A system, method, and readable medium for human glycogen magnetic resonance imaging

Abstract

The application discloses a system, a method and readable medium for human glycogen magnetic resonance imaging. The system comprises a data processor configured to generate a water proton signal intensity map of a relayed nuclear Overhauser effect (rNOE) exchange process between aliphatic protons in glycogen and free water protons in an imaging volume according to water proton signal intensity measurement in each voxel; and obtain a concentration map of glycogen molecules in the imaging volume by using calibration of the water proton signal intensity measurement of the rNOE exchange process. The application realizes non-invasive quantification and high spatiotemporal resolution imaging of glycogen in human tissues by using specific high-sensitivity imaging of glycogen through glycoNOE of the glycogen and magnetic coupling of the glycogen and water.

Description

Technical Field

[0001] This invention relates to the field of medical imaging, and more particularly to a system, method, and readable medium for human glycogen magnetic resonance imaging. Background Technology

[0002] Glycogen, the primary form of glucose storage in mammalian cells, plays a crucial role in maintaining plasma glucose levels and providing fuel for cellular activities and motor function. In humans, most (up to 90%) of absorbed glucose is deposited as muscle glycogen, with a normal concentration of approximately 80 mM (glucose units), while higher concentrations (200–500 mM) are stored in the liver. It is well known that human glycogen metabolism differs significantly from that in animals. To gain a detailed understanding of human glycogen metabolism and to conduct potential clinical assessments of related diseases, a non-invasive method is needed to directly observe human glycogen metabolism.

[0003] While blood glucose levels can be readily measured and used as diagnostic biomarkers, accurate, non-invasive measurement of tissue glycogen concentration is challenging in itself, yet of considerable interest for both basic science and clinical applications. Reliable measurement of liver glycogen content has been problematic due to its confined location within cells. Biochemical assays of glycogen following tissue biopsy are the oldest method, but their invasiveness and the potential for regional variability within the liver limit their clinical applicability.

[0004] For human glycogen imaging methods, the most commonly used is to use 13 The use of 12C magnetic resonance spectroscopy (MRS) to detect glycogen in vivo has greatly advanced the quantitative understanding of glycogen metabolism in animals and humans. It was demonstrated approximately 35 years ago that... 13 C10 NMR spectroscopy offers the ability to non-invasively measure glycogen in the liver and muscle. Numerous follow-up studies in patients with type 1 and type 2 diabetes have yielded valuable insights into the dysregulation of muscle and liver glycogen metabolism in diabetic states. Despite these advances, due to… 13 The low natural abundance and low gyromagnetic ratio of the C nucleus, and the typically required... 13 The cost of C-labeled isotopes, in vivo 13 CNMR spectroscopy remains hampered by its inherently low signal-to-noise ratio (SNR). Furthermore, the vast majority of clinical MRI scanners lack… 13 The clinical application of this C-detection technology may be limited to the research field.

[0005] Therefore, existing technologies still need to be improved and developed. Summary of the Invention

[0006] The inventors discovered that existing methods for detecting glycogen in the human body have the following two main problems:

[0007] On the one hand, due to 13 C MRS has low detection sensitivity, and total glycogen levels are typically measured in large volumes at natural abundance (1.1%) with low temporal and spatial resolution. 13 The additional hardware requirements of C-measurement further limit its widespread application. Therefore, a more sensitive method capable of rapid and localized detection of glycogen in the human body remains needed.

[0008] On the other hand, saturation transfer (ST) or chemical exchange saturation transfer (CEST) MRI has emerged as an emerging method, successfully achieving high-resolution detection of several metabolites, including glycogen, creatine (Cr), and phosphocreatine (PCr). In CEST MRI, the Z-spectrum (similar to...) 1 ¹H MR spectroscopy revealed enhancements in molecules from various sources, including those that exchange protons with water (e.g., -NH, -NH₂, -OH molecules show signal enhancement through proton exchange with water via hydroxyl groups). Glycogen CEST (glycoCEST) experiments have been previously explored, but the rapid exchange of hydroxyl protons presents a significant challenge to the accurate extraction of glycogen CEST signals from the spectrum. A glycogen imaging method using the rNOE mechanism (glycoNOE) was proposed in ST experiments and successfully applied to in vivo mapping of glycogen in mouse liver. However, validation of glycoNOE MRI for glycogen detection in humans remains lacking.

[0009] In view of the shortcomings of the prior art, the main objective of this invention is to verify the visibility of glycoNOE in the human body, which is of great help for the clinical application of high-resolution, specific detection of glycogen. This invention found that in the Z-spectrum of human calf muscle obtained at 5T, glycogen (at -1 ppm away from water) and creatine phosphate (PCr, at +1.95 ppm and +2.5 ppm) were clearly visible. Therefore, the quantification and mapping of glycogen NOE and PCr-CEST signals in human calf muscle were investigated. This provides an opportunity for non-invasive study of human energy metabolism (glycogen and PCr) at high spatial and temporal resolution.

[0010] The technical solution of the present invention is as follows:

[0011] A first aspect of the present invention provides a system for magnetic resonance imaging of human glycogen, such as Figure 1 As shown, it includes:

[0012] The main magnet 1 is configured to provide a sufficiently uniform magnetic field across the imaging volume;

[0013] Magnetic gradient coil 2 is configured to generate spatial coding in a magnetic field;

[0014] Radio frequency coil 3 is configured to acquire one or more water proton signal intensity measurements at each of a plurality of voxels within an imaging volume, wherein the water proton signal intensity measurements are acquired at each voxel at one or more irradiation frequencies that are one part per million (ppm) lower than the baseline frequency associated with free water protons.

[0015] The data processor 4 is configured to generate a water proton signal intensity map of the relay nucleus Orstäuhaus effect (rNOE) exchange process between aliphatic protons in glycogen and free water protons in the imaging volume, based on the water proton signal intensity measurement in each voxel; and to obtain a glycogen concentration map in the imaging volume by means of calibration of the water proton signal intensity measurement of the rNOE exchange process.

[0016] Furthermore, the measured water proton signal intensity was obtained after irradiating the imaging volume using a magnetically labeled pulse sequence.

[0017] Furthermore, the magnetic labeling pulse sequence is one of a saturation transfer pulse sequence, an inversion pulse sequence, and an excitation pulse sequence.

[0018] Furthermore, the data processor is further configured to generate a water proton signal intensity map based on multiple resonances associated with the magnetization transfer of aliphatic protons to free water via the rNOE process.

[0019] Furthermore, the imaging volume includes at least one of a tumor, brain, heart, liver, and skeletal muscle.

[0020] Furthermore, the one or more water proton signal intensity measurements are multiple signal intensity measurements, including one or more water proton signal intensity measurements obtained in each voxel at one or more additional irradiation frequencies, said additional irradiation frequencies being lower than the baseline frequency associated with free water protons;

[0021] The data processor is further configured to correct multiple water proton signal intensity measurements based on a mixed direct water saturation and magnetization transfer contrast (MTC) background.

[0022] Furthermore, the glycogen molecule concentration map is sensitive to the concentration of glycogen molecules within each voxel.

[0023] Furthermore, the measurement of water proton signal intensity is obtained before, during, or after the application of exogenous molecules to the imaging volume; wherein, the radio frequency coil is further configured to acquire one or more water proton signal intensity measurements at each of a plurality of voxels within the imaging volume after the application of exogenous molecules, and the data processor is further configured to generate a series of glycogen concentration maps within the imaging volume, the series of concentration maps describing the characteristics of the temporal changes in glycogen concentration within the imaging volume due to the application of exogenous glycogen.

[0024] Furthermore, the administration of exogenous molecules includes the injection of isotopically labeled or unlabeled agents, including polysaccharide molecules or those that can be metabolized or converted into glycogen in the human body.

[0025] Furthermore, the measurement of water proton signal intensity is obtained prior to intervention on endogenous glycogen within the imaging volume; the radio frequency coil is further configured to obtain one or more water proton signal intensity measurements for each of a plurality of voxels within the imaging volume after intervention, and the data processor is further configured to generate a series of concentration maps of glycogen in the imaging volume, the series of concentration maps describing the temporal change in glycogen concentration in the imaging volume due to intervention.

[0026] Furthermore, changes in glycogen concentration include both substantial changes in glycogen concentration and equivalent changes in H proton concentration in glycogen due to isotope labeling and chemical molecular labeling effects.

[0027] Furthermore, the intervention includes one of the following: administration of medicine or chemical compounds, surgery, exercise program, food intake, and fasting program.

[0028] A second aspect of the present invention provides a method for performing human glycogen magnetic resonance imaging based on the system described herein, comprising:

[0029] 1. Determine the linear relationship between glycogen signal and glycogen concentration.

[0030] In one example, glycogen solutions of different concentrations were prepared using phosphate-buffered saline (PBS). In in vitro MRI experiments, the samples were scanned using MRI at a 37°C water bath to establish a linear relationship between glycogen signal and glycogen concentration.

[0031] 2. Based on ST-Speed ​​MRI, glycoNOE is used to perform magnetic resonance imaging of human glycogen, acquiring complete or partial Z-spectrum data.

[0032] In one example, MRI scans of the lower legs were performed on five healthy male volunteers (aged 22 to 32). The experiment was conducted at the UIH Jupiter 5T. CEST experimental data were collected. Of course, other models and types of MRI scanners can be used to quantify and image human glycogen.

[0033] 3. Data Processing

[0034] Using computer code, Z-spectrum experimental data were analyzed: glycogen signals were extracted and quantified from the Z-spectrum; and the concentration distribution of glycogen at specific image locations was obtained using the signal-concentration linear relationship.

[0035] Z-spectrum was obtained using CEST imaging, and glycogen signal was extracted and quantified from the Z-spectrum. Two-step multi-pool Lorentz fitting measurements were used to extract the glycogen signal.

[0036] This invention provides a magnetic resonance imaging (MRI) method that enables quantitative and high-resolution imaging of glycogen in human tissues. Several methods have been developed to determine glycogen concentration in human tissues in order to assess the role of glycogen in liver and brain energy metabolism. One of the most widely used in vivo glycogen measurement techniques is... 13 C10 nuclear magnetic resonance (NMR) spectroscopy. It can be compared with... 13 C-glucose administration, used in combination, allows for the non-invasive assessment of in vivo temporal evolution of glycogen metabolism in the liver and brain of animals or humans over several hours. However, due to its relatively low sensitivity, achieving high spatial resolution (i.e., better than approximately 0.5 cm⁻¹) is not possible. 3 The lack of glycogen resonance is a key issue limiting its application in human glycogen imaging. Currently, non-invasive detection and imaging of human glycogen remains a significant challenge. This invention utilizes the Overhauser effect (glycoNOE) of glycogen relay nuclei to achieve specific and highly sensitive imaging of glycogen through magnetic coupling between glycogen and water, realizing non-invasive quantification and high spatiotemporal resolution imaging of glycogen in human tissues.

[0037] A third aspect of the invention provides a non-transitory machine-readable medium comprising computer-executable instructions for glycogen magnetic resonance imaging, wherein, when executed by a computer, the computer configures a main magnet to provide a sufficiently uniform magnetic field across the imaging volume; configures a magnetic gradient coil to generate spatial coding in the magnetic field; configures a radio frequency coil to acquire one or more water proton signal intensity measurements at each of a plurality of voxels within the imaging volume, wherein the signal intensity measurements are acquired at each voxel at one or more irradiation frequencies lower than a baseline frequency associated with free water protons by one part per million; configures a data processor to generate a water proton signal intensity map of a relay nucleus Orstäuhaus effect (rNOE) exchange process between aliphatic protons in polysaccharide molecules and free water protons in the imaging volume, based on the water proton signal intensity measurements in each voxel; and generates a concentration map of glycogen molecules in the imaging volume using calibration of the water proton signal intensity measurements of the rNOE exchange process.

[0038] Furthermore, the measured water proton signal intensity was obtained after irradiating the imaging volume using a magnetically labeled pulse sequence.

[0039] Furthermore, the magnetic labeling pulse sequence is one of a saturation transfer pulse sequence, an inversion pulse sequence, and an excitation pulse sequence.

[0040] Furthermore, the generation of the water proton signal intensity map is based on multiple resonances related to the magnetization transfer of aliphatic protons to free water via the rNOE process.

[0041] Furthermore, glycogen molecules include one of the following: chemically modified polysaccharide molecules, labeled polysaccharide molecules, polysaccharides linked to binding substrates, endogenous polysaccharides, and exogenous polysaccharides.

[0042] Furthermore, the imaging volume includes at least one of a tumor, brain, heart, liver, and skeletal muscle. The tumor can be an in vivo tumor or an ex vivo human tumor; the brain can be an in vivo human brain or an ex vivo human brain; the heart can be an in vivo human heart or an ex vivo human heart; the liver can be an in vivo human liver or an ex vivo human liver; and the skeletal muscle can be an in vivo human skeletal muscle or an ex vivo human skeletal muscle.

[0043] Further, the one or more water proton signal intensity measurements are multiple signal intensity measurements, including one or more water proton signal intensity measurements obtained in each voxel at one or more additional irradiation frequencies, said additional irradiation frequencies being lower than the baseline frequency associated with free water protons; wherein, when executed by a computer, computer-executable instructions further cause the computer to correct the multiple water proton signal intensity measurements based on a mixed direct water saturation and magnetization transfer contrast (MTC) background.

[0044] Furthermore, the glycogen molecule concentration map is sensitive to the concentration of glycogen molecules within each voxel.

[0045] Furthermore, the water proton signal intensity measurement is obtained before, during, or after the application of the exogenous molecule to the imaging volume; wherein, when executed by the computer, the computer may execute instructions to further cause the computer to obtain one or more water proton signal intensity measurements at each of a plurality of voxels within the imaging volume after the application of the exogenous polysaccharide; and to generate a series of concentration maps of the polysaccharide molecules within the imaging volume, the series of concentration maps describing the characteristics of the temporal change in polysaccharide concentration within the imaging volume due to the application of the exogenous polysaccharide.

[0046] Furthermore, the application of the exogenous polysaccharide includes the injection of a contrast agent comprising polysaccharide molecules.

[0047] Furthermore, the measurement of water proton signal intensity is obtained prior to intervention with endogenous glycogen within the imaging volume, which alters the endogenous glycogen within the imaging volume. During computer execution, computer-executable instructions further enable the computer to obtain one or more water proton signal intensity measurements at each of a plurality of voxels within the imaging volume after the intervention, and to generate a series of concentration maps of polysaccharide molecules in the imaging volume, the series of concentration maps describing the temporal change in polysaccharide concentration within the imaging volume due to the intervention.

[0048] Furthermore, the intervention includes one of the following: administration of a drug or chemical compound, surgery, exercise program, food intake, and fasting program.

[0049] Furthermore, the distribution of glycogen concentration in the body or the change in concentration distribution over time can be used as an indicator for assessing human metabolic function and diagnosing diseases. Attached Figure Description

[0050] Figure 1 This is a schematic diagram of a system structure for human glycogen magnetic resonance imaging.

[0051] Figure 2 MTR′ shown in A asym This shows the mixture of glycogen signal (at -1 ppm, glycoNOE) and the CEST effect. The ROI of the Z-spectrum is shown on the T2 image; Figure 2 The obtained Z-spectrum (black circled line) is fitted with a multi-pool Lorentz line shape, providing the low-field background spectrum (black line), as well as the extracted +1.95ppm signal (gray area) and +2.5ppm signal (gray area), which are shown at the bottom. Then, the corrected MTR′ is constructed by subtracting the Z-spectrum obtained from the upper field from the low-field background spectrum. asym (cMTR′ asym (Triangle line, playing area). Figure 2 (C) through cMTR′ asym The spectrum was fitted to extract the glycogen signal, which was displayed in multiple Lorenz shapes with a gray area at the bottom.

[0052] Figure 3 ST-S MRI of glycogen, Cr, and PCr at 5T (pH 7.3, 37℃) in vitro. glycoNOE ( Figure 3 (A), PCr () Figure 3 (B) and Cr ( Figure 3 The C) signal was quantized from the Z spectrum, B1 = 0.2 μT. The PCr signal was extracted using a single-step multi-pool Lorentz fit. The glycogen signal was extracted using a two-step multi-pool Lorentz fit. Figure 3In the figure, D and E represent the relationship between the measured glycoNOE and PCr signals and their concentrations. Figure 3 F represents the relationship between the measured PCr and glycoNOE signals and B1.

[0053] Figure 4 The values ​​represent glycoNOE signals of -1 ppm, +1.95 ppm, and +2.5 ppm in in vivo skeletal muscle imaging. Figure 4 In the study, a two-step multi-pool Lorentz fitting strategy was used to perform Z-spectrum analysis on the skeletal muscle of volunteers and extract glycoNOE, +1.95ppm, and +2.5ppm signals. Figure 4 The relationship between glycoNOE, +1.95ppm, and +2.5ppm signals in the B-Z spectrum and the B1 power (n=5, ROI is similar to that of plate C). Figure 4 In the middle, C is a T2 image of the human calf muscle, showing the location of the ROI used for Z-spectrum analysis. Figure 4 D and E in the diagram represent B0 and B1, respectively. Figure 4 FH represents the signal map of glycoNOE at +1.95 ppm and +2.5 ppm (for the same object in group A) obtained by fitting voxels individually. B1 = 0.2 μT. Detailed Implementation

[0054] To make the objectives, technical solutions, and effects of this invention clearer and more explicit, the invention is further described in detail below. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0055] This embodiment provides a method for magnetic resonance imaging of human glycogen, the specific steps of which are as follows:

[0056] 1. Determine the linear relationship between glycogen signal and glycogen concentration.

[0057] Various concentrations of Cr (10, 20, 30, 40, and 50 mM, Sigma, St. Louis), PCr (7, 13, 19, 26, and 32 mM, Sigma, St. Louis), and rabbit liver glycogen (20, 40, 60, 80, 100, 200, 300, and 400 mM, Sigma, St. Louis) were prepared in phosphate-buffered saline (PBS) at pH 7.3. During MRI scans, the sample temperature was maintained at 37°C using a water bath. The linear relationship between glycogen magnetic resonance signal and glycogen concentration was determined.

[0058] 2. Saturation-transfer (ST) MRI experiment

[0059] Five healthy male subjects (aged 22 to 32) were studied after obtaining informed consent from the volunteers. All MRI experiments were performed on a UIH Jupiter 5T human scanner (United Imaging Healthcare, Shanghai, China). In the phantom experiments, a 24-channel pure-receive body array coil and a 48-channel pure-receive spine coil were used together. For the human calf muscles, a 24-channel transmit / receive knee joint coil was used. In each experiment, multiple CEST full Z-spectrums with different radio frequency pulse (B1) powers (0.1 to 1 μT) were acquired for a selective slice (5 mm thickness for human, 10 mm thickness for phantom imaging). A 3-s continuous wave (CW) pulse was applied in each 4 s (human) or 7 s (phantom) TR, followed by a single fast spin echo sequence (FSE) readout. The image resolution was set to 96 × 96, and the field of view (FOV) was adjustable. The saturation frequency was -5 to +5 ppm with a step size of 0.1 ppm. After the CEST scan, B1 plots are obtained on selected slices using the Dream sequence.

[0060] 3. Data Processing

[0061] The following combination Figure 2 To elaborate

[0062] All Z-spectrums were analyzed on a voxel-by-voxel basis. First, the B0 shift of the Z-spectrum was corrected using the WASSR method. To compensate for B1 inhomogeneities, a B1 correction method based on the Z-spectrum was performed, followed by smoothed spinal interpolation. Then, a two-step multi-pool Lorentz fitting was developed to extract PCr and glycoNOE signals from the Z-spectrum, such as... Figure 2 As shown.

[0063] Since the -1 ppm glycogen signal is not directly visible in the Z-spectrum of muscle under 5T conditions, it is assumed that this signal is masked by the direct water saturation (DS) background. As a first attempt to visualize the glycogen signal, the negative MTR′ spectrum was calculated. asym ,like Figure 2 (As shown in A).

[0064]

[0065] MTR asym (Δω) is the MTR asymmetric ratio spectrum, which has been widely used in CEST data analysis; Δω is the chemical shift value of the reference water position; S(Δω) is the water intensity after a pre-saturation pulse at the offset frequency Δω. In MTR′ asym A peak of -1 ppm can be seen in the spectrum; however, it may be contaminated by CEST peaks of +1.95 ppm and +2.5 ppm.

[0066] Therefore, a two-step multi-cell Lorentz fitting strategy was employed to extract glycoNOE, +1.95 ppm, and +2.5 ppm signals. First, contamination was avoided by estimating the background (+Δω) of the low-field Z-spectrum using a 6-cell Lorentz fitting (including water, +1.95 ppm, +2.5 ppm, +3.5 ppm, and MTC). 5 ppm, -3.5 ppm, and MTC cells provided signals for +1.95 ppm and +2.5 ppm, and a corrected MTR′. asym (cMTR′ asym This means subtracting the upper Z-spectrum from the "background (Δω)". For example... Figure 2 As shown in B.

[0067]

[0068] Then, cMTR′ asym The spectra were further fitted using a 4-cell Lorentz method, assuming peaks around -1 ppm, -2.8 ppm, -3.8 ppm, and the MTC cell (e.g., Figure 2 (As shown in C). The signals for glycogen (glycoNOE), +1.95 ppm, and +2.5 ppm were estimated as integrals of peak values ​​from -0.6 ppm to -1.6 ppm, 1.5 ppm to 2.5 ppm, and 2 ppm to 3 ppm, respectively. (as shown in C). Figure 2 (As shown in B and C).

[0069] For in vitro PCr and glycogen, PCr signals were obtained by fitting 2-pool Lorentzian linear curves with centers around +1.95 ppm and +2.5 ppm, respectively. Glycogen signals were obtained by fitting cMTR′. asym The spectrum was estimated using a single Lorentz line center around -1 ppm. Data processing was performed using a custom script on a computer.

[0070] 4. The data results are as follows:

[0071] To assess the concentration dependence of glycogen, Cr, and PCr signals at 5T, ST-MRI experiments were first performed on in vitro models with different molar concentrations at pH 7.3 and 37℃. Figure 3 As shown. The Z-spectrum can detect a glycoNOE signal of -1 ppm (e.g. Figure 2 As shown in Figure A), and PCr signals of +1.95ppm and +2.5ppm (as shown in Figure A). Figure 2 (As shown in B). Cr exhibits a broad peak of +1 to +2 ppm in the Z spectrum, indicating a moderate to rapid exchange rate (e.g., as shown in B). Figure 3(As shown in C). After Lorentz fitting, the signal was quantized as the integral of the fitted Lorentz line shape. It was found that the PCr and glycoNOE signals were linearly correlated with molecular concentration within the study range (e.g., ...). Figure 3 (As shown in D and E). When [glycogen] = 80 mM, the glycogen signal was found to be 11.4%*ppm, and the PCr (+1.95 ppm) signal was 5.0%*ppm. When [PCr] = 32 mM, the PCr (+1.95 ppm) signal was 7.0%*ppm. Within the low B1 detection range, the signal in the solution increased with increasing B1 (e.g., ...). Figure 3 (As shown in D).

[0072] Then, ST-S MRI experiments were performed on the skeletal muscle of five healthy male volunteers. This was achieved using a two-step Lorentz fitting method. Figure 2 Signals of -1 ppm (assigned as glycoNOE), +1.95 ppm (from PCr), and +2.5 ppm (from PCr) were extracted from multiple B1 levels in five volunteers. Figure 4 As shown. Within the experimentally measured B1 range, the PCr CEST peak continuously increased with increasing B1 at +1.95 ppm and +2.5 ppm. However, glycoNOE reached its peak at B1 = 0.2 μT, but subsequently decreased with further increases in B1. Therefore, B1 = 0.2 μT was used in subsequent experiments. After fitting each voxel, Figure 4 Mid-CD imaging showed simultaneous imaging of glycogen, +1.95 ppm, and +2.5 ppm signals in skeletal muscle at B1 = 0.2 μT. The image shows the relatively uniform distribution of glycoNOE, +1.95 ppm, and +2.5 ppm signals in the human calf muscle. The combined signal intensities (B1 = 0.2 μT) of the glycoNOE, +1.95, and +2.5 ppm peaks in the five subjects were 10.6 ± 1.7% * ppm, 5.3 ± 1.2% * ppm, and 9.9 ± 1.1% * ppm, respectively (e.g., ...). Figure 4 (As shown in B), these correspond to signal intensities of 74.1±11.7 mM glycogen, 31.7±7.3 mM glycogen, and 42.4±4.8 mM MPCr in in vitro solutions, respectively.

[0073] In summary, this invention provides the first-ever imaging of glycogen in human skeletal muscle. Glycogen is one of the most abundant energy substrates, widely distributed across various tissues. The sustained supply of intracellular ATP is maintained through several metabolic processes, including the breakdown of PCr into Cr and Pi, the breakdown of glycogen into lactate, and the oxidative phosphorylation of carbohydrates (glycogen and glucose). Abnormalities in energy metabolism manifest in numerous diseases, including metabolic disorders, mitochondrial diseases, and ischemia in various organs such as muscle, heart, brain, and liver. Non-invasive measurements of these energy substrates at high spatiotemporal resolution may contribute to a local understanding of fundamental cellular energy dynamics and the assessment of disease progression.

[0074] Based on in vitro data, the glycogen concentration in human skeletal muscle measured in this invention is approximately 74.1 ± 11.7 mM. These values ​​are consistent with previously reported values ​​obtained through... 13 CMR and biopsy showed human muscle glycogen concentrations of 60 to 110 mM, and through... 31 P MRS and biopsy-detected PCr are approximately 27 to 40 mM consistent. For glycogen, it is known that the glycogen signal is also affected by particle size, and selecting glycogen samples with a similar size distribution to calibrate human data is not easy. Commercial rabbit liver glycogen (Sigma, St. Louis) has been shown to have a size similar to that under in vivo conditions and is therefore used as a reference. PCr shows peaks of +1.95 ppm and +2.5 ppm in the Z-spectrum (…). Figure 3 Previous studies have confirmed that the +2.5 ppm signal in the muscle Z-spectrum primarily originates from PCr. It has been suggested that the +1.95 ppm peak under ultra-high magnetic fields might be caused by PCr + Cr; however, in the 5T experiments of this invention, in vitro data showed a broad Cr signal, which is unlikely to contribute to an isolated +1.95 ppm peak. Therefore, the +1.95 ppm and +2.5 ppm peaks in the in vivo Z-spectrum should primarily originate from PCr.

[0075] As is well known, glycogen Z-spectrum simultaneously exhibits the CEST effect (glycoCEST) from glycogen hydroxyl protons and the NOE effect (glycoNOE) from glycogen aliphatic protons, the latter located around +1 ppm. The broad glycogen CEST effect is challenging to separate due to rapid chemical exchange, especially at lower (5T or 3T) magnetic fields, where it condenses with water peaks. In contrast, the glycogen NOE signal peak at low B1 power can be successfully separated by a two-step Lorentz fit, as shown here. Increasing B1 enhances the glycogen NOE effect in solution. Figure 3Because of its high saturation efficiency, however, the overwhelming MT effect of surrounding tissues and water peaks causes the glycogen peak to merge with the background signal at high B1. Therefore, a low-power B1 (=0.2uT) was determined to be the optimal power for extracting human glycogen signals. At 5T, the inhomogeneity of B1 ( Figure 3 The presence of glycogen (Cp) is also a problem, but its impact on Z-spectrum quantification can be largely compensated for by multiple Z-spectrum datasets. After correction, the concentration maps of glycogen and PCr show a relatively uniform distribution of calf muscle at rest. Figure 4 ).

[0076] Compared with traditional natural abundance 13 Compared to C MRS, based on 1 The glycoNOE method of H exhibits a detection sensitivity gain of 3 to 4 orders of magnitude in detecting total glycogen in the human body. Only 1.1% (natural abundance) of glycogen C1 (single site) is detectable. 13 Glycogen was detected in CMRS, with 100% detection of the four aliphatic protons (H3, H5, H2+H4-1), representing an enhancement of two orders of magnitude compared to proton MRS. This allows for relatively rapid imaging of glycogen at high spatial resolution using saturation transfer MRI. In the current study, a complete Z-spectrum (101 points, from -5 ppm to +5 ppm) was collected in 6 minutes and 48 seconds, with a spatial resolution of approximately 5 mm × 1.5 mm × 1.5 mm. In principle, only a partial Z-spectrum is sufficient to map the distribution of glycogen and PCr to reduce scan time. For example, acquiring a Z-spectrum from -3.5 ppm to +2.5 ppm in a step size of 0.2 ppm (30 points) should be sufficient to map the substrates (glycogen and PCr) well at high spatial resolution.

[0077] It should be understood that the application of the present invention is not limited to the examples above. Those skilled in the art can make improvements or modifications based on the above description, and all such improvements and modifications should fall within the protection scope of the appended claims.

Claims

1. A system for quantitative magnetic resonance imaging of human skeletal muscle glycogen under a 5T magnetic field strength, characterized in that, include: The main magnet is configured to provide a magnetic field over the imaging volume; Magnetic gradient coils are configured to generate spatial codes in a magnetic field; An RF coil is configured to acquire one or more water proton signal intensity measurements at each of a plurality of voxels within an imaging volume, wherein the water proton signal intensity measurements are acquired at each voxel at one or more irradiation frequencies below a baseline frequency associated with free water protons, and the power B1 of the irradiation frequency is 0.2 μT. The system and data processor are configured to, based on the water proton signal intensity measurements in each voxel, employ a two-step multi-pool Lorentz fitting to extract a water proton signal intensity map of the relay nucleus Auschwitz effect exchange process between aliphatic protons in glycogen and free water protons in the imaging volume from the Z-spectrum, wherein the relay nucleus Auschwitz effect is denoted as rNOE; and to obtain a concentration map of glycogen molecules in the imaging volume by calibrating the water proton signal intensity measurements of the rNOE exchange process using an in vitro model based on a similarity to the particle size distribution of endogenous glycogen in human tissue.

2. The system according to claim 1, characterized in that, The water proton signal intensity measurement was obtained after irradiating the imaging volume using a magnetically labeled pulse sequence.

3. The system according to claim 2, characterized in that, The magnetic labeling pulse sequence is one of the following: saturation transfer pulse sequence, inversion pulse sequence, and excitation pulse sequence.

4. The system according to claim 1, characterized in that, The data processor is further configured to generate a water proton signal intensity map based on multiple resonances associated with the magnetization transfer of aliphatic protons to free water via the rNOE process.

5. The system according to claim 1, characterized in that, The imaging volume includes at least one of a tumor, brain, heart, liver, and skeletal muscle.

6. The system according to claim 1, characterized in that, The one or more water proton signal intensity measurements are multiple signal intensity measurements, including one or more water proton signal intensity measurements obtained in each voxel at one or more additional irradiation frequencies, which are lower than the baseline frequency associated with free water protons; The data processor is further configured to correct multiple water proton signal intensity measurements based on a mixed direct water saturation and magnetization transfer contrast background.

7. The system according to claim 1, characterized in that, The glycogen molecule concentration map is sensitive to the concentration of glycogen molecules within each voxel.

8. The system according to claim 1, characterized in that, The measurement of water proton signal intensity is obtained before, during, or after the application of an exogenous molecule to the imaging volume; wherein, the radio frequency coil is further configured to acquire one or more water proton signal intensity measurements at each of a plurality of voxels within the imaging volume after the application of the exogenous molecule, and the data processor is further configured to generate a series of glycogen concentration maps within the imaging volume, the series of concentration maps describing the characteristics of the temporal change in glycogen concentration within the imaging volume due to the application of exogenous glycogen.

9. The system according to claim 8, characterized in that, The administration of exogenous molecules includes the injection of isotopically labeled or unlabeled agents, which may include polysaccharide molecules or those that can be metabolized or converted into glycogen in the human body.

10. The system according to claim 1, characterized in that, The measurement of water proton signal intensity is obtained prior to intervention on endogenous glycogen in the imaging volume; the radio frequency coil is further configured to obtain one or more water proton signal intensity measurements for each of a plurality of voxels in the imaging volume after intervention, and the data processor is further configured to generate a series of concentration maps of glycogen in the imaging volume, the series of concentration maps describing the temporal change in glycogen concentration in the imaging volume due to intervention.

11. The system according to claim 10, characterized in that, The intervention includes one of the following: medication, surgery, exercise program, food intake, and fasting program.

12. A method for performing human glycogen magnetic resonance imaging using the system as described in any one of claims 1-11, characterized in that, Including the following steps: Determine the linear relationship between glycogen signal and glycogen concentration; Based on ST MRI, glycoNOE is used to perform magnetic resonance imaging of human glycogen, acquiring complete or partial Z-spectrum data; Z-spectrum data were analyzed to extract and quantify glycogen signals. The linear relationship between glycogen signals and glycogen concentration was used to obtain the concentration distribution of glycogen at specific image locations.

13. A non-transitory machine-readable medium, characterized in that, Includes computer-executable instructions for glycogen magnetic resonance imaging, which, when executed by a computer, cause the computer to configure a main magnet to provide a magnetic field over the imaging volume; and to configure magnetic gradient coils to generate spatial coding in the magnetic field; A radio frequency coil is configured to acquire one or more water proton signal intensity measurements at each of a plurality of voxels within an imaging volume, wherein the signal intensity measurements are acquired at each voxel at one or more irradiation frequencies lower than the baseline frequency associated with free water protons; a data processor is configured to generate a water proton signal intensity map of the rNOE exchange process between aliphatic protons in glycogen and free water protons in the imaging volume, based on the water proton signal intensity measurements in each voxel; and a concentration map of glycogen molecules in the imaging volume is generated using calibration of the water proton signal intensity measurements of the rNOE exchange process.

14. The non-transitory machine-readable medium according to claim 13, characterized in that, The water proton signal intensity measurement was obtained after irradiating the imaging volume using a magnetically labeled pulse sequence.

15. The non-transitory machine-readable medium according to claim 14, characterized in that, The magnetic labeling pulse sequence is one of the following: saturation transfer pulse sequence, inversion pulse sequence, and excitation pulse sequence.

16. The non-transitory machine-readable medium according to claim 13, characterized in that, The generation of the water proton signal intensity map is based on multiple resonances related to the magnetization transfer of aliphatic protons to free water via the rNOE process.

17. The non-transitory machine-readable medium of claim 13, wherein the polysaccharide molecule comprises one of glycogen molecules, chemically modified polysaccharide molecules, labeled polysaccharide molecules, polysaccharides linked to a binding substrate, endogenous polysaccharides, and exogenous polysaccharides.

18. The non-transitory machine-readable medium of claim 13, wherein the imaging volume comprises at least one of a tumor, brain, heart, liver, and skeletal muscle.

19. The non-transitory machine-readable medium of claim 13, wherein the one or more water proton signal intensity measurements are a plurality of signal intensity measurements, including signal intensity measurements obtained in each voxel at one or more additional irradiation frequencies, said additional irradiation frequencies being lower than the baseline frequency associated with free water protons; wherein, When executed by the computer, the computer-executable instructions further enable the computer to correct multiple water proton signal intensity measurements based on the mixed direct water saturation and magnetization transfer contrast background.

20. The non-transitory machine-readable medium according to claim 13, characterized in that, The glycogen molecule concentration map is sensitive to the concentration of glycogen molecules within each voxel.

21. The non-transitory machine-readable medium according to claim 13, characterized in that, The water proton signal intensity measurement is obtained before, during, or after the application of the exogenous molecule to the imaging volume; wherein, when executed by the computer, the computer may execute instructions to further cause the computer to obtain one or more water proton signal intensity measurements at each of a plurality of voxels within the imaging volume after the application of the exogenous polysaccharide; and to generate a series of concentration maps of the polysaccharide molecules within the imaging volume, the series of concentration maps describing the characteristics of the temporal change in polysaccharide concentration within the imaging volume due to the application of the exogenous polysaccharide.

22. The non-transitory machine-readable medium according to claim 21, characterized in that, The application of the exogenous polysaccharide includes the injection of a contrast agent comprising polysaccharide molecules.

23. The non-transitory machine-readable medium according to claim 13, characterized in that, The measurement of water proton signal intensity is obtained prior to intervention in the endogenous glycogen within the imaging volume, which alters the endogenous glycogen within the imaging volume. During computer execution, computer-executable instructions further enable the computer to obtain one or more water proton signal intensity measurements at each of a plurality of voxels within the imaging volume after the intervention, and to generate a series of concentration maps of polysaccharide molecules in the imaging volume, which describe the temporal change in polysaccharide concentration within the imaging volume due to the intervention.

24. The non-transitory machine-readable medium according to claim 23, characterized in that, The intervention includes one of the following: medication, surgery, exercise program, food intake, and fasting program.

Images