Cardiac magnetic resonance longitudinal relaxation time measurement method, medium, and apparatus

Abstract

The present application relates to a kind of cardiac magnetic resonance longitudinal relaxation time measurement method, medium and equipment.The measurement method includes the following steps: S1 uses MOLLI sequence respectively before and after scanning contrast;Specifically, in the before and after scanning contrast, by setting different initial inversion time TI, setting 3 groups of LL experiment LL1, LL2, LL3 triggered by continuous ECG, enough relaxation curve sampling points are obtained after fusion;And data is collected at the end of diastole of cardiac cycle, and 1 image is obtained per heartbeat;S2 image processing.This measurement method can realize single-pixel high-resolution T1 mapping under the breath holding of selection data acquisition and multi-LL experiment data fusion, give consideration to high accuracy, high resolution and high efficiency, and have good compatibility and practicality, provide new scheme for early diagnosis of heart disease, lesion quantification and prognosis evaluation, have wide clinical application prospect.

Description

Technical Field

[0001] This invention belongs to the field of cardiac magnetic resonance imaging, and specifically relates to methods, media, and equipment for measuring longitudinal relaxation time in cardiac magnetic resonance imaging. Background Technology

[0002] Cardiac magnetic resonance (CMR) provides a non-invasive assessment of myocardial structure and function. T1 (longitudinal relaxation time) quantitatively describes the recovery of the magnetization vector; its value varies with tissue composition and reflects various pathological states of the myocardium. T1 mapping technology calculates the T1 parameter for each pixel by acquiring multiple CMR images and using a signal evolution model.

[0003] The core challenge of cardiac magnetic resonance imaging (MRI) is the contradiction between the long T1 value of myocardium (approximately 1000 ms at 1.5T) and the time constraints caused by cardiac / respiratory motion. Traditional inversion recovery (IR) technology requires a relaxation time of 4-5 times T1 (approximately 4-5 seconds) to complete one magnetization recovery, and a single breath-hold (approximately 20 seconds) can only acquire 4-5 data points, which cannot meet the 6-10 sampling points required for three-parameter fitting, resulting in insufficient accuracy. Although traditional LL technology can efficiently acquire relaxation curves through "single preparation pulse + multiple continuous sampling", the sampling spans the entire cardiac cycle, is greatly affected by cardiac motion, and requires manual definition of the region of interest, making pixel-by-pixel mapping impossible.

[0004] However, the following problems still exist: 1) Due to the heartbeat, the spatial coverage is limited by the acquisition time window; 2) The three-parameter fitting model cannot perfectly characterize the signal recovery process of the MOLLI sequence, so the fitting accuracy is limited; 3) The measurement results are affected by factors such as T2 relaxation and heart rate changes.

[0005] Therefore, existing T1 mapping techniques based on MOLLI sequences still need to be improved in order to better serve the diagnosis and treatment of clinical diseases. Summary of the Invention

[0006] To address the shortcomings of the prior art, this invention aims to provide a method, medium, and device for measuring longitudinal relaxation time of cardiac magnetic resonance imaging based on MOLLI sequences.

[0007] A first aspect of the present invention provides a method for measuring longitudinal relaxation time on cardiac magnetic resonance imaging, comprising the following steps: S1 uses the MOLLI sequence to perform pre- and post-contrast scans. Specifically, in the pre- and post-contrast scans, by setting different initial inversion times TI, three sets of consecutive ECG-triggered LL experiments LL1, LL2, and LL3 are set, and after fusion, a sufficient number of relaxation curve sampling points are obtained. Data is also collected at the end of diastole during the cardiac cycle, with one image acquired per heartbeat. S2 Image processing includes image sorting, curve fitting, and calculation of longitudinal relaxation time T1.

[0008] Preferably, in the pre- and post-contrast scans of step S1, the acquisition window is compressed to <200ms.

[0009] More preferably, by using a single-excitation balanced SSFP sequence as the readout module, combined with sense coding technology, and setting a reduction factor of 2, the acquisition window is compressed to <200ms.

[0010] Preferably, step S1 further includes delayed enhancement scanning.

[0011] Preferably, in step S2, the images are sorted according to the cumulative time t after reversal, specifically using the following formula: , where n is the image number in the LL experiment and RR is the interval between heartbeats.

[0012] Preferably, in step S2, a three-parameter fitting method is used for curve fitting. More preferably, the Levenberg-Marquardt algorithm is used for three-parameter fitting, and the fitting formula is as follows: Where y is the signal strength, Epigenetics in LL experiments .

[0013] Preferably, by formula Transformation to obtain reality value.

[0014] A second aspect of the present invention also provides a computer-readable storage medium storing computer instructions for causing a processor to execute the above-described method for measuring longitudinal relaxation time of cardiac magnetic resonance imaging.

[0015] In a third aspect, the present invention also provides a magnetic resonance imaging device that uses the above-described method to measure the longitudinal relaxation time of the heart.

[0016] The beneficial effects of this invention are as follows: The measurement method of this invention, through an improved scheme of "selective data acquisition + fusion of multiple LL experimental data," overcomes the limitations of motion interference and mapping capabilities in traditional LL technology, achieving pixel-by-pixel high-resolution T1 mapping under a single breath-hold, significantly improving efficiency, patient tolerance, and clinical examination efficiency. It balances high accuracy, high resolution, and high efficiency, with small measurement errors, clear image quality, and good patient tolerance. Furthermore, the measurement method of this invention is applicable to both pre- and post-angiography scenarios, quantifying T1 changes in normal myocardium and diseased tissues, exhibiting good compatibility and practicality. It provides a new approach for the early diagnosis, lesion quantification, and prognostic assessment of cardiac diseases, and has broad clinical application prospects. Detailed Implementation

[0017] To enable those skilled in the art to better understand the technical solution of the present invention, the present invention will be further described in detail below with reference to embodiments.

[0018] Due to the excellent reproducibility and accuracy of the MOLLI (Modified Look-Locker Inversion Recovery) sequence, it has gained widespread clinical application. This invention provides a method for measuring longitudinal relaxation time on cardiac magnetic resonance imaging based on MOLLI, comprising the following steps: S1 uses the MOLLI sequence to perform pre- and post-contrast scans. Specifically, in the pre- and post-contrast scans, by setting different initial inversion times TI, three sets of consecutive ECG-triggered LL experiments LL1, LL2, and LL3 are set, and after fusion, a sufficient number of relaxation curve sampling points are obtained. Data is also collected at the end of diastole during the cardiac cycle, with one image acquired per heartbeat. S2 Image processing includes image sorting, curve fitting, and calculation of longitudinal relaxation time T1.

[0019] By specifically optimizing the traditional Look-Locker (LL) technique, data is collected only at the end of diastole when motion is minimal, avoiding image blurring and registration errors caused by cardiac contraction / diastole. Furthermore, addressing the limitation of sampling times imposed by cardiac motion in traditional LL techniques (which use a single inversion pulse followed by multiple consecutive samplings) to obtain relaxation curves, three sets of LL experiments with different initial inversion times (TI) accumulate a sufficient number of relaxation curve sampling points to meet the accuracy requirements of pixel-by-pixel T1 calculation. Through selective sampling and multi-set data fusion, the sampling efficiency of LL technology is preserved while eliminating motion interference and supplementing the data points needed for fitting.

[0020] In step S1, the method of the present invention accumulates sufficient sampling points within a single breath-hold by “concatenating 3 independent LL experiments”. The core difference between the 3 LL experiments is that the initial inversion time (TI) is different. LL experiments with different TIs are equivalent to “starting sampling from different time points in the relaxation process”. In this way, all subsequent sampling points are reordered according to the “cumulative time after inversion” and can be spliced ​​into a complete T1 relaxation curve, providing a data basis for pixel-by-pixel fitting.

[0021] In one specific embodiment, the initial inversion times (TI) for the three LL experiments are as follows: LL1 (TI=100ms): Acquire signals during the early relaxation phase (the initial stage of the magnetization vector recovering positively from the -Z axis); LL2 (TI=200ms): Acquires signals during the relaxation phase; LL3 (TI=350ms): Acquires signals during the later relaxation phase.

[0022] The number of sampling times for each experiment was as follows: LL1 and LL2 each had 3 single-shot readouts, and LL3 had 5 readouts, for a total of 11 sampling points, which met the 6-10 data points required for subsequent three-parameter fitting.

[0023] In an optional embodiment, at least 4 seconds of interference-free magnetization recovery time is set between each LL experiment to ensure that the magnetization vector of the previous experiment is fully restored to thermal equilibrium, avoid inter-group interference, and the recovery time can be adaptively adjusted according to the subject's heart rate.

[0024] Preferably, in the pre- and post-contrast scanning of step S1, the acquisition window is compressed to <200ms to reduce motion interference. Preferably, by utilizing the spatial sensitivity differences of the phased array coils, the amount of data acquired is reduced through "undersampling k-space," compressing the acquisition window of a single readout to <200ms. Specifically, by using a single-excitation balanced SSFP sequence as the readout module, combined with sense coding technology, and setting a reduction factor of 2, the acquisition window is compressed to <200ms.

[0025] Preferably, step S1 further includes delayed enhancement scanning.

[0026] In step S2, data processing mainly involves the conversion from signal intensity to T1 value, that is, calculating the T1 value of each pixel based on the signal intensity acquired in step S1. This mainly includes steps such as image sorting, signal correction, curve fitting, and calculation of the longitudinal relaxation time T1.

[0027] (1) Image sorting mainly involves establishing a "time-signal" correspondence.

[0028] The "cumulative time after reversal" (t) for each sampling point is crucial for calculating T1. Preferably, the sorting formula is: t = TI + (n 1) × RR, where: TI is the initial inversion time of this LL experiment (e.g., 100 / 200 / 350ms); n is the sampling sequence number in this experiment (1-3 or 1-5); RR is the cardiac cycle (the time from one R wave to the next R wave, calculated by ECG). The 11 sampling points from the 3 experiments are sorted in ascending order of t-value to form a complete "relaxation time-signal intensity" curve.

[0029] (2) Signal correction: Corrects signal distortion in amplitude images; During the T1 relaxation process, when the magnetization vector recovers from the -Z axis to the +Z axis, the signal undergoes a change from "negative → zero-crossing → positive," causing the amplitude image to lose polarity information and resulting in fitting bias. Therefore, signal correction is necessary. Existing methods can be used for correction, such as the method proposed by Nekolla et al.: First, generate multiple sets of polarity combination datasets: for example, in the first set, all signals are set to positive; in the second set, the first signal is set to negative and the rest are positive; in the third set, the first two signals are set to negative and the rest are positive, and so on; then select the optimal dataset: perform curve fitting on each dataset and calculate the chi-square value (χ²). 2 The dataset with the lowest chi-square value best reflects the signal changes in the actual relaxation process.

[0030] (3) Curve fitting: Apparent T1* can be obtained by curve fitting. Since the signal intensity change during T1 relaxation follows the exponential recovery law, the Levenberg-Marquardt algorithm is preferably used to perform curve fitting on the "time-signal" curve for each pixel. The specific formula is as follows: y=A B×exp( t / T1*) Where: y is the signal strength; A is the maximum signal strength after relaxation equilibrium; B is the signal dynamic range coefficient (AB is the signal strength at the relaxation start point); t is the cumulative time after reversal; T1* is the "apparent T1" in the LL experiment.

[0031] (4) Calculation of the true T1 value: The true T1 value is obtained by correcting the systematic bias of the apparent T1*.

[0032] Traditional LL techniques have confirmed that there is a fixed conversion relationship between the real T1 and the apparent T1*. Therefore, the method of this invention directly adopts the mature formula: T1=T1 The conversion is performed using ×(AB-1). By fitting the A and B coefficients, the relaxation perturbation deviation caused by the LL technique is corrected, and the true T1 value of each pixel is finally obtained, forming a high-resolution T1 mapping map.

[0033] In summary, the method of this invention integrates three sets of LL experiments with different TI values ​​to obtain multiple sampling points within a single breath-hold, meeting the fitting requirements and solving the problem of insufficient sampling efficiency in traditional methods. Furthermore, it compresses the sampling time, reduces motion interference, and then combines image sorting, signal correction, curve fitting, and T1* correction to obtain a true and accurate T1 value, achieving single breath-hold, high resolution, and high accuracy.

[0034] An embodiment of the present invention provides a computer-readable storage medium storing computer instructions for causing a processor to execute the above-described method for measuring longitudinal relaxation time of cardiac magnetic resonance imaging.

[0035] Embodiments of the present invention also provide a magnetic resonance imaging device that uses the cardiac magnetic resonance longitudinal relaxation time measurement method of the present invention to efficiently and accurately obtain the T1 values ​​of normal myocardium and diseased (such as myocardial infarction) tissues, so as to facilitate the early diagnosis, lesion quantification and prognostic assessment of heart diseases.

[0036] This invention is not limited to the applications listed in the specification and embodiments. It can be applied to various fields suitable for this invention. Without departing from the spirit and essence of this invention, those skilled in the art can easily make other modifications and variations, but these corresponding modifications and variations should all fall within the protection scope claimed by this invention.

[0037] The above description is only a part of the embodiments of the present invention, and is not intended to limit the implementation and protection scope of the present invention. Those skilled in the art should realize that any equivalent substitutions and obvious changes made based on the content of the present invention specification should be included within the protection scope of the present invention.

Claims

1. A method for measuring longitudinal relaxation time on cardiac magnetic resonance imaging, characterized in that, Includes the following steps: S1 uses the MOLLI sequence for pre- and post-contrast scans; In the pre- and post-contrast scans, by setting different initial reversal times TI, three sets of consecutive ECG-triggered LL experiments LL1, LL2, and LL3 were set, and after fusion, a sufficient number of relaxation curve sampling points were obtained. Furthermore, data is collected at the end of diastole during the cardiac cycle, with one image acquired per heartbeat; S2 Image processing includes image sorting, signal correction, curve fitting, and calculation of longitudinal relaxation time T1.

2. The measurement method as described in claim 1, characterized in that, In step S1, during the pre- and post-contrast scans, the acquisition window is compressed to <200ms.

3. The measurement method as described in claim 2, characterized in that, By using a single-shot balanced SSFP sequence as the readout module, combined with sense coding technology, and setting a reduction factor of 2, the acquisition window is compressed to <200ms.

4. The imaging method as described in claim 1, characterized in that, Step S1 also includes delayed enhancement scanning.

5. The measurement method as described in claim 1, characterized in that, In step S2, the images are sorted according to the cumulative time t after reversal, using the following formula: , where n is the image number in the LL experiment and RR is the interval between heartbeats.

6. The measurement method as described in claim 1, characterized in that, In step S2, curve fitting is performed using three-parameter fitting.

7. The measurement method as described in claim 6, characterized in that, In step S2, the Levenberg-Marquardt algorithm is used for three-parameter fitting, and the fitting formula is as follows: Where y is the signal strength, Epigenetics in LL experiments .

8. The measurement method as described in claim 7, characterized in that, Through formula Transformation to obtain reality value.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the method for measuring longitudinal relaxation time of cardiac magnetic resonance imaging as described in any one of claims 1-8.

10. A magnetic resonance imaging device, characterized in that, The longitudinal relaxation time of cardiac magnetic resonance is measured using the method described in any one of claims 1-8.

Images