carries out image reconstruction；

Fitting module, the image fitting T1 for being reconstructed described in use quantitatively scheme；

The data acquisition module is additionally operable to apply the inversion pulse of non-layer choosing, is returned with the phase gradient of quickly disturbing of radial sample track Wave low-angle excitation sequence is acquired in real time, is sampled to the inversion recovery process of signal, the radial direction sample track Fast spoiled gradient echo low-angle excitation sequence uses multi-layer intercrossed acquisition mode, and at least two can be acquired after an inversion pulse Layer or more image；

The wherein described radial sample track uses small gold angle sampling configuration, and the azimuth of i-th sampling line is (i-1) ψ_NDegree, The azimuth of i.e. every sampling line increases a fixed angle ψ_N, whereinN determines ψ_N Size, for N by empirically determined, sampling line number mesh must be more than N.

6. device as claimed in claim 5, which is characterized in that the data acquisition module is additionally operable to during breathing is held one's breath Carry out data acquisition, the most short TE and TR of use.

7. device as claimed in claim 5, which is characterized in that the selecting module is additionally operable to, according to sliding window size, return It selects to Gu property acquisition time and is located at the sampling line of diastole as the K space center line rebuild.

8. device as claimed in claim 7, which is characterized in that described image rebuilds module and is additionally operable to use the acquisition time Sampling line positioned at diastole determines the position of reconstruction window as the K space center line rebuild, to the reconstruction window, K space data is weighted using K- spatial weighting image comparison methods, data are rebuild using quick algorithm for reconstructing.