Method and apparatus for processing and visualizing magnetic resonance imaging of cerebral blood flow in spinocerebellar degeneration

Abstract

The application can detect the whole brain blood flow change of the patient with spinocerebellar degeneration and optimize the cerebellum visualization. The application carries out CBF reconstruction on ASL data, carries out nonlinear co-registration on the whole brain CBF image to the PET template, and spatially standardizes to the MNI space, divides the CBF value of each voxel by the whole brain average value to obtain the normalized CBF and smooth; construct a general linear model in SPM8, and carry out whole brain statistical analysis at the voxel level; extract the t value image of the significant cluster and its mask, and use the mask to extract the cluster average CBF value for regional analysis. The significant cluster t value image of the cerebellum is overlaid to the coronal SUIT template, the t value image is mapped to the SUIT plane graph in the SUIT toolbox, and the outline of the t value plane graph is overlaid to the cerebellar anatomical and functional orientation plane graph respectively, so that the anatomical and functional detail visualization of the cerebellar region is realized.

Description

Technical Field

[0001] This invention relates to the field of medical image processing technology, and in particular to a method and device for processing and visualizing magnetic resonance imaging of cerebral blood flow in spinocerebellar degeneration. Background Technology

[0002] Several positron emission tomography (PET) and single-photon emission computed tomography (SPECT) studies have demonstrated reduced cerebral blood flow (CBF) in individuals with spinocerebellar degeneration (SCD). Although PET is considered the gold standard for CBF measurement, the ionizing radiation and contrast agent injection involved in PET / SPECT are bottlenecks limiting its widespread clinical application.

[0003] Today, as a completely non-invasive and radiation-free magnetic resonance imaging (MRI) technique, arterial spin labeling (ASL) imaging, which uses magnetically labeled blood instead of exogenous contrast agents as flow tracers, is gradually becoming an alternative method for measuring blood flow, providing MR-based CBF quantification. Currently, pseudo-continuous arterial spin labeling (PCASL) is the preferred labeling method in clinical practice due to its high signal-to-noise ratio.

[0004] ASL MRI exhibits good correlation and concordance with PET and SPECT, and can be used to assess abnormal cerebral perfusion in neurodegenerative diseases such as Parkinson's disease (PD) and multiple system atrophy (MSA), but it is rarely used in the study of spinal cord-cerebellar ataxia (SCD). A previous retrospective study assessed cerebellar blood flow in SCD patients using PCASL MRI measurements, and another applied ASL to spinocerebellar ataxia type 3 (SCA3) and hereditary spastic paraplegia (HSP). However, these studies only analyzed CBF values ​​extracted from predefined regions of interest (ROIs) without considering the whole brain. As a data-driven approach, voxel-based whole-brain analysis can investigate the spatial specificity of CBF changes without being limited to specific brain regions. Furthermore, given the importance of visualization in neuroimaging, CBF results located in the infratentorial space particularly require more appropriate presentation. Summary of the Invention

[0005] To overcome the shortcomings of existing technologies, the technical problem to be solved by this invention is to provide a method for processing and visualizing cerebral blood flow magnetic resonance imaging in spinocerebellar degeneration. This method can accurately detect spatially specific changes in cerebral blood flow (CBF) patterns throughout the whole brain (rather than limited to specific brain regions) in patients with spinocerebellar degeneration. It can automatically perform voxel-based and region-based quantitative analysis of standardized CBF values. At the same time, it optimizes the visualization of significant cerebellar clusters, which is beneficial for displaying the anatomical and functional details of affected cerebellar lobules and nuclei.

[0006] The technical solution of this invention is: a method for processing and visualizing magnetic resonance imaging of cerebral blood flow in spinocerebellar degeneration, comprising the following steps:

[0007] (1) 3D PCASL data were acquired using a 3.0T MR scanner equipped with an 8-channel head coil;

[0008] (2) After subtracting the labeled image from the ASL contrast image, the ASL difference image is calculated, and then the CBF image is generated by GE Functool infusion software;

[0009] (3) The CBF images of all subjects were nonlinearly co-registered to the PET perfusion template using a one-step registration method, and then spatially warped to the standard MNI space while resampling the voxel size.

[0010] (4) Based on hypothesis testing, a general linear model was constructed and a two-sample t-test was used to perform voxel-level statistical analysis on the standardized CBF images. Age and gender were used as covariates in the statistical model, and the whole-brain mask was used to limit the scope of CBF analysis to the whole-brain level.

[0011] (5) Extract the t-value map of each significant result block of the whole brain analysis in the xjView toolbox and save the corresponding block as a binary mask; for each subject, load the mask into the DPABI toolbox, extract and calculate the average standardized CBF value of the ROI of each significant result block;

[0012] (6) In MRICron, the t-value map of the significant mass located in the cerebellum is overlaid on the cerebellum SUIT template, and the cerebral blood flow involvement of the cerebellar lobules and nuclei is displayed in detail in the coronal view; in the SUIT toolbox of SPM, the t-value map of the significant mass of the cerebellum is mapped to the surface-based cerebellum SUIT template to obtain the t-value plane map of the cerebellar mass, and then the outline of the t-value plane map is overlaid on the cerebellar anatomical and functional orientation plane maps respectively.

[0013] This invention acquires three-dimensional pseudo-continuous arterial spin labeling (3D PCASL) data of spinocerebellar degeneration patients and healthy controls using a 3.0T MR scanner equipped with an 8-channel head coil. Brain blood flow (CBF) reconstruction is performed on the raw 3D PCASL data. Subsequently, all subjects' whole-brain CBF images are nonlinearly co-registered to a PET template and spatially normalized to the MNI standard space. The CBF value of each voxel in the whole brain is then divided by the whole-brain mean to normalize the data, achieving CBF image standardization. The standardized CBF images are then smoothed. A general linear model is constructed within SPM8 to perform voxel-level whole-brain statistical analysis of the CBF images to accurately obtain whole-brain CBF changes. Within xjView, t-value maps of significant clusters obtained after whole-brain statistical analysis are extracted. The process involves several steps: First, a block mask is used. Within DPABI, the CBF values ​​of the corresponding cerebellar masses are extracted using the mask. Then, within MRICron, the t-value maps of significant cerebellar masses are overlaid on the cerebellar SUIT template. In the coronal view, this allows for a more detailed visualization of CBF involvement in the cerebellar lobules and nuclei. Finally, within the SUIT toolbox, the t-value maps of significant cerebellar masses are mapped onto a surface-based cerebellar SUIT template to obtain a t-value planar map of the cerebellar masses. The outline of this t-value planar map is then overlaid on the cerebellar anatomical and functional orientation maps to further illustrate the anatomical distribution and functional impairment of the affected cerebellar lobules. This entire process requires no manual drawing and automatically performs voxel- and region-based quantitative analysis of cerebral blood flow, enabling visualization of the anatomical and functional details of the cerebellar regions.

[0014] A data processing device for altering patterns of whole-brain cerebral blood flow (CBF) is also provided, comprising:

[0015] The data acquisition module is configured to acquire 3D PCASL data using a 3.0T MR scanner equipped with an 8-channel head coil;

[0016] The CBF image generation module is configured to subtract the labeled image from the ASL contrast image, calculate the ASL difference image, and then generate the CBF image using GE Functool infusion software;

[0017] The spatial normalization processing module is configured to use a one-step registration method to nonlinearly co-register all subjects' CBF images to the PET perfusion template, then spatially warp to the standard MNI space, while resampling the voxel size.

[0018] The voxel-based whole-brain standardized CBF data analysis module is configured to construct a general linear model based on hypothesis testing and use a two-sample t-test to perform voxel-level statistical analysis on the standardized CBF images. Age and gender are used as covariates in the statistical model, and a whole-brain mask is used to limit the scope of CBF analysis to the whole-brain level.

[0019] The region analysis module is configured to extract the t-value map of each significant outcome clique from the above whole-brain analysis in the xjView toolbox and save the corresponding clique as a binary mask; for each subject, the mask is loaded into the DPABI toolbox to extract and calculate the average standardized CBF value of the ROI of each significant outcome clique.

[0020] The cerebellar salience mass visualization module is configured to, within MRICron, overlay the t-value map of a salience mass located in the cerebellum onto a cerebellar SUIT template, displaying detailed cerebral blood flow involvement in the cerebellar lobules and nuclei in a coronal view; within the SUIT toolbox of SPM, the t-value map of the cerebellar salience mass is mapped onto a surface-based cerebellar SUIT template to obtain a t-value planar map of the cerebellar mass, and then the outline of the t-value planar map is overlaid onto the cerebellar anatomical and functional orientation planar maps respectively. Attached Figure Description

[0021] Figure 1 Detailed visualization of significant CBF results clusters based on voxels in MRICron is shown. The two-sample t-test threshold used was P < 0.05 (FWE corrected), and the cluster size was 10 voxels. Warm colors indicate voxels with significantly reduced CBF in the SCD group compared to the HC group. (A) Cerebellar cluster 1 is shown in yellow / red, overlaid on the coronal SUIT atlas of the cerebellum. The names of the cerebellar lobes and cerebellar nuclei are marked in red on the right cerebellum. (B) Top row: magnified midbrain cluster 2, shown in yellow / red; Bottom row: midbrain cluster 2 covering the AAL3 brain region. Abbreviations: HC, healthy controls; SCD, spinocerebellar degeneration; CBF, cerebral blood flow; Roman numeral IX, cerebellar hemisphere lobule number; D, dentate nucleus; F, apical nucleus; Raphe_D, dorsal raphe nucleus; L, left; R, right; x, y, z: spatial coordinates.

[0022] Figure 2A planar visualization of Cluster 1, a significant CBF outcome in the cerebellum, is shown. (A) A t-plot of normalized CBF intergroup comparisons (HC>SCD) projected onto the cerebellar SUIT planar view. Warm colors represent areas of significant CBF decrease in SCD patients. (B) The outline curve of the outcome in planar view (A) is shown as a white dashed line overlaid on a planar view of the anatomical distribution of the cerebellar IX lobes. For VI-X lobes, the bilateral cerebellar hemispheres and vermis are shown in slightly different colors. (C) The outline curve of the outcome in planar view (A) is shown as a white dashed line overlaid on a sensorimotor orientation planar view of the cerebellum associated with activation of the hand, foot, and tongue. Abbreviations: HC, healthy controls; SCD, spinocerebellar degeneration; L, left; R, right; V, vermis; H, cerebellar hemisphere; Roman numeral IX, cerebellar lobule number.

[0023] Figure 3 The results show that there were statistically significant differences in region-based normalized CBF between the SCD and HC groups in both Cluster 1 and Cluster 2. (SCD, spinocerebellar degeneration. ** indicates P < 0.0001.)

[0024] Figure 4 The correlation between standardized CBF values ​​and clinical characteristics is shown. Standardized CBF in cluster 1 was negatively correlated with both the SARA total score (A) and the SDS standardized score (B). For cluster 2, standardized CBF was also negatively correlated with both the SARA total score (C) and the ICARS total score (D). Abbreviations: CBF, Cerebral Blood Flow; SARA, Ataxia Assessment and Rating Scale; SDS, Self-Rating Depression Scale; ICARS, International Collaborative Ataxia Rating Scale.

[0025] Figure 5 A flowchart of the method for processing and visualizing cerebral blood flow in spinocerebellar degeneration according to the present invention is shown. Detailed Implementation

[0026] like Figure 5 As shown, the data processing method for this pattern of altered cerebral blood flow (CBF) includes the following steps:

[0027] (1) 3D PCASL data were acquired using a 3.0T MR scanner equipped with an 8-channel head coil;

[0028] (2) After subtracting the labeled image from the ASL contrast image, the ASL difference image is calculated, and then the CBF image is generated by GE Functool infusion software;

[0029] (3) The CBF images of all subjects were nonlinearly co-registered to the PET perfusion template using a one-step registration method, and then spatially warped to the standard MNI space while resampling the voxel size.

[0030] (4) Based on hypothesis testing, a general linear model was constructed and a two-sample t-test was used to perform voxel-level statistical analysis on the standardized CBF images. Age and gender were used as covariates in the statistical model, and the whole-brain mask was used to limit the scope of CBF analysis to the whole-brain level.

[0031] (5) Extract the t-value map of each significant result block of the whole brain analysis in the xjView toolbox and save the corresponding block as a binary mask; for each subject, load the mask into the DPABI toolbox, extract and calculate the average standardized CBF value of the ROI of each significant result block;

[0032] (6) In MRICron, the t-value map of the significant mass located in the cerebellum is overlaid on the cerebellum SUIT template, and the cerebral blood flow involvement of the cerebellar lobules and nuclei is displayed in detail in the coronal view; in the SUIT toolbox of SPM, the t-value map of the significant mass of the cerebellum is mapped to the surface-based cerebellum SUIT template to obtain the t-value plane map of the cerebellar mass, and then the outline of the t-value plane map is overlaid on the cerebellar anatomical and functional orientation plane maps respectively.

[0033] This invention acquires three-dimensional pseudo-continuous arterial spin labeling (3D PCASL) data of spinocerebellar degeneration patients and healthy controls using a 3.0T MR scanner equipped with an 8-channel head coil. Brain blood flow (CBF) reconstruction is performed on the raw 3D PCASL data. Subsequently, all subjects' whole-brain CBF images are nonlinearly co-registered to a PET template and spatially normalized to the MNI standard space. The CBF value of each voxel in the whole brain is then divided by the whole-brain mean to normalize the data, achieving CBF image standardization. The standardized CBF images are then smoothed. A general linear model is constructed within SPM8 to perform voxel-level whole-brain statistical analysis of the CBF images to accurately obtain whole-brain CBF changes. Within xjView, t-value maps of significant clusters obtained after whole-brain statistical analysis are extracted. The process involves several steps: First, a block mask is used. Within DPABI, the CBF values ​​of the corresponding cerebellar masses are extracted using the mask. Then, within MRICron, the t-value maps of significant cerebellar masses are overlaid on the cerebellar SUIT template. In the coronal view, this allows for a more detailed visualization of CBF involvement in the cerebellar lobules and nuclei. Finally, within the SUIT toolbox, the t-value maps of significant cerebellar masses are mapped onto a surface-based cerebellar SUIT template to obtain a t-value planar map of the cerebellar masses. The outline of this t-value planar map is then overlaid on the cerebellar anatomical and functional orientation maps to further illustrate the anatomical distribution and functional impairment of the affected cerebellar lobules. This entire process requires no manual drawing and automatically performs voxel- and region-based quantitative analysis of cerebral blood flow, enabling visualization of the anatomical and functional details of the cerebellar regions.

[0034] Preferably, the method further includes step (7), performing statistical analysis of demographic and clinical data in SPSS software: assessing the normality of the data distribution by Shapiro-Wilk normality test, comparing the age difference between patients with spinocerebellar degeneration and healthy controls by a two-sample t-test, and using the Pearson chi-square χ2 test for gender differences, with the significance threshold set at P<0.05.

[0035] Preferably, in step (1), the acquisition parameters of the resting-state 3D PCASL are: layer thickness = 4mm, interlayer spacing = 4mm, repetition time TR = 4,817ms, echo time TE = 14.6ms, flip angle = 111°, post-marking delay time = 1,525ms, sampling = 1024, field of view FOV = 240mm × 240mm, voxel size = 1.875mm × 1.875mm, number of excitations NEX = 3, matrix = 128 × 128, and the total acquisition time is 6min 54s.

[0036] Preferably, in step (3), the CBF images of all subjects are first nonlinearly co-registered to the PET perfusion template in SPM8 using a one-step registration method, and then spatially warped to the Montreal Standard Space (MNI). At the same time, the voxel size is resampled to 2mm×2mm×2mm. After spatial standardization, the CBF value of each voxel in the whole brain is divided by the whole brain mean to normalize the data, and a full-width half-height Gaussian kernel of 8mm×8mm×8mm is applied for spatial smoothing to optimize the whole brain statistical analysis.

[0037] Preferably, in step (4), the FWE method is used for multiple comparison correction, and the voxel level threshold for correction is P<0.05 and the mass size is >10 voxels.

[0038] Preferably, in step (4), the third edition of the Automatic Anatomical Marker AAL3 Atlas is used in MRIcron software to anatomically mark the CBF significant result mass located in the midbrain, and the spatially unbiased infratentorial template SUIT toolbox is used in SPM to overlay the mass located in the cerebellum onto the SUIT template.

[0039] Preferably, in step (5), a two-sample t-test is used in SPSS to perform an ROI-based intergroup comparison of the mean standardized CBF for each significant outcome block in patients and controls; for the patient group, the mean standardized CBF value based on ROI is compared with clinical characteristics (SARA total score, ICARS total score, SAS standard score, SDS standard score, SRSS total score) using Pearson or Spearman correlation analysis; the significance threshold is set to P<0.05.

[0040] Preferably, in step (6), the t-value planar contour of the significant cerebellar mass is extracted in Photoshop software, and the contour is superimposed on the cerebellar anatomy and functional orientation planar maps respectively.

[0041] Those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium. When executed, the program includes the steps of the methods of the above embodiments. The storage medium can be ROM / RAM, magnetic disk, optical disk, memory card, etc. Therefore, corresponding to the method of the present invention, the present invention also includes a data processing device for changing the pattern of whole-brain cerebral blood flow (CBF). This device is typically represented in the form of functional modules corresponding to the steps of the method. The device includes:

[0042] The data acquisition module is configured to acquire 3D PCASL data using a 3.0T MR scanner equipped with an 8-channel head coil;

[0043] The CBF image generation module is configured to subtract the labeled image from the ASL contrast image, calculate the ASL difference image, and then generate the CBF image using GE Functool perfusion software; the spatial normalization processing module is configured to use a one-step registration method to nonlinearly co-register all subjects' CBF images to the PET perfusion template, then spatially warp to the standard MNI space, while resampling the voxel size.

[0044] The voxel-based whole-brain standardized CBF data analysis module is configured to construct a general linear model based on hypothesis testing and use a two-sample t-test to perform voxel-level statistical analysis on the standardized CBF images. Age and gender are used as covariates in the statistical model, and a whole-brain mask is used to limit the scope of CBF analysis to the whole-brain level.

[0045] The region analysis module is configured to extract the t-value map of each significant outcome clique from the above whole-brain analysis in the xjView toolbox and save the corresponding clique as a binary mask; for each subject, the mask is loaded into the DPABI toolbox to extract and calculate the average standardized CBF value of the ROI of each significant outcome clique.

[0046] The cerebellar salience mass visualization module is configured to, within MRICron, overlay the t-value map of a salience mass located in the cerebellum onto a cerebellar SUIT template, displaying detailed cerebral blood flow involvement in the cerebellar lobules and nuclei in a coronal view; within the SUIT toolbox of SPM, the t-value map of the cerebellar salience mass is mapped onto a surface-based cerebellar SUIT template to obtain a t-value planar map of the cerebellar mass, and then the outline of the t-value planar map is overlaid onto the cerebellar anatomical and functional orientation planar maps respectively.

[0047] The present invention is described in more detail below, specifically including:

[0048] 1. Subjects

[0049] Thirty patients with subarachnoid depression (SCD) (20 males and 10 females; mean age, 45.6 ± 12.8 years; age range, 25–67 years; median disease duration, 6 years) and 30 age- and sex-matched healthy controls (HC) (18 males and 12 females; mean age, 45.6 ± 14.9 years; age range, 24–68 years) were prospectively recruited at the China-Japan Friendship Hospital. Of the 30 SCD patients, 23 were diagnosed with subarachnoid carcinoma (SCA1, n = 1; SCA2, n = 2; and SCA3, n = 20), and the remaining 7 patients had MSA-C (n = 4) and ARCA (ANO10-ARCA, n = 1; SPG7-ARCA, n = 1; NPC1-ARCA, n = 1). The inclusion criteria for patients were: (1) diagnosis of SCD or a specific subtype by a trained neurologist based on neurological manifestations, genetic testing, family history, and routine brain MRI results; (2) age ≥ 18 years; and (3) right-handed. The exclusion criteria for patients were: (1) contraindications to MRI examination; (2) history of other neurological diseases such as head trauma, stroke, or tumors; and (3) artifacts in the MRI images. HCs were recruited from the local community based on the following inclusion criteria: (1) no history of neuropsychiatric disorders; (2) age ≥ 18 years; and (3) right-handed. The exclusion criteria for HCs were the same as those for SCD patients.

[0050] 2. Clinical features

[0051] On the day of the MRI scan, demographic information was collected and neurological assessments were performed prior to MR data acquisition. The Ataxia Assessment and Rating Scale (SARA) and the International Collaborative Ataxia Rating Scale (ICARS) were used to quantify neurological presentation and reflect disease severity in patients with SCD. In addition, all patients were assessed for depression, anxiety, and sleep status using the Self-Rating Anxiety Scale (SAS), Self-Rating Depression Scale (SDS), and Self-Rating Sleep Scale (SRSS). For the SAS and SDS, standardized scores, rather than total scores, were used to assess anxiety and depression. A patient was considered to be in a state of anxiety or depression only if their standardized SAS score was ≥50 or their standardized SDS score was ≥53.

[0052] 3. MRI Data Acquisition

[0053] All subjects underwent MRI scans on a 3.0T MR scanner equipped with an 8-channel head coil (General Electric, Discovery MR750, Milwaukee, WI, United States). The acquisition parameters for resting-state 3D PCASL were: slice thickness = 4 mm, slice spacing = 4 mm, repetition time (TR) = 4,817 ms, echo time (TE) = 14.6 ms, flip angle = 111°, post-marking delay = 1,525 ms, sampling = 1024, field of view (FOV) = 240 mm × 240 mm, voxel size = 1.875 mm × 1.875 mm, number of excitations (NEX) = 3, matrix = 128 × 128, and total acquisition time was 6 min 54 s. Earplugs and padding were provided to all subjects to reduce noise, and foam padding was provided to restrict head movement. During the ASL scan, all subjects were instructed to close their eyes, not to fall asleep, relax, and not think about anything particular.

[0054] 4. CBF Reconstruction and Post-processing

[0055] After subtracting the labeled image from the contrast image, the PCASL difference image is calculated. Then, the CBF map is generated from the ASL difference image using the GE Functool infusion software.

[0056] CBF image processing and further voxel-based analysis were performed using MATLAB's Statistical Parametric Mapping 8 (SPM8, https: / / www.fil.ion.ucl.ac.uk / spm / software / ). First, a one-step registration method was used to nonlinearly co-register all participants' CBF images to the PET perfusion template. Then, the images were spatially warped to the standard MNI space, and the voxel size was resampled to 2mm × 2mm × 2mm. After spatial standardization, the CBF value of each voxel in the whole brain was divided by the whole brain mean for data normalization. Spatial smoothing was then performed using an 8mm × 8mm × 8mm full width at half height (FWHM) Gaussian kernel to optimize whole-brain statistical analysis.

[0057] 5. Statistical Analysis

[0058] 5.1 Whole-brain analysis of CBF

[0059] In SPM8, a general linear model was constructed based on hypothesis testing, and a two-sample t-test was used to perform voxel-level statistical analysis on the standardized CBF images. Age and gender were included as covariates in the statistical model, and a whole-brain mask was used to limit the CBF analysis to the whole-brain level. Multiple comparison correction was performed using the FWE method, with the corrected voxel-level thresholds being P < 0.05 and mass size > 10 voxels.

[0060] The SUIT atlas was used in MRIcron (http: / / www.mccauslandcenter.sc.edu / crnl / tools) software to visualize the CBF-significant masses in the cerebellum in detail. The t-value maps of the significant masses in the cerebellum were overlaid on the cerebellar SUIT template, and the cerebral blood flow involvement of the cerebellar lobules and nuclei was displayed in detail in the coronal view. Figure 1 A); Anatomical marking of CBF-signature masses in the midbrain was performed using the third edition of the Automated Anatomical Marking Atlas (AAL3). Figure 1 B); Further, using the Spatially Unbiased Subtentorial Template (SUIT) toolkit (version 3.4, https: / / www.diedrichsenlab.org / imaging / suit_flatmap.htm) in SPM, the t-value maps of cerebellar salient outcome patches are mapped to a surface-based cerebellar SUIT template to obtain t-value planar maps of cerebellar patches. Figure 2 A); Extract the t-value plane map outline of the significant cerebellar mass in Photoshop software, and overlay the outline of the t-value plane map onto the cerebellar anatomical and functional orientation plane maps respectively. Figure 2 B and Figure 2 C) Above.

[0061] 5.2 Regional Analysis of CBF

[0062] The t-value maps of each significant outcome clique from the whole-brain analysis were extracted using the xjView toolbox (version 10.0, https: / / www.alivelearn.net / xjview), and the corresponding cliques were saved as binary masks. For each subject, the masks were loaded into the DPABI (version 5.1) toolbox to extract and calculate the mean standardized CBF value for each significant clique, which was further used for analysis based on the region of interest (ROI). Figure 3 ).

[0063] Two-sample t-tests were used for ROI-based intergroup comparisons. For the SCD group, Pearson or Spearman correlation analysis was performed between ROI-based CBF values ​​and clinical characteristics (including disease duration, SARA score, ICARS score, SDS standard score, SAS standard score, and SRSS score). Figure 4 A, Figure 4 B Figure 4 C and Figure 4 D). Statistical analysis was performed in SPSS software (version 26.0), with a significance threshold set at P < 0.05.

[0064] 5.3 Other Statistical Analysis

[0065] The remaining statistical analyses were performed using SPSS software (version 26.0). The normality of the data distribution was assessed using the Shapiro-Wilk normality test. Demographic comparisons between SCD patients and HC patients were performed using a two-sample t-test for age and a Pearson chi-square (χ²) test for sex. A significance threshold was set at P < 0.05.

[0066] We investigated changes in whole-brain standardized brain flow (CBF) in patients with spinocerebellar degeneration (SCD) using non-invasive ASL-MRI. We found decreased CBF in multiple cerebellar lobules, deep cerebellar nuclei, and the brainstem in SCD patients. This hypoperfusion in these areas was associated with disease severity and depression scores, potentially suggesting deficits in motor and emotional functions. MRI-based CBF values ​​may be a promising neuroimaging biomarker, reflecting the severity of in vivo neurodegeneration at the whole-brain level and indicating mood changes.

[0067] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Any simple modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for processing and visualizing cerebral blood flow in spinocerebellar degeneration using magnetic resonance imaging, characterized by: It includes the following steps: (1) Three-dimensional pseudo-continuous arterial spin labeling (3D-PCASL) data were acquired from spinocerebellar degeneration patients and healthy controls using a magnetic resonance scanner; (2) Subtract the labeled image from the contrast image of ASL to obtain the ASL difference image, and then generate the CBF image using GE Functool infusion software; (3) CBF image processing and further voxel-based analysis were performed in MATLAB's SPM 8; (4) Perform voxel-based whole-brain standardized CBF analysis: In SPM8, based on hypothesis testing, a general linear model is constructed and a two-sample t-test is used to perform voxel-level statistical analysis on the standardized CBF images. Age and gender are used as covariates in the statistical model, and a whole-brain mask is used to limit the scope of CBF analysis to the whole-brain level. (5) Perform CBF region analysis: In the xjView toolbox, extract the t-value map of each significant result block of the whole brain analysis above, and save the corresponding block as a binary mask; for each subject, load the mask into the DPABI toolbox, extract and calculate the average standardized CBF value of the ROI of each significant result block; (6) Visualization of cerebellar CBF changes: In MRICron, the t-value map of the significant mass in the cerebellum is overlaid on the cerebellar SUIT template to show the cerebral blood flow involvement of the cerebellar lobules and cerebellar nuclei in detail in the coronal view; In the SUIT toolbox of SPM, the t-value map of the significant mass in the cerebellum is mapped onto the surface-based cerebellar SUIT template to obtain the t-value plane map of the cerebellar mass, and then the outline of the t-value plane map is overlaid on the cerebellar anatomical and functional orientation plane maps to show the anatomical distribution and functional impairment of the affected cerebellar lobules in more detail.

2. The method for processing and visualizing cerebral blood flow in spinocerebellar degeneration according to claim 1, characterized in that: The method also includes step (7), performing statistical analysis of demographic and clinical data in SPSS software: the normality of the data distribution was assessed by the Shapiro-Wilk normality test, the age difference between patients with spinocerebellar degeneration and healthy controls was compared by a two-sample t-test, the sex difference was compared by the Pearson chi-square χ2 test, and the significance threshold was set as P<0.

05.

3. The method for processing and visualizing cerebral blood flow in spinocerebellar degeneration according to claim 2, characterized in that: In step (1), the acquisition parameters of 3D PCASL are as follows: layer thickness = 4mm, interlayer spacing = 4mm, repetition time TR = 4,817ms, echo time TE = 14.6ms, flip angle = 111°, post-marking delay time = 1,525ms, sampling = 1024, field of view FOV = 240mm × 240mm, voxel size = 1.875mm × 1.875mm, number of excitations NEX = 3, matrix = 128 × 128, and the total acquisition time is 6min 54s.

4. The method for processing and visualizing cerebral blood flow in spinocerebellar degeneration according to claim 3, characterized in that: In step (3), firstly, the CBF images of all subjects are nonlinearly co-registered to the PET perfusion template in SPM8 using a one-step registration method, and then spatially warped to the Montreal Standard Space MNI, while resampling the voxel size to 2mm×2mm×2mm. After spatial standardization, the CBF value of each voxel in the whole brain was divided by the whole brain mean to normalize the data, and spatial smoothing was performed using an 8mm×8mm×8mm full-width half-height Gaussian kernel to optimize the whole brain statistical analysis.

5. The method for processing and visualizing cerebral blood flow in spinocerebellar degeneration according to claim 4, characterized in that: In step (4), the FWE method is used for multiple comparison correction, and the voxel level threshold for correction is P<0.05 and the mass size is >10 voxels.

6. The method for processing and visualizing cerebral blood flow in spinocerebellar degeneration according to claim 5, characterized in that: In step (4), the third edition of the Automatic Anatomical Marker AAL3 Atlas is used in MRIcron software to anatomically mark the CBF significant result mass located in the midbrain, and the spatially unbiased infratentorial template SUIT toolbox is used in SPM to overlay the mass located in the cerebellum onto the SUIT template.

7. The method for processing and visualizing cerebral blood flow in spinocerebellar degeneration according to claim 6, characterized in that: In step (5), a two-sample t-test was used in SPSS to perform an ROI-based intergroup comparison of the mean standardized CBF for each significant outcome block in patients and controls; for the patient group, the mean standardized CBF value based on ROI was compared with clinical characteristics (SARA total score, ICARS total score, SAS standard score, SDS standard score, SRSS total score) using Pearson or Spearman correlation analysis; the significance threshold was set as P<0.

05.

8. The method for processing and visualizing cerebral blood flow in spinocerebellar degeneration according to claim 7, characterized in that: In step (6), the t-value planar outline of the significant cerebellar mass is extracted in Photoshop software, and the outline is superimposed on the cerebellar anatomy and functional orientation planar maps respectively.

9. A device for processing and visualizing magnetic resonance imaging of cerebral blood flow in spinocerebellar degeneration, characterized in that: It includes: The data acquisition module is configured to acquire 3D PCASL data using a 3.0T MR scanner equipped with an 8-channel head coil; The CBF image generation module is configured to subtract the labeled image from the ASL contrast image, calculate the ASL difference image, and then generate the CBF image using GE Functool perfusion software; the spatial normalization processing module is configured to use a one-step registration method to nonlinearly co-register all subjects' CBF images to the PET perfusion template, then spatially warp to the standard MNI space, while resampling the voxel size. The voxel-based whole-brain standardized CBF data analysis module is configured to construct a general linear model based on hypothesis testing and use a two-sample t-test to perform voxel-level statistical analysis on the standardized CBF images. Age and gender are used as covariates in the statistical model, and a whole-brain mask is used to limit the scope of CBF analysis to the whole-brain level. The region analysis module is configured to extract the t-value map of each significant outcome clique from the above whole-brain analysis in the xjView toolbox and save the corresponding clique as a binary mask; for each subject, the mask is loaded into the DPABI toolbox to extract and calculate the average standardized CBF value of the ROI of each significant outcome clique. The cerebellar salience mass visualization module is configured to, within MRICron, overlay the t-value map of a salience mass located in the cerebellum onto a cerebellar SUIT template, displaying detailed cerebral blood flow involvement in the cerebellar lobules and nuclei in a coronal view; within the SUIT toolbox of SPM, the t-value map of the cerebellar salience mass is mapped onto a surface-based cerebellar SUIT template to obtain a t-value planar map of the cerebellar mass, and then the outline of the t-value planar map is overlaid onto the cerebellar anatomical and functional orientation planar maps respectively.

Images