WO2022188026A1 - 一种磁共振多参数定量方法及其应用 - Google Patents

一种磁共振多参数定量方法及其应用 Download PDF

Info

Publication number
WO2022188026A1
WO2022188026A1 PCT/CN2021/079714 CN2021079714W WO2022188026A1 WO 2022188026 A1 WO2022188026 A1 WO 2022188026A1 CN 2021079714 W CN2021079714 W CN 2021079714W WO 2022188026 A1 WO2022188026 A1 WO 2022188026A1
Authority
WO
WIPO (PCT)
Prior art keywords
echo
water
signal
magnetic resonance
fat
Prior art date
Application number
PCT/CN2021/079714
Other languages
English (en)
French (fr)
Inventor
彭浩
邹超
程传力
刘新
郑海荣
Original Assignee
深圳高性能医疗器械国家研究院有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳高性能医疗器械国家研究院有限公司 filed Critical 深圳高性能医疗器械国家研究院有限公司
Priority to PCT/CN2021/079714 priority Critical patent/WO2022188026A1/zh
Publication of WO2022188026A1 publication Critical patent/WO2022188026A1/zh

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/05—Detecting, measuring or recording for diagnosis by means of electric currents or magnetic fields; Measuring using microwaves or radio waves 
    • A61B5/055—Detecting, measuring or recording for diagnosis by means of electric currents or magnetic fields; Measuring using microwaves or radio waves  involving electronic [EMR] or nuclear [NMR] magnetic resonance, e.g. magnetic resonance imaging

Definitions

  • the present application belongs to the technical field of image imaging, and in particular relates to a multi-parameter quantitative method of magnetic resonance and its application.
  • Ectopic fat is defined as the accumulation of excess fatty acids in non-adipose tissue. Ectopic fat can cause cellular inflammatory response and apoptosis through processes such as oxidative stress. Magnetic resonance water-fat separation technology can accurately quantify fat accumulation, and its accuracy has been verified in related studies. But mere quantification of adipose fat accumulation is not enough clinically, other physiological processes related to fat accumulation are very important for the prognosis and management of the disease.
  • Tamada et al. proposed the variable flip angle gradient echo T 1 quantification method framework, combined with multi-echo acquisition to simultaneously quantify T 1 , PDFF and R 2 *.
  • a 4.5-fold speedup was achieved through parallel imaging and compressed sensing, enabling whole-liver data acquisition to be completed within one breath-hold.
  • the specific contents of the method are as follows: the six-echo gradient echo sequences around the phase are collected separately at different flip angles to form the original data, and the total acquisition time is 18 seconds through acceleration; the multi-echo signals at different flip angles are based on the method proposed by Reeder et al.
  • the IDEAL method calculates the water-fat signal and R 2 *; according to the water signal at different flip angles, the longitudinal relaxation time of the water tissue is inversely calculated.
  • the present application Based on the problems that the water-fat separation results of the existing quantitative technology are unstable and easily affected by the uneven B1 field, the present application provides a magnetic resonance multi-parameter quantitative method and its application.
  • the present application provides a multi-parameter quantitative method for magnetic resonance, which includes the following steps: Step 1: acquire data in sequence, obtain a stimulated echo signal and a free decay signal, according to the stimulated echo signal The wave signal and the free attenuation signal are calculated to obtain the radio frequency transmission field information; Step 2: perform water-fat separation calculation on the data to obtain the water signal under different flip angles; Step 3: According to the water signal and the radio frequency transmission field information to calculate the longitudinal relaxation time of the water tissue.
  • the data in step 1 is multi-echo gradient echo data.
  • step 1 the multi-echo gradient echo data is collected in sequence, and the water-fat separation calculation is performed on the multi-echo gradient echo data to obtain the water signal and the water signal under different flip angles.
  • Main magnetic field information according to the main magnetic field information, the proton density fraction of fat and the quantitative result of T 2 * are obtained by calculation.
  • the sequence includes a first part, a second part and a third part
  • the first part is a double-focus echo acquisition mode sequence
  • the second part is a multi-echo gradient detour echo sequence at a first different flip angle
  • the third part is a multi-echo echo sequence at a second different flip angle.
  • the echo gradient goes around the echo sequence; the echo intervals of the second part and the third part are kept the same.
  • Another implementation manner provided by the present application is: based on the radio frequency emission field calibration method of the double-focus echo acquisition mode sequence, the stimulated echo signal and the free attenuation signal are acquired, and the free attenuation signal is The water signal and the lipid signal are in phase.
  • the water-fat separation is double-flip angle water-fat separation.
  • the water-fat separation includes dividing the candidate solutions of the main magnetic field B 0 field map into a water-dominated solution and a fat-dominated solution;
  • the field map was corrected, and the longitudinal relaxation time was quantified using the water signal at different flip angles.
  • the solution of the field diagram includes a correct solution and an inverse solution, the correct solution corresponds to the inverse solution, and for any combination of equidistant echo times, the The relationship between the correct solution and the inverse solution is determined by the water-fat ratio, and is related to the choice of echo time;
  • ⁇ t is the correct solution
  • ⁇ a is the inverse solution
  • ⁇ TE is the echo interval time
  • f F is the offset of the fat peak relative to the water peak (in ppm)
  • TE 2 is the echo time corresponding to the second echo signal
  • W t and F t are the real water and fat signals, respectively .
  • the present application also provides an application of a magnetic resonance multi-parameter quantification method, which is applied to whole liver multi-parameter quantification.
  • the magnetic resonance multi-parameter quantification method provided in this application proposes an imaging sequence by optimizing the existing commercial gradient echo (GRE) sequence, which can complete one imaging sequence when B1+/main magnetic field B0 is not uniform.
  • GRE commercial gradient echo
  • the magnetic resonance multi-parameter quantification method provided in the present application is a method that can stably complete the quantification of whole liver T 1 /PDFF/T 2 *, while correcting the influence of the B1+/B0 field on the results.
  • the magnetic resonance multi-parameter quantification method provided in the present application uses the radio frequency emission field calibration method to calibrate the B 1 + inhomogeneity, so as to improve the accuracy of T 1 quantification.
  • the magnetic resonance multi-parameter quantification method studied the signal model of the three-point water-fat separation, and obtains the theoretical solution of the field diagram of the three-point water-fat separation.
  • the magnetic resonance multi-parameter quantitative method provided in this application based on the influence of the field diagram theoretical solution formula, proposes a method for resolving the ambiguity of water and fat based on double flip angle information, and completes accurate water and fat in the abdomen and spine. separation.
  • the multi-parameter quantification method of magnetic resonance provided by this application innovatively proposes to introduce the prior information of double flip angle to solve the problem of water-fat ambiguity; at the same time, considering the unevenness of the radio frequency transmission field, the quantification of T 1 is corrected.
  • the application of the magnetic resonance multi-parameter quantification method provided in the present application compared with the quantitative results of T 1 before and after B 1 correction, the method proposed in the present application has a significant improvement.
  • Fig. 1 is the sequence sequence diagram of the magnetic resonance multi-parameter quantitative method of the present application
  • Fig. 2 is the schematic flow sheet of the water-fat separation method of the magnetic resonance multi-parameter quantitative method of the present application
  • Fig. 3 is the schematic diagram of the water-fat mimic experiment result of the present application.
  • FIG. 4 is a schematic diagram of the scanning results of volunteers of the present application.
  • Transverse relaxation is a process in which the magnetic moment gradually disappears from a maximum value in the transverse xy plane. Also called T2 relaxation. Transverse relaxation also takes a long time, so the transverse magnetic moment decays to 37% of the time. transverse relaxation time
  • T1 relaxation refers to the time required for the longitudinal magnetization vector to increase from zero to 63% of its maximum value after a 90° radio frequency pulse, also known as T1 relaxation.
  • T1 relaxation time is one of the important imaging parameters of magnetic resonance imaging. The length of the T1 value is reflected on the fluorescent screen, and it is displayed as the difference between the gray level and the dark.
  • Gradient echo is an echo signal generated by reversing the direction of the gradient field. Gradient echo is also called field echo (field echo). A GRE sequence always begins with an RF pulse less than 90°.
  • T 2 * time is the time required for the magnetization vector intensity to decay from the maximum value to 37% after phase dispersion due to factors such as magnetic field inhomogeneity when the transverse magnetization decays, and the T 2 * of the tissue is less than the T 2 relaxation time of the tissue .
  • the multi-parameter quantitative method of magnetic resonance is often used to obtain abdominal water tissue signals specifically to exclude the influence of fat on subsequent data processing.
  • the present application provides a multi-parameter quantitative method for magnetic resonance, which includes the following steps:
  • Step 1 collect data in sequence, obtain a stimulated echo signal and a free attenuation signal, and calculate and obtain radio frequency emission field information according to the stimulated echo signal and the free attenuation signal;
  • Step 2 perform water-fat separation and calculation on the data to obtain water signals under different flip angles
  • Step 3 Calculate the longitudinal relaxation time of the water tissue according to the water signal and the radio frequency transmission field information.
  • the data in the step 1 is multi-echo gradient echo data.
  • the multi-echo gradient echo data is collected in sequence, and the water-fat separation calculation is performed on the multi-echo gradient echo data to obtain the water signal and main magnetic field information under different flip angles.
  • the magnetic field information was calculated to yield the fat proton density fraction and T2* quantification.
  • the sequence includes a first part, a second part and a third part; the first part is a double-focus echo acquisition mode sequence, and the second part is the multi-echo gradient detours under the first different flip angles
  • the echo sequence, the third part is a multi-echo gradient reversing echo sequence under the second different flip angle; the echo interval of the second part and the third part are consistent.
  • the sequence consists of the following three parts: A: B 1 + field calibration method based on DREAM sequence, acquisition of stimulated echo signal (STE) and free decay signal (FID), and water and lipid signals in FID same phase. B and C: Multi-echo gradient reversing echo sequences at different flip angles. Signals acquired at equal intervals (interval time ⁇ TE) were read out using unipolar. The echo spacing in B and C remains the same, but the second echo time is shifted back by ⁇ TE/2. The layer selection gradient and phase encoding gradient are omitted here.
  • the RF emission field B 1+ can be obtained by the following formula:
  • S STE* and S FID are respectively the complex conjugate and free decay signal of the collected stimulated echo signal, which are collected from part A in the sequence diagram 1. Therefore, when the set flip angle is ⁇ , the actual flip angle size is B 1 ⁇ .
  • the double flip angle water-fat separation method proposed in this application (such as the subsequent 'water-fat separation' section and Fig. 2 described) to complete the separation of water and fat, and obtain the water signals W 1 and W 2 under different flip angles.
  • the longitudinal relaxation time constant T 1 is obtained by fitting the following formula:
  • the signal model is as follows:
  • W and F are the amplitudes of the water-fat signal, respectively, and ⁇ is the main magnetic field inhomogeneity.
  • Signals at low flip angles can be used to estimate the proton-density fat fraction with a suitable water-fat separation method:
  • the ratio of the water-fat signal is also affected by the flip angle, and T 1 is also encoded into the water-fat signal at the same time (considering the B 1 inhomogeneity):
  • T1 fitting problem can be transformed into a linear fitting problem:
  • Equation (4) Equation (4) can be transformed into:
  • ⁇ a is also the solution of formula (4), but the final result of water and fat will be opposite, which is the classic water-fat ambiguity problem in the water-fat separation problem.
  • the relationship between ⁇ a and ⁇ t is determined by the water-fat ratio and is related to the choice of TE.
  • the real solution P t of the field is fixed, but it can be seen from formula (11) that the decomposition and inverse solutions of the field map will be inconsistent at different flip angles due to the different ratios of the water-fat signals. which is:
  • the stimulated echo signal and the free attenuation signal are acquired, and the water signal and the lipid signal in the free attenuation signal are in phase.
  • the water-fat separation is double-flip angle water-fat separation.
  • the water-fat separation includes dividing the candidate solutions of the main magnetic field B 0 field map into a water-dominated solution and a fat-dominated solution;
  • the first type of pixel point is defined as a sub-region according to the set formed by spatial continuity, and the correct field diagram of the current region is determined according to the method for "regional solution field diagram ambiguity" described later;
  • the field map was corrected, and the longitudinal relaxation time was quantified using the water signal at different flip angles.
  • Equation (11) for the signal model of Equation (4), there are two sets of solutions in most cases. Changing the echo time combination and the flip angle will bring about the change of the decomposition and inverse solution P a .
  • Two sets of solutions for the signal model (4) can be calculated, but it is not directly known which solution is the correct solution for the field diagram. Dividing the candidate solutions of the field map into two categories, Pw and Pf , results in water-dominated and fat-dominated solutions, respectively.
  • k is the pixel index in the 3D direction, and
  • D f (r) is defined in the same way.
  • the pixel r is defined as "the first type of pixel", and the rest of the pixels are defined It is "the second type of pixel”. All “first-class pixels” are shown in Figure 2c, and such pixels will be preferentially processed.
  • the "first-type pixel points” are clustered based on spatial continuity to reduce the amount of calculation and improve the stability of the method.
  • the set of pixels of the first type formed by spatial continuity is defined as a sub-region.
  • the field diagrams for these regions can all be derived from Pw and Pf simultaneously. Using the information in equation (12), the field map ambiguity of the entire region can be resolved. when:
  • the field diagram of the remaining pixels is solved by region growth, taking into account the six-peak fat model and Attenuation, further correction of the field map.
  • the water signal at different flip angles was further applied for T 1 quantification.
  • the solution of the field diagram includes a correct solution and a split-inverse solution
  • the correct solution corresponds to the split-inverse solution
  • the correct solution and the split-inverse solution are The relationship between them is determined by the water-fat ratio and is related to the choice of echo time; the decomposition and inverse solutions are inconsistent at different flip angles.
  • ⁇ t is the correct solution
  • ⁇ a is the inverse solution
  • ⁇ TE is the echo interval time
  • f F is the offset of the fat peak relative to the water peak (in ppm)
  • TE 2 is the echo time corresponding to the second echo signal
  • W t and F t are the real water and fat signals, respectively .
  • the present application also provides an application of a magnetic resonance multi-parameter quantification method, which is applied to whole liver multi-parameter quantification.
  • a and b are the results of fat quantification, and the method of the present application has a high consistency with the existing method.
  • c, d are the results of T 1 quantification, and the method of the present application and the T 1 quantification method based on inversion recovery have also achieved relatively consistent results.
  • the method of the present application can complete multi-parameter magnetic resonance imaging in multiple parts of the human body.
  • the water-fat separation method described in this application can not only be applied to the double-flip angle triple-echo GRE sequence, but also can be applied to other quantitative sequences under the premise of retaining the method framework.
  • the method of the present application is calibrated for B 1 + inhomogeneity by using the radio frequency emission field calibration method to improve the accuracy of T 1 quantification; the method of the present application is studied for the three-point water-fat separation signal model, and obtained Field diagram theoretical solution of three-point water-fat separation; based on the influence of flip angle on the formula of field diagram theoretical solution, the present application proposes a method for resolving water-fat ambiguity based on double flip angle information, and completes it in the abdomen and spine accurate water-fat separation.

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Physics & Mathematics (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Engineering & Computer Science (AREA)
  • Medical Informatics (AREA)
  • Biophysics (AREA)
  • Pathology (AREA)
  • High Energy & Nuclear Physics (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Radiology & Medical Imaging (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • General Health & Medical Sciences (AREA)
  • Public Health (AREA)
  • Veterinary Medicine (AREA)
  • Magnetic Resonance Imaging Apparatus (AREA)

Abstract

本申请属于图像成像技术领域，特别是涉及一种磁共振多参数定量方法及其应用。现有的方法压缩感知重建计算复杂，只能离线完成；而且没有考虑射频发射场B1+不均匀对T1定量的影响。本申请提供了一种磁共振多参数定量方法，所述方法包括如下步骤：步骤1：序列采集数据，得到受激回波信号和自由衰减信号，根据所述受激回波信号和所述自由衰减信号计算得到射频发射场信息；步骤2：对所述数据进行水脂分离计算得到不同翻转角下的水信号；步骤3：根据所述水信号和所述射频发射场信息计算水组织的纵向弛豫时间。利用射频发射场校准方法针对B1 +不均匀性进行了校准，以提高T1定量的准确性。

Description

一种磁共振多参数定量方法及其应用 技术领域
本申请属于图像成像技术领域，特别是涉及一种磁共振多参数定量方法及其应用。
背景技术
异位脂肪的定义是过量的脂肪酸在非脂肪组织中的积累。异位脂肪会通过氧化应激反应等过程造成细胞的炎性反应以及细胞凋亡。磁共振水脂分离技术可以准确的定量脂肪囤积，并且其准确性已在相关研究中已得到了验证。但是仅仅定量脂肪脂肪囤积在临床上是不够的，其他与脂肪囤积相关的生理过程对于疾病的预后和管理相当重要。
近年来有许多研究利用多参数的磁共振成像，包括脂肪质子密度分数(proton density fat fraction，PDFF)，纵向弛豫时间T 1，T 2*，以全面描述脂肪囤积以及其他的生理变化。腹部多参数磁共振成像的挑战主要在于以下三点：单次屏息内全肝覆盖，高场(3T及以上)射频发射场B 1+不均匀性，水脂二义性。单次屏息内完成全肝覆盖对扫描时间提出了较高的要求，射频发射场B 1+不均匀性会决定纵向弛豫时间T 1定量的精度，而水脂分离的准确性是方法后续定量正确的基础。
腹部多参数磁共振成像已有较多的相关研究。Tamada等人在2018年提出了可变翻转角梯度回波T 1定量方法框架下，结合多回波采集以同时定量T 1，PDFF和R 2*。通过并行成像和压缩感知的方法完成了4.5倍的加速，使得全肝数据采集可以在一次屏息内完成。方法具体内容如下：不同翻转角下六回波绕相梯度回波序列分别采集以形成原始数据，通过加速使得总采集时间为18秒；不同翻转角下的多回波信号根据Reeder等人提出的IDEAL方法算出水脂信号以及R 2*；依据不同翻转角下的水信号反算出水组织的纵向弛豫时间。
但是现有的方法压缩感知重建计算复杂，只能离线完成；而且没有考虑射频发射场B 1+不均匀对T 1定量的影响。
发明内容
1.要解决的技术问题
基于现有定量技术水脂分离结果不稳定且易受到B1场不均匀影响的问题，本申请提供了一种磁共振多参数定量方法及其应用。
2.技术方案
为了达到上述的目的，本申请提供了一种磁共振多参数定量方法，所述方法包括如下步骤：步骤1：序列采集数据，得到受激回波信号和自由衰减信号，根据所述受激回波信号和 所述自由衰减信号计算得到射频发射场信息；步骤2：对所述数据进行水脂分离计算得到不同翻转角下的水信号；步骤3：根据所述水信号和所述射频发射场信息计算水组织的纵向弛豫时间。
本申请提供的另一种实施方式为：所述步骤1中数据为多回波梯度回波数据。
本申请提供的另一种实施方式为：所述步骤1中序列采集多回波梯度回波数据，对所述多回波梯度回波数据进行水脂分离计算得到不同翻转角下的水信号和主磁场信息，根据所述主磁场信息计算得到脂肪质子密度分数和T 2*定量结果。
本申请提供的另一种实施方式为：所述序列包括第一部分、第二部分和第三部分；
所述第一部分为双聚焦回波采集模式序列，所述第二部分为第一不同翻转角下的多回波梯度绕向回波序列，所述第三部分为第二不同翻转角下的多回波梯度绕向回波序列；所述第二部分和所述第三部分的回波间隔保持一致。
本申请提供的另一种实施方式为：基于所述双聚焦回波采集模式序列的射频发射场校准方法，采集所述受激回波信号和所述自由衰减信号，并且所述自由衰减信号中水信号和脂信号同相位。
本申请提供的另一种实施方式为：所述水脂分离为双翻转角水脂分离。
本申请提供的另一种实施方式为：所述水脂分离包括将主磁场B 0场图的候选解分为水占主导的解和脂肪占主导的解；
将像素点分为第一类像素点和第二类像素点，对所述第一类像素点以空间连续性为依据进行聚类；
将所述第一类像素点依据空间连续性形成的集合定义为子区域，根据所述子区域确定当前区域的正确场图解；
对场图进行修正，采用不同翻转角下的水信号进行纵向弛豫时间定量。
本申请提供的另一种实施方式为：所述场图的解包括正确解和分反解，所述正确解与所述分反解相对应，对于任意的等间距回波时间组合，所述正确解与所述分反解之间的关系由水脂比例决定，并且与回波时间的选择有关；
在不同翻转角下，所述分反解不一致。
本申请提供的另一种实施方式为：所述正确解与所述分反解可以用以下公式表示：
Figure PCTCN2021079714-appb-000001
其中，ψ t为正确解，ψ a为分反解，
Figure PCTCN2021079714-appb-000002
ΔTE为回 波间隔时间，f F脂肪峰相对水峰的偏移(单位ppm)，TE 2为第二个回波信号对应的回波时间，W t和F t分别为真实的水和脂肪信号。
本申请还提供一种磁共振多参数定量方法的应用，将所述的磁共振多参数定量方法应用于全肝多参数定量。
3.有益效果
与现有技术相比，本申请提供的一种磁共振多参数定量方法的有益效果在于：
本申请提供的磁共振多参数定量方法，通过对现有商用梯度回波(gradient echo，GRE)序列进行优化，提出一种成像序列，能够在B1+/主磁场B0不均匀的情况下，完成一次屏息内的全肝T1/PDFF/T2*定量。
本申请提供的磁共振多参数定量方法，为一种能够稳定完成全肝T 1/PDFF/T 2*定量的方法，同时校正B1+/B0场对结果的影响。
本申请提供的磁共振多参数定量方法，利用射频发射场校准方法针对B 1 +不均匀性进行了校准，以提高T 1定量的准确性。
本申请提供的磁共振多参数定量方法，针对三点水脂分离信号模型进行了研究，并得出了三点水脂分离的场图理论解。
本申请提供的磁共振多参数定量方法，基于场图理论解公式的影响，提出了一种基于双翻转角信息解水脂二义性的方法，并且在腹部和脊柱中完成了准确的水脂分离。
本申请提供的磁共振多参数定量方法，创新性的提出了引入双翻转角先验信息解决水脂二义性的问题；同时考虑了射频发射场的不均匀，对T 1定量进行了校正。
本申请提供的磁共振多参数定量方法的应用，对比B 1校正前后的T 1定量结果，本申请提出的方法有着明显的提升。
附图说明
图1是本申请的磁共振多参数定量方法的序列时序示意图；
图2是本申请的磁共振多参数定量方法的水脂分离方法流程示意图；
图3是本申请的水脂仿体实验结果示意图；
图4是本申请的志愿者扫描结果示意图。
具体实施方式
在下文中，将参考附图对本申请的具体实施例进行详细地描述，依照这些详细的描述，所属领域技术人员能够清楚地理解本申请，并能够实施本申请。在不违背本申请原理的情况下，各个不同的实施例中的特征可以进行组合以获得新的实施方式，或者替代某些实施例中 的某些特征，获得其它优选的实施方式。
横向弛豫是在横向xy平面上，磁矩由最大值逐渐消失的过程。也称T 2弛豫。横向弛豫也需要很长时间，所以在横向磁矩衰减到37％的时间。称为横向弛豫时间
纵向弛豫指90°射频脉冲后纵向磁化矢量由零增长到它的最大值的63％所需要的时间,又称T 1弛豫，T 1弛豫时间是磁共振成像的重要成像参数之一。T 1值长短反映在荧光屏上，显示为灰度明暗的差异，T 1时间短呈亮的灰度，T 1时间长则呈暗淡的灰度。
IDEAL水脂分离方法的稳定性有待提升。
梯度回波(GRE，gradient echo)就是通过有关梯度场方向的翻转而产生的回波信号。梯度回波又叫场回波(filed echo)，它与自旋回波的主要区别在于二者产生回波的激励方式不同。GRE序列总是以一个小于90°的RF脉冲开始。
T 2*时间为横向磁化衰减时由于磁场不均匀度等因素导致更快散相后磁化矢量强度由最大值衰减到37％所需的时间，组织的T 2*小于组织的T 2弛豫时间。
在腹部参数定量中，经常会利用磁共振多参数定量方法特异性的获取腹部水组织信号，以排除脂肪对于后续数据处理的影响。
参见图1～4，本申请提供一种磁共振多参数定量方法，所述方法包括如下步骤：
步骤1：序列采集数据，得到受激回波信号和自由衰减信号，根据所述受激回波信号和所述自由衰减信号计算得到射频发射场信息；
步骤2：对所述数据进行水脂分离计算得到不同翻转角下的水信号；
步骤3：根据所述水信号和所述射频发射场信息计算水组织的纵向弛豫时间。
进一步地，所述步骤1中数据为多回波梯度回波数据。
进一步地，所述步骤1中序列采集多回波梯度回波数据，对所述多回波梯度回波数据进行水脂分离计算得到不同翻转角下的水信号和主磁场信息，根据所述主磁场信息计算得到脂肪质子密度分数和T 2*定量结果。
进一步地，所述序列包括第一部分、第二部分和第三部分；所述第一部分为双聚焦回波采集模式序列，所述第二部分为第一不同翻转角下的多回波梯度绕向回波序列，所述第三部分为第二不同翻转角下的多回波梯度绕向回波序列；所述第二部分和所述第三部分的回波间隔保持一致。
如图1所示，序列包含以下三个部分：A:基于DREAM序列的B 1+场校准方法，采集受激回波信号(STE)和自由衰减信号(FID)，并且FID中水、脂信号同相位。B和C:不同翻转角下的多回波梯度绕向回波序列。等间距采集(间隔时间为ΔTE)的信号使用单极读出。B与C 中的回波间隔保持一致，但第二个回波时间会往后挪ΔTE/2。层选梯度和相位编码梯度此处省略。
信号模型
射频发射场B 1+可以通过以下公式得出：
Figure PCTCN2021079714-appb-000003
其中S STE*和S FID分别为采集到的受激回波信号的复共轭和自由衰减信号，由序列图1中的A部分采集得到。因而当设定的翻转角为θ时，实际的翻转角大小为B 1·θ。
对于由图1中的B-C部分采集得到的，不同翻转角下的梯度绕向回波序列信号，使用本申请所提出的双翻转角水脂分离方法(如后续‘水脂分离’部分以及图2所述)，完成水脂分离，得到不同翻转角下的水信号W 1和W 2。通过下面所述公式，拟合得到纵向弛豫时间常数T 1：
Figure PCTCN2021079714-appb-000004
Figure PCTCN2021079714-appb-000005
对等间距采样(回波时间[TE 2-ΔTE,TE 2,TE 2+ΔTE])的信号，其信号模型如下所示：
Figure PCTCN2021079714-appb-000006
其中W和F分别为水脂信号的幅度，ψ为主磁场不均匀性。通过合适的水脂分离方法，低翻转角下的信号可以用来估计质子密度脂肪分数：
Figure PCTCN2021079714-appb-000007
双翻转角采集中，水脂信号的比例同时也会受到翻转角的影响，而T 1也会同时编码到水脂信号中(考虑B 1不均匀性)：
Figure PCTCN2021079714-appb-000008
Figure PCTCN2021079714-appb-000009
其中TR为重复时间，M w和M f与水脂稳态纵轴磁化矢量成正比；B 1为射频发射场不均匀；θ为翻转角。T 1拟合问题可以被转换为线性拟合问题：
Figure PCTCN2021079714-appb-000010
其中C为常数项。只要计算得到E 1，纵向磁化矢量可以通过以下公式表达：
T 1＝-TR/ln(E 1)  (9)
水脂二义性问题
需首先采集两组不同翻转角下的多回波信号。对等间距采样的信号公式(4)中，令
Figure PCTCN2021079714-appb-000011
公式(4)可以转化为：
Figure PCTCN2021079714-appb-000012
以上公式为基于对称回波时间[-ΔTE,0,ΔTE]的信号模型正确解分反解之间的理论关系式。因此，对于任意的等间距回波时间组合，假设ψ t为场不均匀性的正确解，与之对应的分反解可用以下公式表示：
Figure PCTCN2021079714-appb-000013
ψ a也是公式(4)的解，但会造成最终的水脂结果相反，此即为水脂分离问题中经典的水脂二义性问题。ψ a与ψ t之间的关系由水脂比例决定，并且与TE的选择有关。为了避免解卷绕的问题，相位因子P＝e iψ将在后续部分中取代场ψ。场的真实解P t固定，但是由公式(11)可知，在不同翻转角下由于水脂信号比例不同，场图的分反解会不一致。即：
P t(θ 1)＝P t(θ 2)
P a(θ 1)≠P a(θ 2)      (12)
进一步地，基于所述双聚焦回波采集模式序列的射频发射场校准方法，采集所述受激回波信号和所述自由衰减信号，并且所述自由衰减信号中水信号和脂信号同相位。
进一步地，所述水脂分离为双翻转角水脂分离。引入双翻转角的先验信息解决水脂二义性的水脂分离方法。
进一步地，所述水脂分离包括将主磁场B 0场图的候选解分为水占主导的解和脂肪占主导的解；
将像素点分为第一类像素点和第二类像素点，对所述第一类像素点以空间连续性为依据进行聚类；
将所述第一类像素点依据空间连续性形成的集合定义为子区域，根据后续所述“区域性 解场图二义性”的方法确定当前区域的正确场图解；
对场图进行修正，采用不同翻转角下的水信号进行纵向弛豫时间定量。
如图2所示，场图候选解计算：
如公式(11)所示，对于公式(4)的信号模型，绝大多数情况下都有两组解。改变回波时间组合以及翻转角，均会带来分反解P a的变化。信号模型(4)的两组解可算出，但无法直接得知哪个解是场图的正确解。将场图的候选解分为两类，P w和P f，分别会得到水占主导和脂肪占主导的解。如图2b中所示的”Candidate Phasor Calculation”部分。
像素点分类：
所有像素点依据以下标准分为两类：1.像素点与其3D空间内的6领域的场图解光滑；2.6领域内至少有一个像素点的场图解与当前像素点存在跳变。定义P w解中，6领域内最大的场图变化为D w(r)：
D w(r)＝max k{|angle[P w(r)×conj(P w(r k))]|}  (13)
k为3D方向6领域内的像素点索引，|·|表示取绝对值，D f(r)以相同的方式定义。当某个像素点不同翻转角下的D w(r)和D f(r)均小于一个给定的阈值D T，定义像素点r为”第一类像素点”，其余的像素点被定义为”第二类像素点”。所有的”第一类像素点”如图2c所示，这类像素点会被优先处理。通过MATLAB命令”conncomp”，对”第一类像素点”以空间连续性为依据进行了聚类，以减少计算量，并提升方法稳定性。
区域性解场图二义性：
第一类像素点依据空间连续性形成的集合被定义为子区域。这些区域的场图解可以全部同时来自于P w和P f。利用(12)式中的信息，可以解整个区域的场图二义性。当：
Figure PCTCN2021079714-appb-000014
其中Θ j为第j个子区域内的所有像素点，对子区域Θ j的像素点，即选择P w为场图解：
Figure PCTCN2021079714-appb-000015
反之亦然。
场图修正及水脂分离：
余下像素点的场图解通过区域增长的方式解决，并考虑六峰脂肪模型及
Figure PCTCN2021079714-appb-000016
衰减，对场图进一进修正。不同翻转角下的水信号被进一步应用于T 1定量。
进一步地，所述场图的解包括正确解和分反解，所述正确解与所述分反解相对应，对于任意的等间距回波时间组合，所述正确解与所述分反解之间的关系由水脂比例决定，并且与回波时间的选择有关；在不同翻转角下，所述分反解不一致。
进一步地，所述正确解与所述分反解可以用以下公式表示：
Figure PCTCN2021079714-appb-000017
其中，ψ t为正确解，ψ a为分反解，
Figure PCTCN2021079714-appb-000018
ΔTE为回波间隔时间，f F脂肪峰相对水峰的偏移(单位ppm)，TE 2为第二个回波信号对应的回波时间，W t和F t分别为真实的水和脂肪信号。
本申请还提供一种磁共振多参数定量方法的应用，将所述的磁共振多参数定量方法应用于全肝多参数定量。
为了验证所提出的方法，本申请在水脂仿体以及志愿者中均进行了实验。如图3所示，在水脂仿体中，用六回波水脂分离方法TREE[7]做对比，验证方法脂肪定量的准确性；对于T 1定量，用反转恢复快速回波序列(IR-fSE)结合水脂分离方法进行了验证。B-A分析及线性回归拟合结果均证明了方法的准确性。
此外，还征集了两名志愿者，扫描了腹部及脊柱，并完成了一次屏息内的全肝多参数定量。定量结果如图4所示。
在仿体及在体实验中，均得到了准确的结果。
如图3所示a，b为脂肪定量结果，本申请方法与现有方法具有较高的一致性。c，d为T 1定量结果，本申请方法与基于反转恢复的T 1定量方法亦取得了较为一致的结果。
如图4所示对比B1校正前后的T 1定量结果，本申请方法有着明显的提升。
如结果所示，本申请方法可以在人体多个部位完成多参数磁共振成像。本申请所述水脂分离方法不仅可以用于双翻转角三回波GRE序列，亦可以在保留方法框架的前提下应用于其他定量序列。
本申请的方法利用射频发射场校准方法针对B 1+不均匀性进行了校准，以提高T 1定量的准确性；本本申请的方法针对三点水脂分离信号模型进行了研究，并得出了三点水脂分离的场图理论解；基于翻转角对场图理论解公式的影响，本申请提出了一种基于双翻转角信息解水脂二义性的方法，并且在腹部和脊柱中完成了准确的水脂分离。
尽管在上文中参考特定的实施例对本申请进行了描述，但是所属领域技术人员应当理解，在本申请公开的原理和范围内，可以针对本申请公开的配置和细节做出许多修改。本申请的保护范围由所附的权利要求来确定，并且权利要求意在涵盖权利要求中技术特征的等同物文字意义或范围所包含的全部修改。

Claims (10)

  1. 一种磁共振多参数定量方法，其特征在于：所述方法包括如下步骤：
    步骤1：序列采集数据，得到受激回波信号和自由衰减信号，根据所述受激回波信号和所述自由衰减信号计算得到射频发射场信息；
    步骤2：对所述数据进行水脂分离计算得到不同翻转角下的水信号；
    步骤3：根据所述水信号和所述射频发射场信息计算水组织的纵向弛豫时间。
  2. 如权利要求1所述的磁共振多参数定量方法，其特征在于：所述步骤1中数据为多回波梯度回波数据。
  3. 如权利要求2所述的磁共振多参数定量方法，其特征在于：所述步骤1中序列采集多回波梯度回波数据，对所述多回波梯度回波数据进行水脂分离计算得到不同翻转角下的水信号和主磁场信息，根据所述主磁场信息计算得到脂肪质子密度分数和T 2*定量结果。
  4. 如权利要求1所述的磁共振多参数定量方法，其特征在于：所述序列包括第一部分、第二部分和第三部分；
    所述第一部分为双聚焦回波采集模式序列，所述第二部分为第一不同翻转角下的多回波梯度绕向回波序列，所述第三部分为第二不同翻转角下的多回波梯度绕向回波序列；所述第二部分和所述第三部分的回波间隔保持一致。
  5. 如权利要求4所述的磁共振多参数定量方法，其特征在于：基于所述双聚焦回波采集模式序列的射频发射场校准方法，采集所述受激回波信号和所述自由衰减信号，并且所述自由衰减信号中水信号和脂信号同相位。
  6. 如权利要求1～5中任一项所述的磁共振多参数定量方法，其特征在于：所述水脂分离为双翻转角水脂分离。
  7. 如权利要求6所述的磁共振多参数定量方法，其特征在于：所述水脂分离包括将主磁场B 0场图的候选解分为水占主导的解和脂肪占主导的解；
    将像素点分为第一类像素点和第二类像素点，对所述第一类像素点以空间连续性为依据进行聚类；
    将所述第一类像素点依据空间连续性形成的集合定义为子区域，根据所述子区域确定当前区域的正确场图解；
    对场图进行修正，采用不同翻转角下的水信号进行纵向弛豫时间定量。
  8. 如权利要求6所述的磁共振多参数定量方法，其特征在于：所述场图的解包括正确解和分反解，所述正确解与所述分反解相对应，对于任意的等间距回波时间组合，所述正确解与所述分反解之间的关系由水脂比例决定，并且与回波时间的选择有关；
    在不同翻转角下，所述分反解不一致。
  9. 如权利要求8所述的磁共振多参数定量方法，其特征在于：所述正确解与所述分反解可以用以下公式表示：
    Figure PCTCN2021079714-appb-100001
    其中，ψ t为正确解，ψ a为分反解，
    Figure PCTCN2021079714-appb-100002
    ΔTE为回波间隔时间，f F脂肪峰相对水峰的偏移(单位ppm)，TE 2为第二个回波信号对应的回波时间，W t和F t分别为真实的水和脂肪信号。
  10. 一种磁共振多参数定量方法的应用，其特征在于：将权利要求1～9中任一项所述的磁共振多参数定量方法应用于全肝、胰腺、脾脏、脑多参数定量。
PCT/CN2021/079714 2021-03-09 2021-03-09 一种磁共振多参数定量方法及其应用 WO2022188026A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2021/079714 WO2022188026A1 (zh) 2021-03-09 2021-03-09 一种磁共振多参数定量方法及其应用

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2021/079714 WO2022188026A1 (zh) 2021-03-09 2021-03-09 一种磁共振多参数定量方法及其应用

Publications (1)

Publication Number Publication Date
WO2022188026A1 true WO2022188026A1 (zh) 2022-09-15

Family

ID=83227349

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/079714 WO2022188026A1 (zh) 2021-03-09 2021-03-09 一种磁共振多参数定量方法及其应用

Country Status (1)

Country Link
WO (1) WO2022188026A1 (zh)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108802648A (zh) * 2018-04-03 2018-11-13 上海东软医疗科技有限公司 一种基于梯度回波的磁共振定量成像方法和装置
EP3531154A1 (en) * 2018-02-22 2019-08-28 Koninklijke Philips N.V. Dixon mr imaging using a multi-gradient-echo sequence
CN110786854A (zh) * 2019-11-05 2020-02-14 广州互云医院管理有限公司 一种水脂混合体系下的反转回复序列t1测量方法
CN112435266A (zh) * 2020-11-10 2021-03-02 中国科学院深圳先进技术研究院 一种图像分割方法、终端设备及计算机可读存储介质

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP3531154A1 (en) * 2018-02-22 2019-08-28 Koninklijke Philips N.V. Dixon mr imaging using a multi-gradient-echo sequence
CN108802648A (zh) * 2018-04-03 2018-11-13 上海东软医疗科技有限公司 一种基于梯度回波的磁共振定量成像方法和装置
CN110786854A (zh) * 2019-11-05 2020-02-14 广州互云医院管理有限公司 一种水脂混合体系下的反转回复序列t1测量方法
CN112435266A (zh) * 2020-11-10 2021-03-02 中国科学院深圳先进技术研究院 一种图像分割方法、终端设备及计算机可读存储介质

Similar Documents

Publication Publication Date Title
CN113017596B (zh) 一种磁共振多参数定量方法及其应用
Budde et al. Ultra-high resolution imaging of the human brain using acquisition-weighted imaging at 9.4 T
Srinivasan et al. TE-averaged two-dimensional proton spectroscopic imaging of glutamate at 3 T
Foxley et al. Improving diffusion-weighted imaging of post-mortem human brains: SSFP at 7 T
Sung et al. Measurement and characterization of RF nonuniformity over the heart at 3T using body coil transmission
US10371779B2 (en) Apparatus and method for magnetic resonance imaging with high spatial temporal resolutions
US9008399B2 (en) System and method for joint reconstruction of spatial and spectral imaging data
WO2020037814A1 (zh) 基于多板块同时激发的等体素磁共振扩散成像方法及装置
CA2964945A1 (en) System and method for producing distortion free magnetic resonance images using dual-echo echo-planar imaging
US20110025327A1 (en) Method for radiofrequency mapping in magnetic resonance imaging
Ho et al. Placental magnetic resonance imaging in chronic hypertension: a case-control study
Lévy et al. Intravoxel Incoherent Motion at 7 Tesla to quantify human spinal cord perfusion: Limitations and promises
US11796620B2 (en) Free-breathing MRI with motion compensation
Wang et al. A referenceless Nyquist ghost correction workflow for echo planar imaging of hyperpolarized [1‐13C] pyruvate and [1‐13C] lactate
WO2018184056A1 (en) Magnetic resonance imaging method and apparatus
WO2022188026A1 (zh) 一种磁共振多参数定量方法及其应用
Murata et al. Effect of hybrid of compressed sensing and parallel imaging on the quantitative values measured by 3D quantitative synthetic MRI: A phantom study
WO2020214725A1 (en) System and method for free-breathing quantitative multiparametric mri
US10228434B2 (en) Multi-shot echo planar imaging using reordered segments and RF excitation pulse phase and slice profiles matched across interleaves
Strasser et al. Improving D‐2‐hydroxyglutarate MR spectroscopic imaging in mutant isocitrate dehydrogenase glioma patients with multiplexed RF‐receive/B0‐shim array coils at 3 T
CN109814056B (zh) 一种获取磁共振精准定量图像的方法
Belsley et al. Optimal flip angles for in vivo liver 3D T1 mapping and B1+ mapping at 3T
US20190254596A1 (en) Method for generating a magnetic resonance image dataset, computer program product, data medium, and magnetic resonance system
Milford et al. A novel method for T2 quantification in presence of B1 inhomogeneities
WO2020223567A1 (en) Simultaneous multi-slice mrsi using density weighted concentric ring acquisition

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21929508

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21929508

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 23.01.2024)