WO2020228641A1 - 一种超短回波时间磁共振指纹弛豫时间测量方法 - Google Patents

一种超短回波时间磁共振指纹弛豫时间测量方法 Download PDF

Info

Publication number
WO2020228641A1
WO2020228641A1 PCT/CN2020/089424 CN2020089424W WO2020228641A1 WO 2020228641 A1 WO2020228641 A1 WO 2020228641A1 CN 2020089424 W CN2020089424 W CN 2020089424W WO 2020228641 A1 WO2020228641 A1 WO 2020228641A1
Authority
WO
WIPO (PCT)
Prior art keywords
time
magnetic resonance
signal
fingerprint
echo
Prior art date
Legal status (The legal status 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 status listed.)
Ceased
Application number
PCT/CN2020/089424
Other languages
English (en)
French (fr)
Inventor
何宏建
李庆
叶慧慧
曹笑之
钟健晖
丁秋萍
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhejiang University ZJU
Original Assignee
Zhejiang University ZJU
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 Zhejiang University ZJU filed Critical Zhejiang University ZJU
Priority to US17/263,549 priority Critical patent/US11353531B2/en
Publication of WO2020228641A1 publication Critical patent/WO2020228641A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R33/00—Arrangements or instruments for measuring magnetic variables
    • G01R33/20—Arrangements or instruments for measuring magnetic variables involving magnetic resonance
    • G01R33/44—Arrangements or instruments for measuring magnetic variables involving magnetic resonance using nuclear magnetic resonance [NMR]
    • G01R33/48—NMR imaging systems
    • G01R33/4816—NMR imaging of samples with ultrashort relaxation times such as solid samples, e.g. MRI using ultrashort TE [UTE], single point imaging, constant time imaging
    • 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
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/45—For evaluating or diagnosing the musculoskeletal system or teeth
    • A61B5/4504—Bones
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R33/00—Arrangements or instruments for measuring magnetic variables
    • G01R33/20—Arrangements or instruments for measuring magnetic variables involving magnetic resonance
    • G01R33/44—Arrangements or instruments for measuring magnetic variables involving magnetic resonance using nuclear magnetic resonance [NMR]
    • G01R33/448—Relaxometry, i.e. quantification of relaxation times or spin density
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R33/00—Arrangements or instruments for measuring magnetic variables
    • G01R33/20—Arrangements or instruments for measuring magnetic variables involving magnetic resonance
    • G01R33/44—Arrangements or instruments for measuring magnetic variables involving magnetic resonance using nuclear magnetic resonance [NMR]
    • G01R33/46—NMR spectroscopy
    • G01R33/4641—Sequences for NMR spectroscopy of samples with ultrashort relaxation times such as solid samples
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R33/00—Arrangements or instruments for measuring magnetic variables
    • G01R33/20—Arrangements or instruments for measuring magnetic variables involving magnetic resonance
    • G01R33/44—Arrangements or instruments for measuring magnetic variables involving magnetic resonance using nuclear magnetic resonance [NMR]
    • G01R33/48—NMR imaging systems
    • G01R33/4818—MR characterised by data acquisition along a specific k-space trajectory or by the temporal order of k-space coverage, e.g. centric or segmented coverage of k-space
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R33/00—Arrangements or instruments for measuring magnetic variables
    • G01R33/20—Arrangements or instruments for measuring magnetic variables involving magnetic resonance
    • G01R33/44—Arrangements or instruments for measuring magnetic variables involving magnetic resonance using nuclear magnetic resonance [NMR]
    • G01R33/48—NMR imaging systems
    • G01R33/50—NMR imaging systems based on the determination of relaxation times, e.g. T1 measurement by IR sequences; T2 measurement by multiple-echo sequences
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R33/00—Arrangements or instruments for measuring magnetic variables
    • G01R33/20—Arrangements or instruments for measuring magnetic variables involving magnetic resonance
    • G01R33/44—Arrangements or instruments for measuring magnetic variables involving magnetic resonance using nuclear magnetic resonance [NMR]
    • G01R33/48—NMR imaging systems
    • G01R33/54—Signal processing systems, e.g. using pulse sequences ; Generation or control of pulse sequences; Operator console
    • G01R33/56—Image enhancement or correction, e.g. subtraction or averaging techniques, e.g. improvement of signal-to-noise ratio and resolution
    • G01R33/561—Image enhancement or correction, e.g. subtraction or averaging techniques, e.g. improvement of signal-to-noise ratio and resolution by reduction of the scanning time, i.e. fast acquiring systems, e.g. using echo-planar pulse sequences
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R33/00—Arrangements or instruments for measuring magnetic variables
    • G01R33/20—Arrangements or instruments for measuring magnetic variables involving magnetic resonance
    • G01R33/44—Arrangements or instruments for measuring magnetic variables involving magnetic resonance using nuclear magnetic resonance [NMR]
    • G01R33/48—NMR imaging systems
    • G01R33/54—Signal processing systems, e.g. using pulse sequences ; Generation or control of pulse sequences; Operator console
    • G01R33/56—Image enhancement or correction, e.g. subtraction or averaging techniques, e.g. improvement of signal-to-noise ratio and resolution
    • G01R33/565—Correction of image distortions, e.g. due to magnetic field inhomogeneities
    • G01R33/56563—Correction of image distortions, e.g. due to magnetic field inhomogeneities caused by a distortion of the main magnetic field B0, e.g. temporal variation of the magnitude or spatial inhomogeneity of B0

Definitions

  • the invention relates to the technical field of information processing, in particular to a method for measuring the relaxation time of an ultra-short echo time magnetic resonance fingerprint.
  • Magnetic resonance technology provides rich soft tissue imaging contrast.
  • traditional magnetic resonance technology is basically unable to detect bone, Achilles tendon, meniscus and myelin sheath with short T2 or even ultra-short T2 (about 1-10ms).
  • Ultrashort echo technology and zero echo imaging technology have been widely used in imaging of the above-mentioned tissues by shortening the echo time. Because the scanning time of single-point technology and multi-point technology is too long, they are rarely used in the research of subjects.
  • the readout trajectory from the center to the outside is used for signal acquisition, including: semi-projection radial trajectory and spiral trajectory, the signal excitation part adopts half pulse excitation method in 2D imaging, and hard pulse excitation in 3D imaging .
  • Bone compact water content is regarded as a new type of bone compact quality measurement index. Quantitative calculation of bone compact water content requires the measurement of bone tissue T1 and T2*, but it takes up to 1 hour to measure bone tissue T1 and T2*. In order to shorten the scanning time, Abbasi-Rad et al. used the method of double repetition time to measure T1 combined with prior T2* information, but this method obviously ignores the difference in T2* between healthy volunteers and patients.
  • UTE and ZTE technologies are not only used in the diagnosis of skeletal muscle diseases in magnetic resonance systems, but also can provide pseudo-CT (pseudo-CT, pCT) images for PET attenuation correction for multimodal PET (positron emission tomography)/MRI systems.
  • the long T2 suppression pulse can be used as the signal excitation pulse during signal excitation, or the short T2 selection pulse for double inversion recovery, or two echoes can be used in image reconstruction Images with different times are poor.
  • Wiesinger et al. reconstructed the pCT image from the proton density images collected by the ZTE sequence by image segmentation. Since soft tissues, bones and cavities have different relaxation times and proton densities, quantifying multiple tissue parameters can also be used to enhance bone tissue structure.
  • Magnetic resonance fingerprint imaging (magnetic resonance fingerprinting, MRF) technology can achieve multi-parameter simultaneous quantitative imaging, and this method can be applied to quantify bone tissue.
  • MRF technology models different quantitative indicators of tissues, including: T1, T2, T2*, and proton density (PD), to MRF by changing parameters such as the signal flip angle (FA), repetition time TR, and echo time TE. The signal curve is changing.
  • FA signal flip angle
  • TR repetition time
  • TE echo time TE
  • the signal curve is changing.
  • MRF technology has shown clinical potential in brain and abdominal scans.
  • the existing MRF technology still has challenges in quantifying ultra-short T2 tissues.
  • the ultra-short T2 tissues usually have low proton density, resulting in low overall signal strength, which reduces the accuracy of MRF dictionary recognition.
  • the width of the readout window needs to be limited to 0.81T2 in 2D imaging. For bone compactness, the optimized sampling window width is only a few hundred microseconds.
  • each radial readout contains less data.
  • each MRF time point needs to collect multiple radial trace signals.
  • the purpose of the present invention is to provide an ultra-short echo time magnetic resonance fingerprint relaxation time measurement method based on the sine wave echo time in view of the defects of the prior art.
  • An ultrashort echo time magnetic resonance fingerprint relaxation time measurement method including the following steps:
  • phase of a series of down-sampled images in S4 and S3 is the echo time (TE) of sinusoidal fluctuations.
  • TE echo time
  • the optimization of the parameters in S1 is to select the parameters with the smallest mean square error between the simulation results and the theoretical value through the MATLAB simulation method, including: the minimum echo time, the maximum echo time and the sinusoidal fluctuation period, as the experimental parameters.
  • NUFFT non-uniform fast Fourier transform
  • T1 Longitudinal relaxation time, which refers to the time required for the longitudinal magnetization vector to recover from zero to 67% of the total signal strength
  • Transverse relaxation time refers to the time required for the transverse magnetization vector to decay from 100% to 37%; according to the length of the transverse relaxation time, the tissue can be divided into ultra-short T2 tissue (T2 ⁇ 1ms), short T2 tissue ( 1ms ⁇ T2 ⁇ 10ms), and long T2 organization (10ms ⁇ T2). Bone is a common ultra-short T2 tissue, and soft tissues such as muscle belong to long T2 tissue;
  • T2* Effective transverse relaxation time, which refers to the time required for the transverse magnetization vector to decay from 100% to 37% in the presence of magnetic field inhomogeneity
  • Proton density refers to the content of hydrogen protons
  • B0 main magnetic field strength, the B0 diagram usually only shows the difference between the magnetic field distribution and the main magnetic field;
  • FA flip angle
  • TE echo time, which refers to the time interval between the signal excitation center and the echo center. In UTE technology, TE is the time from the end of the excitation pulse to the start of the readout gradient;
  • TR repetition time, refers to the time interval between two consecutive excitations of the sequence.
  • the beneficial effects of the present invention are as follows: Compared with the prior art, the present invention firstly realizes the T1 and T2 relaxation of the short T2 and ultrashort T2 tissues by the ultrashort echo magnetic resonance fingerprint imaging technology. Simultaneous quantitative measurement of time. By proposing the coding mode of sine wave echo time magnetic resonance fingerprint signal, the distinguishing ability and quantitative accuracy of magnetic resonance fingerprint signal for short T2 and ultrashort T2 tissues are improved.
  • the present invention proposes a method based on amplitude modulation and demodulation to realize the direct demodulation estimation of the B0 field without adding additional dictionary calculation and reconstruction calculation amount.
  • the present invention proposes the Mz/T1 method based on the generated longitudinal magnetization vector diagram to suppress the long T2 signal, and at the same time generate a bone enhanced image.
  • Figure 1a is a coding mode of flip angle (FA) and echo time (TE) in a sequence scan of the present invention
  • FIG. 1b is a schematic diagram of a sequence of an ultrashort echo magnetic resonance fingerprint imaging technology in a repetition time (TR) of the present invention
  • FIG. 2 is a schematic diagram of the effect of the sinusoidal wave echo time method of the present invention on improving the difference between signal curves;
  • Figure 3a is a schematic diagram of the simulated water membrane structure of the present invention.
  • Figure 3b is a list of tissue relaxation parameters of the simulated water model of the present invention.
  • Fig. 4a shows the influence of different echo time change ranges of the present invention on the measurement of tissue T1 and T2 relaxation time.
  • the left and right columns of images are the results of repeated scanning of the sequence in Fig. 1a for 3 and 6 times respectively;
  • Fig. 6a is a result analysis diagram of Fig. 5 based on the region of interest according to the present invention.
  • Fig. 6b is a detailed measurement parameter list of the region of interest in Fig. 6a of the present invention.
  • FIG. 7 is a schematic diagram of the results of the scanning experiment of the skeletal muscle system of the subject of the present invention.
  • Figure 8 shows the change of the longitudinal magnetization vector of different tissues according to the present invention
  • FIG. 9 is a schematic diagram of the results of the brain scan experiment of the present invention and the comparison with the CT image.
  • the signal encoding parameters of magnetic resonance fingerprinting (magnetic resonance fingerprinting, MRF), and the two-dimensional ultrashort response based on fast imaging with steady-state precession (FISP) Schematic diagram of the sequence of wave magnetic resonance fingerprint imaging technology (ultrashort echo time magnetic resonance fingerprinting, UTE-MRF).
  • MRF magnetic resonance fingerprinting
  • FISP steady-state precession
  • a UTE-MRF unit contains 480 high-magnification down-sampled images, and the flip angle (FA) and echo time (TE) in the image sequence are continuously changing.
  • FA is composed of four groups of half-period sine waves, in which the peak FA intensity is 32°, 22°, 60°, and 10°, and the minimum FA is 5°.
  • TE changes according to a sinusoidal waveform, where the minimum TE is 0.05ms, the maximum TE is 0.6ms, and the TE fluctuation period is 120 (unit: TR). TR is fixed at 6ms.
  • UTE-MRF adopts ramp sampling technology
  • the readout window width is 0.79ms, including: 0.64ms gradient plateau time and 0.15ms ramp time, and the readout bandwidth is 1780Hz/pixel.
  • a waiting time of 3s is set after the first UTE-MRF unit, in order to restore the proton to the initial state before the second MRF unit.
  • a delay of 3s is repeated each time, and the purpose is to restore the longitudinal magnetization vector.
  • SNR signal-to-noise-ratio
  • the spoke-shaped (radial) readout track rotates according to a small golden angle (23.62°). At the same time point of different repetition times of MRF, the radial trajectory is evenly distributed on the unit circle.
  • UTE-MRF a sinusoidal TE mode is used to improve the sensitivity of MRF to short or even ultra-short T2 tissues.
  • FIG 2 it shows the signal curves of six different organizations using EPG (extended phase graph) simulation.
  • TE is constant (the first two columns in Figure 2), the normalized signals of tissues with different relaxation times are difficult to distinguish, even when TE is minimized to 0.05 ms. But in the TE mode with sinusoidal changes (minimum TE is 0.05ms, maximum TE is 2ms), the signal curves of each tissue can be distinguished intuitively.
  • the changing TE will introduce a phase that varies with the TE time and space position in the MRF image sequence.
  • the extra phase introduced by magnetic field inhomogeneity can be solved by: a) Pre-scan the B0 map and compensate the phase caused by B0, b) Model the non-resonant effect in the MRF dictionary Go in.
  • the present invention proposes a B0 estimation method without a dictionary (dictionary free) through the sinusoidal TE variation mode.
  • the cumulative phase in the MRF image sequence is modulated by the non-uniform field B off and the TE time of the sinusoidal fluctuation.
  • the modulated carrier is ( ⁇ sin( ⁇ )+ ⁇ ),
  • the noise term in the above formula includes physiological noise, thermal noise and other noise related to MRI systems.
  • Each row of S represents a window, and the value of the element is 1 only when the signal is in the window, and 0 otherwise. From the first row to the last row in the S matrix, the window moves from left to right. In the case of a sliding window width of 4, S can be written as:
  • phase image noise term n processed by the sliding window can be almost ignored, so the noise term after the sliding window is considered to be zero.
  • the time integral within the change period of integer multiples of TE can be regarded as a low-pass filter, so that the phase terms of the carrier cos(2 ⁇ ) and sin( ⁇ ) will be filtered out, and B off is derived as:
  • mT is the phase integration period
  • m is the number of cycles
  • T 2 ⁇ / ⁇
  • ⁇ is the frequency of the sine wave
  • ⁇ (TE max -TE min )/2
  • ⁇ (TE max +TE min )/2
  • TE max and TE min are the maximum and minimum echo time respectively.
  • the vertical recovery time T1 of the dictionary ranges from 10ms to 3000ms, specifically: [10:10:400,400:20:2000,2000:40:3000]ms.
  • the transverse relaxation time T2 ranges from 0.1 ms to 300 ms, specifically: [0.1:0.1:5,5:5:150,150:10:300] ms.
  • a two-step dictionary design method is adopted, and the area of all curve envelopes in the dictionary is normalized to 1.
  • Step 1 Use non-uniform fast Fourier transform (NUFFT) algorithm to reconstruct multi-channel images from multi-channel k-space, and then synthesize a single image using adaptive coil superposition method. Then, the images from the positive and negative polarity gradient excitations are directly and complexly added to obtain a complete image of the selected layer.
  • Step two use the formula Estimate the B0 image from the phase part of the image in step 1. In order to reduce the interference of downsampling and aliasing and reduce the loss of MRF signal specificity, the sliding window width is set to 20.
  • Step 3 In order to remove the phase interference caused by the field inhomogeneity, the field inhomogeneity estimated in step 2 is compensated into the complex MRF image sequence.
  • Step 4 Normalize the image sequence of the MRF and find the dictionary curve that is most similar to the scan signal by dot multiplication with the dictionary curve. Because bone compact has a lower proton density and ultra-short T2 value, compared to bone marrow, radial downsampling artifacts interfere more with its recognition. Therefore, before bone signal recognition, the method of partial volume dictionary recognition is used to estimate and remove the bone marrow components in the signal, and then the method of MRF dictionary recognition is used to reconstruct a multi-parameter quantitative map.
  • MRF uses an adiabatic inversion pulse to make the image signal of MRF show the effect of soft tissue suppression when the long T2 tissue inversion returns to zero.
  • this chapter proposes an image reconstruction method for bone enhancement.
  • bone tissue has an ultra-short T2, which results in a very rapid signal attenuation in the transverse plane, the properties of short T1 make it recover faster in the longitudinal plane than the long T2 tissue, so the bone tissue exhibits a high signal in the longitudinal magnetization vector diagram .
  • the MRF technology can quantify the T1 and T2 relaxation times of the tissue, through the reverse look-up method, the transverse and longitudinal magnetization vector diagrams at any time in the MRF scan can be obtained.
  • the image in which the ultra-short T2 bone signal is enhanced in the longitudinal magnetization vector diagram is selected as the skull enhancement map.
  • the ideal T2' value is shown in Figure 3(b), so that the T2* of the muscle is 25ms, the T2* of the Achilles tendon is 2.3ms, the T2* of all bone water is 0.7ms, and the T2* of bone free water is 2.4 ms.
  • the subjects of the experiment were a self-made agar water film to simulate soft tissue, and a rubber sieve to simulate ultra-short T2 tissue.
  • the magnetic resonance signal was collected by a 20-channel head coil.
  • the agar water film contains 7 test tubes, which are composed of MnCl 2 agar gels of different concentrations to simulate the difference between T1 and T2 in tissues.
  • the collection time of the gold standard scan is as follows: T1: 124 minutes, T2*: 74 minutes, and T2: 36 minutes.
  • the experimental results of the water film are shown in Figures 5 and 6a and 6b.
  • FIG. 7 The brain scan of a patient with facial neuroma, where the scanning coil uses a 64-channel head coil. Considering the safety of the scan when the patient is scanned, the amplitude of the FA sequence is halved.
  • Figure 9 is a comparison between the bone enhancement image reconstructed from the patient’s scan results and the patient’s CT image.
  • CT data was collected on the Philips iCT instrument (Brilliance iCT, Philips Healthcare, The Netherlands) of the local hospital during preoperative navigation.
  • Philips iCT instrument Bailliance iCT, Philips Healthcare, The Netherlands
  • the imaging layer thickness of the water film is 6mm, and the layer thickness of the human body scan is 7mm.
  • the minimum TE value (0.05ms) remains unchanged during brain scan.
  • the multi-parameter quantitative results map was reconstructed by MATLAB R2014a (The MathWorks, MA) on Linux server (Core i7Intel Xeon 2.8GHz CPUs and 64GB RAM).
  • the above experiments were all completed on Siemens Prisma scanner.
  • the parameters in the embodiment are defined as follows:
  • T1 Longitudinal relaxation time, which refers to the time required for the longitudinal magnetization vector to recover from zero to 67% of the total signal strength
  • Transverse relaxation time refers to the time required for the transverse magnetization vector to decay from 100% to 37%; according to the length of the transverse relaxation time, the tissue can be divided into ultra-short T2 tissue (T2 ⁇ 1ms), short T2 tissue ( 1ms ⁇ T2 ⁇ 10ms), and long T2 organization (10ms ⁇ T2). Bone is a common ultra-short T2 tissue, and soft tissues such as muscle belong to long T2 tissue;
  • T2* Effective transverse relaxation time, which refers to the time required for the transverse magnetization vector to decay from 100% to 37% in the presence of magnetic field inhomogeneity
  • Proton density refers to the content of hydrogen protons
  • B0 main magnetic field strength, the B0 diagram usually only shows the difference between the magnetic field distribution and the main magnetic field;
  • FA flip angle
  • TE echo time, which refers to the time interval between the signal excitation center and the echo center. In UTE technology, TE is the time from the end of the excitation pulse to the start of the readout gradient;
  • TR repetition time, refers to the time interval between two consecutive excitations of the sequence.

Landscapes

  • Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • High Energy & Nuclear Physics (AREA)
  • Condensed Matter Physics & Semiconductors (AREA)
  • General Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • General Health & Medical Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Veterinary Medicine (AREA)
  • Public Health (AREA)
  • Animal Behavior & Ethology (AREA)
  • Surgery (AREA)
  • Molecular Biology (AREA)
  • Biophysics (AREA)
  • Pathology (AREA)
  • Medical Informatics (AREA)
  • Radiology & Medical Imaging (AREA)
  • Spectroscopy & Molecular Physics (AREA)
  • Rheumatology (AREA)
  • Orthopedic Medicine & Surgery (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Dentistry (AREA)
  • Signal Processing (AREA)
  • Magnetic Resonance Imaging Apparatus (AREA)

Abstract

一种超短回波时间磁共振指纹弛豫时间测量方法，采用半脉冲激发和半投影读出缩短TE实现超短T2时间信号的采集，图像采集和重建基于磁共振指纹成像技术。引入正弦波动的回波时间变化模式，提高了磁共振指纹信号对短T2和超短T2组织的区分能力，实现短T2和超短T2组织以及长T2组织的同时多参数定量成像。通过正弦波动的TE将磁场的不均匀性调制进指纹信号的相位信息中，根据调幅信号解调原理直接重建B0图，并在指纹信号中对B0场引起的相位改变进行补偿，提高信号识别的准确性。在磁共振骨骼肌系统成像中，实现对骨（超短T2）和肌肉（长T2）组织的同时T1，T2，PD和B0的定量测量；也可用于脑部成像，测量灰/白质弛豫时间以及颅骨结构。

Description

一种超短回波时间磁共振指纹弛豫时间测量方法 技术领域
本发明涉及信息处理技术领域，尤其涉及一种超短回波时间磁共振指纹弛豫时间测量方法。
背景技术
磁共振技术提供了丰富的软组织成像对比度。但传统磁共振技术基本无法检测具有短T2甚至超短T2(约1-10ms)的骨、跟腱、半月板以及髓鞘壁。超短回波技术以及零回波成像技术通过缩短回波时间，已经被广泛的应用在上述组织的成像中。由于单点技术和多点技术的扫描时间过长，很少应用在被试的研究中。为了提高k空间的采样效率，信号采集时采用中心向外的读出轨迹，包括：半投影radial轨迹和spiral轨迹，信号激发部分二维成像中采用半脉冲激发方法，三维成像中采用硬脉冲激发。
定量超短回波技术被应用于研究大量的骨关节疾病，包括：关节软骨退变、半月板撕裂以及年龄相关的骨密质退化以及骨质酥松。骨密质水含量被视为新型的骨密质质量的衡量指标，定量计算骨密质水含量需要测量骨组织的T1和T2*，但测量骨组织的T1和T2*需要长达1小时。为了缩短扫描时间，Abbasi-Rad等人利用双重复时间测量T1的方法结合先验的T2*信息，但该方法显然忽略了健康志愿者和病人之间T2*的差异。
UTE和ZTE技术不仅应用于磁共振系统对骨骼肌疾病进行诊断，而且可以为多模态PET(positron emission tomography)/MRI系统提供用于PET衰减矫正的伪CT(pseudo-CT，pCT)图。为了增强骨组织的信号抑制背景长T2组织的信号，信号激励时可采用长T2抑制脉冲作为信号激发脉冲，或双反转恢复的短T2选择脉冲，也可在图像重建时利用两幅回波时间不同的图像做差。Wiesinger等人通过图像分割的方法从ZTE序列采集的质子密度图像中重建出pCT图。由于软组织、骨和空腔具有不同的弛豫时间和质子密度，同时定量多组织参数也可用于增强骨组织结构。
磁共振指纹成像(magnetic resonance fingerprinting,MRF)技术可实现多参数的同时定量成像，该方法可以应用于对骨组织的定量。MRF技术通过改变信号翻转角(flip angle,FA)、重复时间TR以及回波时间TE等参数将组织的不同定量指标，包括：T1、T2、T2*和质子密度(PD)等建模到MRF信号曲线的变化中。此外，MRF技术在脑部以及腹部扫描中已展现出临床潜力。但是现有MRF技术对超短T2的组织进行定量仍存在挑战。
首先，由于传统MRF技术的最短回波时间为数毫秒，使MRF难以检测超短T2的组织信号。
其次，超短T2的组织通常质子密度都低使得整体信号强度低，降低了MRF字典识别的准确性。第三，为了避免T2*模糊对读出信号的影响，二维成像中需要将读出窗口宽度限制在0.81T2，对于骨密质，优化的采样窗宽仅数百微秒。
最后，与传统MRF的spiral读出相比，每次radial读出包含更少的数据量，为了降低单个MRF时间点的降采样伪影，每个MRF时间点需要采集多条radial轨迹的信号。
发明内容
本发明的目的是针对现有技术的缺陷，提供了一种基于正弦波动回波时间的超短回波时间磁共振指纹弛豫时间测量方法。
为了实现以上目的，本发明采用以下技术方案：
一种超短回波时间磁共振指纹弛豫时间测量方法，包括以下步骤：
S1、建立回波磁共振指纹成像序列设计与实现，包括正弦波动回波时间参数的优化；
S2、利用S1的建立回波磁共振指纹成像序列通过磁共振扫描仪对被试进行扫描，获得原始k空间数据；
S3、将S2中的原始k空间数据重建成一系列的降采样图像；
S4、S3中一系列降采样图像的相位是正弦波动的回波时间(echo time,TE) 调制了B0场不均匀性的结果：Phase＝2pi·B off·(αsin(ωτ)+β)+n，
其中，B off为由场不均匀性和化学位移引起的频率偏移，单位：Hz，α,β和ω是TE的采样参数，α＝(TE max-TE min)/2，β＝(TE max+TE min)/2,TE max和TE min分别为最大和最小回波时间；ω为正弦波的频率；τ为时间向量[1,2,…,F] T，时间单位是一个重复时间(TR)；n代表了噪声项，B off的解调通过乘以载波sin(ωτ)再低通滤波的方式实现；
S5、采用滑动窗技术先对指纹信号进行滑动平均，然后对S4进行幅度解调，即通过乘以载波sin(ωτ)再低通滤波的方式，计算出B off：
Figure PCTCN2020089424-appb-000001
Figure PCTCN2020089424-appb-000002
其中，dPhase为S4中乘以sin(ωτ)的相位信号，mT为相位积分周期，m为周期数，π为圆周率，T＝2π/ω，S为F×F的矩阵，包含的元素非0即1，S的每行代表一个窗口，有且仅当信号在窗口内时元素的值为1，否则为0，S矩阵中从第一行到最后一行，窗口从左向右移动；
S6、将S5计算出的B off补偿到滑动窗处理后的指纹信号中。然后通过磁共振指纹成像技术中字典识别方法从采集的指纹信号中重建出多参数定量图；
S7、根据S6重建出的组织弛豫时间，查招字典，获得对应的纵向磁化矢量变化曲线，选取骨组织对比度最高的图像作为骨增强图，记为Mz，使用Mz/T1作为骨增强图像输出，起到抑制长T1组织的目的，其中T1为S6中输出的纵向弛豫时间定量结果。
进一步的，S1中参数的优化为，通过MATLAB仿真方法选取仿真结果与理论值均方跟误差最小的参数，包括：最小回波时间、最大回波时间和正弦波动周期，作为实验参数。
进一步的，S3中通过非均匀快速傅里叶变换(non-uniform fast Fourier transform,NUFFT)重建算法将S2中的原始k空间数据重建成一系列的降采样图像。
进一步的，S5中在滑动窗宽为4的情况下，可以将S写成：
Figure PCTCN2020089424-appb-000003
方法中参数定义如下：
T1:纵向弛豫时间，指纵向磁化矢量从零恢复至总信号强度的67％所需的时间；
T2:横向弛豫时间，指横向磁化矢量从100％衰减至37％所需的时间；根据横向弛豫时间的长短，可以将组织划分为超短T2组织(T2≤1ms)，短T2组织(1ms<T2≤10ms)，和长T2组织(10ms<T2)。骨是常见的超短T2组织，肌肉等软组织属于长T2组织；
T2*:有效横向弛豫时间，指存在磁场不均匀性的情况下，横向磁化矢量从100％衰减至37％所需的时间；
T2′：磁场不均匀性引起的横向弛豫时间改变，其中1/T2＝1/T2*+1/T2′；
PD:质子密度(proton density)，指氢质子的含量；
B0：主磁场强度，B0图中通常仅显示磁场分布相对于主磁场的差异；
FA:翻转角(flip angle)，指激发脉冲使磁化矢量沿主磁场方向偏离的角度。FA＝90°时，磁化矢量垂直于主磁场方向；
TE：回波时间(echo time)，指信号激发中心到回波中心之间的时间间隔。在UTE技术中，TE为激发脉冲的终点到读出梯度起点的时间；
TR:重复时间(repetition time)，指序列两次相邻激发之间的时间间隔。
采用本发明技术方案，本发明明的有益效果为：与现有技术相比，本发明首先超短回波磁共振指纹成像技术，实现对短T2和超短T2组织的T1、T2弛豫，时间同时定量测量，通过提出正弦波动回波时间磁共振指纹信号的编码模式，提高了磁共振指纹信号对短T2和超短T2组织的区分能力和定量准确性。本发明中提出基于幅度调制解调的方法实现B0场的直接解调估计，无需增加额外的字典计算和重建计算量。最后，本发明根据生成的纵向磁化矢量图提出 Mz/T1方法抑制长T2信号，同时生成骨增强图像。
附图说明
图1a为本发明的一次序列扫描中翻转角(flip angle,FA)和回波时间(echo time,TE)的编码模式；
图1b为本发明的一个回波时间(repetition time,TR)内超短回波磁共振指纹成像技术的序列示意图；
图2为本发明的正弦波动回波时间方法对提高信号曲线之间的差异效果示意图；
图3a为本发明的仿真水膜结构示意图；
图3b为本发明的仿真水模各个组织弛豫参数列表；
图4a为本发明的不同回波时间改变范围对组织T1,T2弛豫时间测量的影响，左右两列图像分别是图1a中的序列重复扫描3次和6次的结果；
图4b为本发明的图4a中最大TE＝0.6ms情况下，仿真结果与理论结果之间的对比图以及残差图。
图5为本发明的水模的UTE-MRF方法与金标准方法测量结果的对比图；
图6a为本发明的基于感兴趣区域对图5的结果分析图；
图6b为本发明的图6a中感兴趣区域的详细测量参数列表；
图7为本发明的被试骨骼肌系统扫描实验结果示意图；
图8为本发明的不同组织纵向磁化矢量的改变；
图9为本发明的被试脑部扫描实验结果及于CT图的对比示意图。
具体实施方式
结合附图对本发明具体方案具体实施例作进一步的阐述。
如图1a、1b所示，为磁共振指纹(magnetic resonance fingerprinting,MRF)的信号编码参数，以及基于稳态自由感应快速成像序列(fast imaging with steady-state precession,FISP)的二维超短回波磁共振指纹成像技术(ultrashort echo time magnetic resonance fingerprinting,UTE-MRF)的序列示意图。为了降低 回波时间，采用半脉冲激发技术。将脉冲持续时间为1.2ms，时间带宽积为6的sinc脉冲分成两个半脉冲。通过变速率选择性激发算法(variable-rate selective excitation,VERSE)将RF的峰值能量和截断的梯度轨迹平缓降低至0，生成的半脉冲持续时间为0.7ms。半脉冲施加的同时伴随着正负双极性梯度脉冲以实现完整的选层，同时提高层选梯度对涡流的鲁棒性。一个UTE-MRF单元包含了480幅高倍降采样的图像，图像序列中翻转角(flip angle,FA)和回波时间(echo time,TE)都在连续变化。FA由四组半周期的正弦波型组成，其中峰值FA强度分别为32°,22°,60°,和10°，且最小FA为5°。TE按照正弦波形改变，其中最小TE为0.05ms，最大TE为0.6ms，TE的波动周期为120(单位：TR)。TR固定在6ms。
为了缩短回波时间，UTE-MRF中采用斜坡采样技术，读出窗宽0.79ms，包含：0.64ms的梯度平台时间和0.15ms的斜坡时间，读出带宽为1780Hz/pixel。第一个UTE-MRF单元之后设置3s的等待时长，目的是在第二个MRF单元之前使质子恢复的初始状态。重复扫描时，每次重复间隔3s的延时，目的也是让纵向磁化矢量恢复。为了最大化信噪比(signal-to-noise-ratio,SNR)，同时将扫描时间控制在1分钟内，水膜、脚踝和脑组织扫描时重复次数为5。小腿骨密质定量时，为了提高SNR额外多重复了一次，总扫描时间增加至68s。为了增加MRF图像之间的不一致性同时降低涡流对图像质量的影响，轮辐状(radial)读出轨迹按照小黄金角(23.62°)旋转。在MRF的不同重复次数的相同时间点，radial轨迹均匀的分布在单位圆上。
现有的MRF技术研究中，通过引入回波时间的变化要么为了提高MRF对T2*的敏感度，用于区分水和脂肪的信号。UTE-MRF技术中，使用正弦变化的TE模式来提高MRF对短甚至超短T2组织的敏感度。
如图2所示，显示了六种不同组织利用EPG(extended phase graph，扩展相位图)仿真的信号曲线，这些组织的T1＝180ms，T2＝0.5,1,2,5以及10ms，采用图1(b)中的FA变化模式。当TE为常数时(图2前两列)，不同弛豫时间组织 的归一化信号难以区分，即使TE最小化到0.05ms时。但是在正弦变化的TE模式下(最小TE为0.05ms，最大TE为2ms)，可以直观的区分各个组织的信号曲线。但在磁场不均匀时，变化的TE会在MRF图像序列中引入随TE时间和空间位置变化的相位。
为了避免损失相位的信息，磁场非均匀性引入的额外相位可以通过一下方式解决：a)通过预扫描B0图，并对B0引起的相位进行补偿，b)在MRF字典中将非共振效应建模进去。
但上述方法在MRF字典计算和识别方面都需要增加计算量。本发明通过正弦的TE变化模式，提出无需字典(dictionary free)的B0估计方法。MRF图像序列中的累积相位被非均匀场B off和正弦波动的TE时间调制，调制的载波为(αsin(ωτ)+β)，
Phase＝2pi·B off·(αsin(ωτ)+β)+n，
其中，B off为由场不均匀性和化学位移引起的频率偏移，单位：Hz，α,β和ω是TE的采样参数，α＝(TE max-TE min)/2，β＝(TE max+TE min)/2,TE max和TE min分别为最大和最小回波时间；ω为正弦波的频率；τ为时间向量[1,2,…,F] T，以重复时间(repetition time,TR)为单位；n代表了噪声项，B off的解调通过乘以载波sin(ωτ)再低通滤波的方式实现：
dPhase＝(2π·B off·(αsin(ωτ)+β)+n)·sin(ωτ)。
进一步将dPhase写成：
Figure PCTCN2020089424-appb-000004
其中，B off为由场不均匀性和化学位移引起的频率偏移，单位：Hz，α,β和ω是TE的采样参数，α＝(TE max-TE min)/2，β＝(TE max+TE min)/2,TE max和TE min分别为最大和最小回波时间；ω为正弦波的频率；τ为时间向量[1,2,…,F] T，时间单位是一个重复时间(repetition time,TR)；n代表了噪声项。上式中的噪声项包含了生理噪声、热噪声以及其他MRI系统相关的噪声。但由于MRI图像序列中每张图重建自一个TR内采集的k空间按数据，因此噪声项n中的大部分信号源 自图像的降采样伪影。为了降低k空间降采样引起的相位噪声，对上式施加滑动窗算法，对相位图像序列进行滑动平均，具体方法是等式dPhase＝(2π·B off·(αsin(ωτ)+β)+n)·sin(ωτ)左右乘以滑动窗矩阵：
Figure PCTCN2020089424-appb-000005
其中，S为F×F的矩阵，包含的元素非0即1；B off为由场不均匀性和化学位移引起的频率偏移，单位：Hz；α,β和ω是TE的采样参数，α＝(TE max-TE min)/2，β＝(TE max+TE min)/2,TE max和TE min分别为最大和最小回波时间；ω为正弦波的频率；τ为时间向量[1,2,…,F] T，时间单位是一个重复时间(TR)；n代表了噪声项。S的每行代表一个窗口，有且仅当信号在窗口内时元素的值为1，否则为0。S矩阵中从第一行到最后一行，窗口从左向右移动。在滑动窗宽为4的情况下，可以将S写成：
Figure PCTCN2020089424-appb-000006
通过滑动窗处理的相位图像噪声项n的影响几乎可以忽略，因此认为滑动窗后的噪声项为0。整数倍TE变化周期内的时间积分可视为低通滤波器，从而载波为cos(2ωτ)和sin(ωτ)的相位项将被滤除，从而导出B off为：
Figure PCTCN2020089424-appb-000007
其中，mT为相位积分周期，m为周期数，T＝2π/ω，ω为正弦波频率，α＝(TE max-TE min)/2，β＝(TE max+TE min)/2,TE max和TE min分别为最大和最小回波时间。
MRF的字典根据图1(b)中的FA模式和正弦的TE变化模式(最小TE＝0.05ms，最大TE＝[0.05:0.05:1.0]ms)，通过相位拓展图(extended phase graph,EPG)的方法计算得到，本示例中将TE的周期设置为120个TR。字典的纵向恢复时间T1的范围从10ms到3000ms，具体为：[10:10:400,400:20:2000,2000:40:3000]ms。横向弛豫时间T2从0.1ms到300ms，具体为：[0.1:0.1:5,5:5:150,150:10:300] ms。考虑到纵向磁化矢量的部分恢复效应，采用一个两步的字典设计方法，字典中所有曲线包络的面积都归一化为1。
通过以下四步重建多参数T1、T2、PD以及B0的定量图谱。步骤一，利用非均匀快速傅里叶变换(non-uniform fast Fourier transform,NUFFT)算法从多通道k空间重建多通道的图像，然后利用自适应线圈叠加的方法合成单幅图像。接着将来自正负极性梯度激励的图像直接复数相加，得到完整选层的图像。步骤二，利用式
Figure PCTCN2020089424-appb-000008
从步骤一中图像的相位部分估计出B0图，为了减小降采样混叠的干扰，且降低MRF信号特异性的损失，滑动窗宽度取20。同时，为了避免IR引起的相位翻转的影响，仅用从240到480个TR的相位图(即两个TE变化周期)。步骤三，为了去除场不均匀性产生的相位干扰，将步骤二估计的场不均匀性补偿进复数的MRF图像序列中。步骤四，MRF的图像序列归一化并通过与字典曲线点乘的方法找出与扫描信号最相似的字典曲线。由于骨密质具有较低的质子密度和超短的T2值，相比于骨髓，radial降采样伪影对其识别的干扰更大。因此在骨信号识别前，先利用部分容积字典识别的方法估计并去除信号中的骨髓成分，然后利用MRF字典识别的方法重建多参数的定量图。
MRF中为了在信号中引入T1加权，使用了绝热反转脉冲，使MRF的图像信号在长T2组织反转恢复至零点时表现出软组织抑制的效果。基于这一特性，本章提出骨增强的图像重建方法。虽然骨组织具有超短T2，造成其横向平面内信号衰减非常快速，但短T1的属性使其在纵向平面的恢复速度快于长T2组织，从而骨组织在纵向磁化矢量图中表现出高信号。因为MRF技术可定量出组织的T1和T2弛豫时间，通过反向查字典的方法，可以从在MRF扫描中任意时间的横向和纵向磁化矢量图。最后，选取纵向磁化矢量图中超短T2骨信号被增强的图作为颅骨增强图。
为了研究正弦的TE采样模式对组织定量准确性的影响，在如图3a、3b的数字水膜(110x 110)上进行了仿真实验。数字水膜包含的组织有：肌肉、跟腱、 骨的全部水信号以及骨的自由水信号。水膜中黑色的区域表示空气。图3(b)中列举了组织的参数特性。理想的MRF图像序列来自EPG算法计算得到的横向磁化矢量。尽管TE的变化提高了MRF对短T2甚至超短T2组织的特异性，也在图像中引入了由TE变化引起的T2*加权。因为通过EPG已经将组织的T2效应模拟进了MRF信号曲线中，T2*的效应通过在理想MRF信号曲线中乘以额外的T2′衰减项实现，因为T2*是T2和T2′的综合效应：exp(TE/T2*)＝exp(TE/T2+TE/T2′)。理想的T2′的值如图3(b)所示，使得肌肉的T2*为25ms，跟腱的T2*为2.3ms，骨头全部水的T2*为0.7ms，骨头自由水的T2*为2.4ms。
在仿真实验中，也考虑了不同组织SNR的差异对定量的影响。将不同噪声强度的复数高斯白噪声加入到MRF的信号中，使得不同组织的SNR如图3(b)所示。MRF每个时间点的图像都通过NUFFT算法将含噪声的MRF横向磁化矢量变换到k空间，其中需要黄金角旋转的radial采样轨迹以及对应密度补偿函数。最小TE固定在0.05ms，最大TE从0.05ms提高到1ms，步长为0.05ms。定量多参数图谱采用上一小节中图像重建部分介绍的方法。结果计算了测量的T1和T2值与理想的T1和T2值之间的均方根误差(root mean square error,RMSE)。仿真实验结果如图4a、4b所示。
实验扫描对象为自制的琼脂水膜用以模拟软组织，以及橡胶筛以模拟超短T2的组织，磁共振信号由20通道的头线圈采集。琼脂水膜包含7只试管，由不同浓度的MnCl 2琼脂凝胶构成，用于模拟组织中的T1和T2差异。还有一只试管充满了植物油(包含94％的大豆油和6％葵花子油)，该植物油的样品有一个主要的共振频率，其中心频率与水之间的偏差约为3.46ppm。金标准T1图由反转恢复超短回波(inversion recovery ultrashort echo time,IR-UTE)序列采集得到，其中TI时间为：50,100,200,400和800ms,TR＝3000ms,TE＝0.05ms,radial条数＝248。金标准的T2*定量图由UTE序列采集得到，其中TE时间分别为：0.05,0.2,0.5,1,2和4ms,TR＝1500ms,radial条数＝248。所有IR-UTE以及UTE序列采集的数据都采用MRF图像重建步骤一的算法进行重建。金标准的T2定量图通 过SE序列测量，其中TE＝25,50,75,100和125ms,TR＝3000ms,重建矩阵＝192x 192,分辨率＝1x 1mm 2,同时采用了6/8的部分傅里叶采集。金标准扫描的采集时间如下：T1：124分钟,T2*：74分钟,以及T2：36分钟。水膜的实验结果如图5和6a、6b所示。
小腿以及跟腱的成像，信号采集使用15通道的膝盖线圈，结果见图7。面神经瘤患者脑部的扫描，其中扫描线圈采用64通道头线圈。患者扫描时考虑到扫描的安全性，将FA序列的幅度减半。图8显示了基于脑组织的弛豫时间模拟的纵向磁化矢量改变曲线，以及在帧数＝130和393时的纵向磁化矢量图。图9为病人扫描结果重建出的骨增强图像与该病人的CT图像进行了对比，其中CT采集的详细参数为：管电压＝80kV,管电流＝365mA,DLP＝97.9mGy·cm,平均辐射剂量＝0.34mSv,层厚＝1.0mm,以及分辨率＝0.5x 0.5mm 2)，CT数据是病人术前导航在当地医院的Philips iCT仪(Brilliance iCT,Philips Healthcare,The Netherlands)上采集得到。利用RadiAnt Dicom Viewer(Medixant Co.,Poland)的三维旋转功能将三维的CT图像与二维的磁共振图像进行配准。
水膜的成像层厚为6mm，人体扫描的层厚为7mm。为了提高颅骨增强图像的图像质量，脑部扫描时回波时间保持最小的TE值(0.05ms)不变。多参数定量结果图由Linux服务器(Core i7Intel Xeon 2.8GHz CPUs and 64GB RAM)上的MATLAB R2014a(The MathWorks,MA)重建得到。水膜、小腿和跟腱的图像重建分辨率为1.0x 1.0mm 2(重建矩阵＝240x 240)，脑组织的分辨率为0.75x0.75mm 2(矩阵大小＝256x 256)。上述实验均在西门子Prisma扫描仪上完成。实施例中参数定义如下：
T1:纵向弛豫时间，指纵向磁化矢量从零恢复至总信号强度的67％所需的时间；
T2:横向弛豫时间，指横向磁化矢量从100％衰减至37％所需的时间；根据横向弛豫时间的长短，可以将组织划分为超短T2组织(T2≤1ms)，短T2组织(1ms<T2≤10ms)，和长T2组织(10ms<T2)。骨是常见的超短T2组织，肌肉 等软组织属于长T2组织；
T2*:有效横向弛豫时间，指存在磁场不均匀性的情况下，横向磁化矢量从100％衰减至37％所需的时间；
T2′：磁场不均匀性引起的横向弛豫时间改变，其中1/T2＝1/T2*+1/T2′；
PD:质子密度(proton density)，指氢质子的含量；
B0：主磁场强度，B0图中通常仅显示磁场分布相对于主磁场的差异；
FA:翻转角(flip angle)，指激发脉冲使磁化矢量沿主磁场方向偏离的角度。FA＝90°时，磁化矢量垂直于主磁场方向；
TE：回波时间(echo time)，指信号激发中心到回波中心之间的时间间隔。在UTE技术中，TE为激发脉冲的终点到读出梯度起点的时间；
TR:重复时间(repetition time)，指序列两次相邻激发之间的时间间隔。
注意，上述仅为本发明的较佳实施例及所运用技术原理。本领域技术人员会理解，本发明不限于这里所述的特定实施例，对本领域技术人员来说能够进行各种明显的变化、重新调整和替代而不会脱离本发明的保护范围。因此，虽然通过以上实施例对本发明进行了较为详细的说明，但是本发明不仅仅限于以上实施例，在不脱离本发明构思的情况下，还可以包括更多其他等效实施例，而本发明的范围由所附的权利要求范围决定。

Claims (4)

  1. 一种超短回波时间磁共振指纹弛豫时间测量方法，其特征在于，包括以下步骤：
    S1、建立回波磁共振指纹成像序列设计与实现，包括正弦波动回波时间参数的优化；
    S2、利用S1的建立回波磁共振指纹成像序列通过磁共振扫描仪对被试进行扫描，获得原始k空间数据；
    S3、将S2中的原始k空间数据重建成一系列的降采样图像；
    S4、S3中一系列降采样图像的相位是正弦波动的回波时间(echo time,TE)调制了B0场不均匀性的结果：Phase＝2pi·B off·(αsin(ωτ)+β)+n，
    其中，B off为由场不均匀性和化学位移引起的频率偏移，单位：Hz，α,β和ω是TE的采样参数，α＝(TE max-TE min)/2，β＝(TE max+TE min)/2,TE max和TE min分别为最大和最小回波时间；ω为正弦波的频率；τ为时间向量[1,2,…,F] T，时间单位是一个重复时间(TR)；n代表了噪声项，B off的解调通过乘以载波sin(ωτ)再低通滤波的方式实现；
    S5、采用滑动窗技术先对指纹信号进行滑动平均，然后对S4进行幅度解调，即通过乘以载波sin(ωτ)再低通滤波的方式，计算出
    Figure PCTCN2020089424-appb-100001
    Figure PCTCN2020089424-appb-100002
    其中，dPhase为S4中乘以sin(ωτ)的相位信号，mT为相位积分周期，m为周期数，π为圆周率，T＝2π/ω，S为F×F的矩阵，包含的元素非0即1，S的每行代表一个窗口，有且仅当信号在窗口内时元素的值为1，否则为0，S矩阵中从第一行到最后一行，窗口从左向右移动；
    S6、将S5计算出的B off补偿到滑动窗处理后的指纹信号中。然后通过磁共振指纹成像技术中字典识别方法从采集的指纹信号中重建出多参数定量图；
    S7、根据S6重建出的组织弛豫时间，查招字典，获得对应的纵向磁化矢量变化曲线，选取骨组织对比度最高的图像作为骨增强图，记为Mz，使用Mz/T1作为骨增强图像输出，起到抑制长T1组织的目的，其中T1为S6中输 出的纵向弛豫时间定量结果。
  2. 如权利要求1所述的一种超短回波时间磁共振指纹弛豫时间测量方法，其特征在于，S1中参数的优化为，通过MATLAB仿真方法选取测量准确性最高的回波时间参数，包括：最小回波时间、最大回波时间和正弦波动周期。
  3. 如权利要求1所述的一种超短回波时间磁共振指纹弛豫时间测量方法，其特征在于，S3中通过非均匀快速傅里叶变换(non-uniform fast Fourier transform,NUFFT)重建算法将S2中的原始k空间数据重建成一系列的降采样图像。
  4. 如权利要求1所述的一种超短回波时间磁共振指纹弛豫时间测量方法，其特征在于，S5中在滑动窗宽为4的情况下，可以将S写成：
    Figure PCTCN2020089424-appb-100003
PCT/CN2020/089424 2019-05-10 2020-05-09 一种超短回波时间磁共振指纹弛豫时间测量方法 Ceased WO2020228641A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US17/263,549 US11353531B2 (en) 2019-05-10 2020-05-09 Method for measuring relaxation time of ultrashort echo time magnetic resonance fingerprinting

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910388149.4A CN110133553B (zh) 2019-05-10 2019-05-10 一种超短回波时间磁共振指纹弛豫时间测量方法
CN201910388149.4 2019-05-10

Publications (1)

Publication Number Publication Date
WO2020228641A1 true WO2020228641A1 (zh) 2020-11-19

Family

ID=67573189

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/089424 Ceased WO2020228641A1 (zh) 2019-05-10 2020-05-09 一种超短回波时间磁共振指纹弛豫时间测量方法

Country Status (3)

Country Link
US (1) US11353531B2 (zh)
CN (1) CN110133553B (zh)
WO (1) WO2020228641A1 (zh)

Families Citing this family (19)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110133553B (zh) 2019-05-10 2020-06-05 浙江大学 一种超短回波时间磁共振指纹弛豫时间测量方法
CN111105472B (zh) * 2019-11-26 2024-03-22 上海联影智能医疗科技有限公司 Pet图像的衰减校正方法、装置、计算机设备和存储介质
CN112946546B (zh) * 2019-12-10 2023-10-27 西门子(深圳)磁共振有限公司 短t2组织的成像方法、系统及磁共振成像系统
CN113495242B (zh) * 2020-04-03 2022-12-20 上海联影医疗科技股份有限公司 相位误差探测方法、装置、磁共振系统及其成像方法
CN111568390B (zh) * 2020-05-15 2023-07-07 杭州佳量医疗科技有限公司 一种减少实时磁共振温度成像中系统温度误差的方法
CN112075934B (zh) * 2020-09-09 2021-07-23 清华大学 用于识别颈动脉斑块的磁共振单序列多参数定量成像系统
CN114325523B (zh) * 2020-09-27 2023-10-03 上海联影医疗科技股份有限公司 T1值确定方法、装置、电子设备和存储介质
CN112754459A (zh) * 2020-12-23 2021-05-07 上海交通大学 骨骼组织定量成像方法、系统、介质及终端
CN113538611A (zh) * 2021-06-21 2021-10-22 复旦大学 一种无监督的心脏磁共振参数定量图像重建方法
CN113391250B (zh) * 2021-07-09 2022-11-29 清华大学 组织属性的多参数定量测试系统及其方法
CN115700103B (zh) * 2021-07-22 2025-04-22 中国科学院深圳先进技术研究院 血脑屏障关闭时间的确定方法、装置、电子设备及介质
CN114217255B (zh) * 2021-11-29 2022-09-20 浙江大学 一种快速的肝脏多参数定量成像方法
CN114690101B (zh) * 2022-03-30 2024-08-02 西门子数字医疗科技(上海)有限公司 磁共振成像中的参数定量检测方法、装置及磁共振扫描仪
CN115137347B (zh) * 2022-05-10 2023-05-26 浙江大学 三维超短回波时间磁共振指纹成像的髓鞘定量成像方法
US12078696B2 (en) 2022-10-06 2024-09-03 GE Precision Healthcare LLC Silent calibration for magnetic resonance imaging
CN115629347B (zh) * 2022-10-20 2023-09-19 无锡鸣石峻致医疗科技有限公司 一种磁共振成像系统中获得梯度轨迹的方法、装置和介质
CN116900536A (zh) * 2023-08-23 2023-10-20 中建安装集团有限公司 一种环管无应力配管方法
CN119444667B (zh) * 2024-09-30 2026-01-09 北京理工大学 一种一站式心脏磁共振检查技术
CN120579601B (zh) * 2025-05-30 2026-03-10 北京邮电大学 一种面向物联网云边协同的个性化联邦学习方法

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105869192A (zh) * 2016-03-28 2016-08-17 浙江大学 一种基于滑动窗的磁共振指纹识别重建技术
CN107194354A (zh) * 2017-05-23 2017-09-22 杭州师范大学 一种用于磁共振指纹成像的快速字典搜索方法
US20180292483A1 (en) * 2017-04-07 2018-10-11 Case Western Reserve University System and method for controlling effects of magnetic field inhomogeneities on magnetic resonance fingerprinting (mrf)
CN109073720A (zh) * 2016-04-26 2018-12-21 皇家飞利浦有限公司 静音磁共振指纹识别
CN110133553A (zh) * 2019-05-10 2019-08-16 浙江大学 一种超短回波时间磁共振指纹弛豫时间测量方法

Family Cites Families (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2009531705A (ja) * 2006-03-31 2009-09-03 コーニンクレッカ フィリップス エレクトロニクス エヌ ヴィ 超短ｔ２＊緩和測定法を用いた細胞測定のためのシステムおよび方法
US8008915B2 (en) * 2007-12-28 2011-08-30 Kabushiki Kaisha Toshiba Magnetic resonance imaging apparatus and magnetic resonance imaging method
EP2548039A1 (en) * 2010-03-18 2013-01-23 Koninklijke Philips Electronics N.V. Simultaneous and dynamic determination of longitudinal and transversal relaxation times of a nuclear spin system
US9513358B2 (en) * 2013-03-12 2016-12-06 Vaposun Inc. Method and apparatus for magnetic resonance imaging
JP6371554B2 (ja) * 2014-03-28 2018-08-08 株式会社日立製作所 磁気共鳴イメージング装置
US10261154B2 (en) * 2014-04-21 2019-04-16 Case Western Reserve University Nuclear magnetic resonance (NMR) fingerprinting tissue classification and image segmentation
CN104161517B (zh) * 2014-06-12 2016-06-15 中国科学院苏州生物医学工程技术研究所 一种用于mri对比度增强的成像方法
CN104215652B (zh) * 2014-08-21 2016-08-31 中国石油天然气股份有限公司 确定油气饱和度的方法和装置
CN104997512A (zh) * 2015-07-07 2015-10-28 中国科学院上海技术物理研究所 一种mri测温磁场漂移相位补偿方法
CN107015181B (zh) * 2017-04-07 2020-01-14 厦门大学 一种在不均匀磁场下测量质子纵向弛豫时间的方法
CN109633500B (zh) * 2018-12-18 2021-01-12 上海联影医疗科技股份有限公司 横向弛豫映射图的确定方法、装置以及磁共振成像设备
GB2580695B (en) * 2019-01-25 2021-01-20 Siemens Healthcare Ltd Method of reconstructing magnetic resonance image data

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105869192A (zh) * 2016-03-28 2016-08-17 浙江大学 一种基于滑动窗的磁共振指纹识别重建技术
CN109073720A (zh) * 2016-04-26 2018-12-21 皇家飞利浦有限公司 静音磁共振指纹识别
US20180292483A1 (en) * 2017-04-07 2018-10-11 Case Western Reserve University System and method for controlling effects of magnetic field inhomogeneities on magnetic resonance fingerprinting (mrf)
CN107194354A (zh) * 2017-05-23 2017-09-22 杭州师范大学 一种用于磁共振指纹成像的快速字典搜索方法
CN110133553A (zh) * 2019-05-10 2019-08-16 浙江大学 一种超短回波时间磁共振指纹弛豫时间测量方法

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
LI, SHA ET AL.: "Research progress on magnetic resonance with ultrashort echo time sequence", CHINESE JOURNAL OF MEDICAL PHYSICS, vol. 36, no. 1,, 31 January 2019 (2019-01-31), ISSN: 1005-202X, DOI: 20200707102928A *
XIAOZHI CAO ET AL.: "Robust Sliding-Window Reconstruction for Accelerating the Acquisition of MR Fingerprinting", MAGNETIC RESONANCE IN MEDICINE, vol. 78, no. 4,, 7 November 2016 (2016-11-07), XP055753411, ISSN: 0740-3194, DOI: 20200707102623A *

Also Published As

Publication number Publication date
CN110133553B (zh) 2020-06-05
US11353531B2 (en) 2022-06-07
CN110133553A (zh) 2019-08-16
US20220057463A1 (en) 2022-02-24

Similar Documents

Publication Publication Date Title
WO2020228641A1 (zh) 一种超短回波时间磁共振指纹弛豫时间测量方法
CN106716167B (zh) 用于评估结构空间频率的基于选择性采样磁共振的方法
CN103597370B (zh) 空间编码的相位对比磁共振成像
JP5917077B2 (ja) 磁気共鳴イメージング装置
CN113017596B (zh) 一种磁共振多参数定量方法及其应用
Li et al. Ultrashort echo time magnetic resonance fingerprinting (UTE‐MRF) for simultaneous quantification of long and ultrashort T2 tissues
US12181551B2 (en) Magnetic resonance fingerprinting imaging method with variable number of echoes
WO2007145193A1 (ja) 磁気共鳴イメージング装置
CN115137347B (zh) 三维超短回波时间磁共振指纹成像的髓鞘定量成像方法
CN115728691A (zh) 一种基于重叠回波的磁共振多参数同时定量成像方法及系统
JP2000157507A (ja) 核磁気共鳴撮影装置
EP4179344B1 (en) Method for generating a magnetic resonance synthetic computer tomography image
Wang et al. Fast magnetic resonance elastography with multiphase radial encoding and harmonic motion sparsity based reconstruction
CN102238909B (zh) 磁共振成像装置以及同步摄像方法
US11675029B2 (en) Retrospective tuning of soft tissue contrast in magnetic resonance imaging
CN117269866B (zh) 基于射频编码脉冲的高信噪比薄层t1参数定量成像方法
CN109839607A (zh) 一种基于变加速敏感度编码的cest图像重建方法和装置
JP4678916B2 (ja) 磁気共鳴イメージング装置
KR20160056271A (ko) 자기 공명 영상을 통해 대상체의 속성을 정량화하기 위한 방법 및 장치
Koopmans fMRI of cortical layers
Sayin Non-cartesian magnetic resonance imaging strategies: improvements in accelerated cardiac magnetic resonance imaging and technical considerations
Marsh Layer-fMRI acquisition and analysis
CN117665676A (zh) 一种双对比度超低场磁共振成像方法
Lin et al. Radial spoke energy for self-navigated motion detection and position-ordered dynamic musculoskeletal MRI
Rafiqul Islam Compressed Sensing Magnetic Resonance Imaging Using Fourier and Non-Fourier Based Bunched Phase Encoding

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: 20804793

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: 20804793

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 20804793

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 27/05/2022)

122 Ep: pct application non-entry in european phase

Ref document number: 20804793

Country of ref document: EP

Kind code of ref document: A1