Method for performing emotion-related disease analysis on TED patient based on MRI image

Abstract

The invention provides a method for carrying out emotion-related disease analysis on a TED patient based on an MRI image, and the method comprises the following steps: S1, carrying out questionnaire evaluation on a tested patient, and evaluating the depression degree and anxiety level of the tested patient through neuropsychological evaluation; s2, performing whole-brain MRI scanning on the tested patient in the S1 by using a scanner to obtain MRI image data of the tested patient; s3, data preprocessing: preprocessing the MRI image data in the S2 to obtain preprocessed MRI data; s4, performing data post-processing including DTI and DKI processing on the preprocessed MRI data; s5, carrying out TBSS analysis on the data processed by DTI and DKI, identifying white matter beams by using a map, and extracting parameter values; and S6, data analysis is carried out, and grouping analysis, correlation analysis and intermediary analysis are carried out on the data in S1 to S5. The data obtained by the analysis method provided by the invention is more comprehensive and persuasive, and a new detection means is provided for TED patients.

Description

Technical Field

[0001] The present invention relates to the technical field of medical image processing, and particularly to a method for analyzing emotion-related diseases of TED patients based on MRI images. Background Art

[0002] Thyroid eye disease (TED) is an organ-specific autoimmune disease, most commonly associated with Graves' disease, presenting as a prominent extrathyroidal manifestation of autoimmune thyroid disease. TED is characterized by orbital inflammation and abnormal tissue hyperplasia, which can lead to symptoms such as periorbital edema, eyelid retraction, proptosis, diplopia, and vision loss. In addition to these orbital manifestations, TED also profoundly affects the patient's mood and mental health. Many studies have established a strong link between autoimmune thyroid disease and mood disorders, including anxiety and depression, with thyroid function thought to play a facilitating role. These symptoms are also frequently observed in TED, exacerbating the overall disease burden. Therefore, understanding the neural mechanisms of these mood disorders is crucial for improving patient prognosis.

[0003] Emerging evidence suggests that thyroid function can affect the central nervous system, potentially leading to changes that impact mood regulation. Recent MRI studies have revealed structural and functional alterations outside the orbit in TED patients, indicating that the central nervous system may be involved in the pathophysiology of TED-related mood disorders. In particular, white matter integrity is crucial for effective neural connectivity and is related to mood regulation. Disruption of white matter tracts may underlie the mood disorders observed in TED and may serve as a neural basis for linking thyroid dysfunction to psychological symptoms.

[0004] Diffusion MRI, including diffusion tensor imaging (DTI) and diffusion kurtosis imaging (DKI), can effectively characterize the neural structure and connectivity of the central nervous system. DTI provides key metrics such as fractional anisotropy (FA) to evaluate the overall directionality of water diffusion within brain tissue. DKI has higher sensitivity and captures the complexity or heterogeneity of the microstructures within brain tissue through mean kurtosis (MK). Collectively, these techniques provide complementary insights into the brain integrity of TED. Tract-based spatial statistics (TBSS) is a widely used tool for analyzing diffusion MRI data by projecting volumetric data onto a white matter skeleton, which helps reduce partial volume effects. It can enhance diffusion MRI analysis by detecting subtle white matter changes at the voxel level, providing high sensitivity, objectivity, and interpretability.

[0005] The prior art usually relies on the radiomics features of conventional MRI sequences to stratify the risk of diseases, while ignoring the exploration of these relationships using TBSS analysis of DTI and DKI metrics, and lacks a systematic study on the prognostic value and biological implications of high-throughput radiomics features derived from DTI and DKI metrics. Summary of the Invention

[0006] Aiming at the deficiencies in the prior art, the purpose of the present invention is to provide a method for analyzing mood-related diseases in TED patients based on MRI images.

[0007] The technical solution of the present invention is as follows: In the first aspect of the present invention, there is provided a method for analyzing mood-related diseases in TED patients based on MRI images, and the analysis method includes the following steps: S1. Conduct a questionnaire assessment on all subject patients, and evaluate the severity of depression and the level of anxiety of the subjects through neuropsychological assessment; S2. Use a scanner to perform whole-brain MRI scans on the subject patients in step S1 to obtain the MRI image data of the subject patients; S3. Perform data preprocessing, and preprocess the MRI image data obtained in step S2 to obtain preprocessed MRI data; S4. Perform data postprocessing on the preprocessed MRI data, including DTI processing and DKI processing; S5. Conduct TBSS analysis on the data after the DTI processing and the DKI processing, including FA map registration, skeleton generation, projection image, MK mapping registration, through statistical analysis, correction and inclusion of covariates, identify white matter tracts with an atlas and extract parameter values; S6. Conduct data analysis, and perform grouped analysis, correlation analysis and mediation analysis on the processed data in steps S1-S5.

[0008] Furthermore, in S1, when conducting a questionnaire assessment on all subject patients and evaluating the severity of depression and the level of anxiety of the subjects through neuropsychological assessment, specifically: The severity of depression is measured using the Hamilton Depression Rating Scale HDRS. The Hamilton Depression Rating Scale HDRS includes several different depression assessment sub-items that score the severity of depression separately, and after scoring the depression assessment sub-items separately, calculate the total depression score. The higher the total depression score, the higher the severity of depression; The severity level of anxiety is measured using the Hamilton Anxiety Rating Scale (HARS). The HARS includes several anxiety assessment sub-items that score the severity of anxiety separately. After scoring the anxiety assessment sub-items separately, the total anxiety score is calculated. The higher the total anxiety score, the higher the severity level of anxiety.

[0009] Further, in step S2, a scanner is used to perform a whole-brain MRI scan on the test patient in step S1 to obtain the MRI image data of the test patient. Specifically: The whole-brain MRI scan is performed using a scanner equipped with a phased array head coil with a preset number of channels, and auxiliary tools including foam padding and earplugs are used during the scan to reduce head movement and noise interference during the scan; During the whole-brain MRI scan, high-resolution sagittal T1-weighted structural scans and whole-brain diffusion imaging are performed in sequence. At the same time, diffusion shells with different b values are set during the whole-brain diffusion imaging, and data of corresponding numbers of diffusion-weighted directions are obtained, and b0 images with specific phase encodings are obtained for distortion correction.

[0010] Further, in step S3, the preprocessing process includes denoising, Gibbs ringing artifact removal, eddy current and motion correction, B0 field inhomogeneity correction, and brain extraction.

[0011] Further, in step S4, data post-processing including DTI processing and DKI processing is performed on the preprocessed MRI data. Specifically: DTI processing is performed. The diffusion tensor is fitted using the error of FSL to generate an FA map; DKI processing is performed. The diffusion kurtosis tensor is estimated using a diffusion kurtosis estimator, and an MK map is generated using a constrained linear least squares quadratic programming algorithm.

[0012] Further, in step S5, TBSS analysis is performed on the data after the DTI processing and the DKI processing, including FA map registration, skeleton generation, projection image, MK mapping registration. After statistical analysis, correction, and inclusion of covariates, white matter bundles are identified using an atlas and parameter values are extracted. Specifically: For the TBSS analysis, all FA maps obtained after the DTI processing are aligned with a preset template through non-linear registration; Based on the aligned FA maps, an average FA image is created and thinned to generate an average FA skeleton representing the centers of all bundles common to the group; The aligned FA image of each test patient is projected onto the average FA skeleton; Apply the transformation matrix and skeleton projection vector obtained from DTI-FA analysis to register the MK map obtained after the DKI processing into the same space as the FA skeleton; Perform statistical analysis of a preset number of permutations using the randomization tool of FSL, apply threshold-free cluster enhancement TFCE, and correct the results of multiple comparisons using family-wise error FWE correction; Incorporate age, gender, and years of education as covariates into the design matrix; Identify white matter bundles using the white matter label atlas available in FSL; For atlas-based analysis, extract the parameter values of a preset number of white matter bundles from the white matter label atlas.

[0013] Further, in step S6, perform data analysis, and conduct grouped analysis, correlation analysis, and mediation analysis on the processed data in steps S1 - S5. Specifically: Conduct grouped analysis, use GraphPad Prism 9 to analyze multi-parameter differences, evaluate normality with the Kolmogorov-Smirnov test, perform an independent two-sample t-test on Gaussian distribution variables, a Mann-Whitney U test on non-Gaussian distribution continuous variables, and a chi-square test on categorical variables, with p < 0.05 as the statistical significance criterion; Conduct correlation analysis, incorporate diffusion metrics and clinical characteristics into Pearson or Spearman correlation analysis to explore their relationships with age, gender, and years of education, as covariates of no interest; Conduct mediation analysis, perform mediation analysis in the PROCESS macro of SPSSAU according to the guidelines of Baron and Kenny, set the significance threshold to p < 0.05, use the bootstrapping method with 1000 resamples to estimate the indirect effect, and obtain a bias-corrected 95% confidence interval through this method to evaluate the statistical significance of the indirect effect. If the interval does not contain zero, the indirect effect is considered significant.

[0014] The second aspect of the present invention provides a system for analyzing mood-related diseases in TED patients based on MRI images. The system is used to execute the method for analyzing mood-related diseases in TED patients based on MRI images described above. The system includes: Questionnaire assessment dataset: Conduct questionnaire assessments on all subject patients, and evaluate the severity of depression and the level of anxiety of the subjects through neuropsychological assessments; MRI scanning device: Use a scanner to perform a whole-brain MRI scan on the test patient in step S1 to obtain the MRI image data of the test patient; Data preprocessing unit: Perform data preprocessing, and preprocess the MRI image data obtained in step S2 to obtain preprocessed MRI data; Data postprocessing unit: Perform data postprocessing on the preprocessed MRI data, including DTI processing and DKI processing; TBSS processing unit: Conduct TBSS analysis on the data after the DTI processing and the DKI processing, including FA map registration, skeleton generation, projection image, MK mapping registration. Through statistical analysis, correction, and inclusion of covariates, identify white matter bundles with an atlas and extract parameter values; Data analysis unit: Perform data analysis, and conduct grouped analysis, correlation analysis, and mediation analysis on the processed data in steps S1 - S5.

[0015] The third aspect of the present invention provides a storage medium for storing a computer program, where the computer program includes the above-mentioned system for analyzing emotion-related diseases in TED patients based on MRI images and / or when the computer program is executed by a processor, it realizes the above-mentioned method for analyzing emotion-related diseases in TED patients based on MRI images.

[0016] The fourth aspect of the present invention provides an electronic device, where the electronic device includes: A memory for storing a computer program, where the computer program includes the above-mentioned system for analyzing emotion-related diseases in TED patients based on MRI images; A processor for executing the computer program stored in the memory, and when the computer program is executed by the processor, it realizes the above-mentioned method for analyzing emotion-related diseases in TED patients based on MRI images.

[0017] Compared with the prior art, the present invention has the following beneficial effects: The method for analyzing mood-related diseases in TED patients based on MRI images provided by the present invention combines DTI and DKI parameters for the first time, and uses TBSS to analyze and compare the white matter integrity of TED patients and healthy controls (HC). Although the preliminary analysis of DTI-FA indicates that widespread white matter changes exist in TED patients, after multiple comparison corrections, these differences are not significant. In contrast, DKI-MK analysis shows that even after TFCE correction, multiple white matter tracts in TED patients are significantly reduced. This indicates that DKI may have higher sensitivity than DTI in detecting subtle white matter changes related to TED. In addition, the MK value within the left white matter tract in the brain is significantly correlated with lower thyroid hormone levels and higher depression scores. Importantly, mediation analysis shows that changes in white matter integrity completely mediate the relationship between thyroid function and depressive symptoms in TED patients. These findings emphasize the potential role of white matter changes as a neural pathway through which thyroid dysfunction may affect the mood state of TED patients. Therefore, the data obtained through the analysis method provided by the present invention is more comprehensive and persuasive, providing a new detection means for TED patients. This not only helps to more deeply understand the pathological mechanism of TED, but also provides an important basis for developing new treatment plans and strategies, with significant clinical application value.

[0018] Compared with the prior art, the present invention has the following disadvantages: 1) The sample size is relatively small, which may affect the universality of the research results, although it exceeds the minimum of 20 participants recommended by Simmons et al.

[0019] 2) Although the connection between white matter tract changes, mood disorders, and thyroid dysfunction in TED patients has been determined, no causal relationship has been established among these factors.

[0020] 3) Due to the initial research design, thyroid function indicators of the HC group were not collected, which limits the ability of this application to comprehensively evaluate the relationship between thyroid function and white matter changes.

[0021] 4) Fourth, this application cannot clearly determine whether the observed white matter changes are specifically related to TED or are the result of mood disorders independent of TED. Future research should include more detailed subgroup analyses, including TED patients with and without anxiety or depression, to further explore the relationship between white matter changes and mood symptoms in TED patients.

[0022] 5) Although the relevant analysis of this application indicates a potential connection between white matter changes and mood disorders in TED patients, these findings are not sufficient to withstand strict multiple comparison corrections. Brief Description of the Drawings

[0023] Other features, objectives, and advantages of the present invention will become more apparent by reading the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 A workflow diagram of the inclusion and exclusion criteria for the TED patient group and the HC group provided in the experimental examples of the present invention; Figure 2 A schematic diagram of the regional results showing differences in the TBSS analysis provided in the experimental examples of the present invention ( Figure 2 In, figure a): A schematic diagram of the uncorrected DTI-FA results; Figure 2 In, figure b): A schematic diagram of the DKI-MK results after TFCE correction (pFWE < 0.05, TFCE correction)); Figure 3 A schematic diagram of the correlation analysis results between the DKI-MK values of TED patients, thyroid function indicators, and questionnaire assessment results provided in the experimental examples of the present invention (the black line represents linear regression, and the confidence interval is 95% (shaded area)); Figure 4 A schematic diagram of the mediation analysis results provided in the experimental examples of the present invention; Figure 5 A schematic diagram of the relationship between white matter changes mediating the relationship between thyroid function and depression in thyroid eye disease provided in the experimental examples of the present invention. Detailed implementation manners

[0024] The present invention will be described in detail below in conjunction with specific embodiments.

[0025] The following embodiments will help those skilled in the art to further understand the present invention, but do not limit the present invention in any form. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several changes and improvements can still be made, and these all belong to the protection scope of the present invention.

[0026] The first embodiment This embodiment provides a method for analyzing mood-related diseases in TED patients based on MRI images, including the following steps: S1. Conduct a questionnaire assessment on all test patients, and evaluate the severity of depression and the level of anxiety of the subjects through neuropsychological assessment. The specific implementation operation is as follows: The severity of depression was measured using the Hamilton Depression Rating Scale (HDRS). The Hamilton Depression Rating Scale (HDRS) includes several different depression assessment sub-items that score the severity of depression separately. After scoring the depression assessment sub-items separately, the total depression score is calculated. The higher the total depression score, the higher the severity of depression. The 17-item Hamilton Depression Rating Scale (HDRS) was used for measurement. Depending on the symptoms, each item of the HDRS scores from 0 to 4 or 0 to 2, and the total score ranges from 0 to 52. The higher the score, the more severe the depressive symptoms. The cut-off points are generally divided into the following categories: mild (8 - 13), moderate (14 - 18), severe (19 - 22), and extremely severe (≥23) depression. The severity of anxiety was measured using the Hamilton Anxiety Rating Scale (HARS). The Hamilton Anxiety Rating Scale (HARS) includes several different anxiety assessment sub-items that score the severity of anxiety separately. After scoring the anxiety assessment sub-items separately, the total anxiety score is calculated. The higher the total anxiety score, the higher the severity of anxiety. The 14-item Hamilton Anxiety Rating Scale (HARS) was used to assess the anxiety level. Each item of the HARS scores from 0 to 4 points, and the total score ranges from 0 to 56 points. Similar to the HDRS, the higher the score, the higher the severity of anxiety. The cut-off points are generally divided into the following categories: a total score of ≥21 points indicates obvious anxiety, a score of ≥14 points indicates unobvious anxiety, a score of ≥7 indicates possible anxiety, and a score of <7 indicates no anxiety symptoms.

[0027] S2. Use a scanner to perform a whole-brain MRI scan on the test patient in step S1 to obtain the MRI image data of the test patient. The specific implementation operation is as follows: Whole-brain MRI scans were performed using a 3.0 Tesla scanner (Magnetom Vida, Siemens, Erlangen, Germany) equipped with a 64-channel phased array head coil. Foam padding and earplugs were used to minimize head movement and reduce scanning noise. During the whole-brain MRI scan, high-resolution sagittal T1-weighted structural scans and whole-brain diffusion imaging were performed in sequence. The parameters of the high-resolution sagittal T1-weighted structural scans were as follows: 3D MPRAGE sequence, isotropic spatial resolution of 0.8 mm, matrix of 320×320, FOV of 256×256 mm², TR / TE of 2400 / 2.38 ms. At the same time, diffusion shells with different b values were set during the whole-brain diffusion imaging, and data with corresponding numbers of diffusion-weighted directions were acquired, and b0 images with specific phase encodings were acquired for distortion correction. The b values were 1000 and 3000 s / mm², with 30 and 64 diffusion-weighted directions respectively, and 30 volumes using PA phase encoding. In addition, 3 b0 images with AP phase encoding were obtained for TOPUP distortion correction. The SMS factor of the EPI readout was 3, the Grappa acceleration factor was 2, the anisotropic resolution was 2.0 mm, 72 axial slices, matrix of 110×110, FOV of 220×220 mm², TR / TE of 7600 / 139 ms. The total scan time for each participant was 27 minutes and 38 seconds.

[0028] S3. Perform data preprocessing, and the specific implementation operations are as follows: Use the MRtrix3 software package to preprocess the MRI image data obtained in step S2 to obtain preprocessed MRI data; the preprocessing process includes denoising, Gibbs ringing artifact removal, eddy current and motion correction, B0 field inhomogeneity correction, and brain extraction.

[0029] S4. Perform data post-processing on the preprocessed MRI data, including DTI processing and DKI processing. The specific operations are as follows: For DTI processing, use the error of FSL to fit the diffusion tensor to generate an FA map. For DKI processing, use a diffusion kurtosis estimator (DKE) (http: / / www.nitrc.org / projects / dke) to estimate the diffusion kurtosis tensor, and use the constrained linear least squares quadratic programming (CLLS-QP) algorithm to generate an MK map.

[0030] S5. Perform TBSS analysis on the data after the DTI processing and the DKI processing, including FA map registration, skeleton generation, image projection, and MK mapping registration. After statistical analysis, correction, and inclusion of covariates, identify white matter bundles using an atlas and extract parameter values. The specific operations are as follows: Perform the TBSS analysis, and align all the FA maps obtained after the DTI processing with a preset template (FMRIB58_FA_1mm template) through non-linear registration. Based on the aligned FA maps, create an average FA image and thin it to generate an average FA skeleton representing all the bundle centers common to the group, with the threshold set to 0.2. Project the aligned FA image of each subject patient onto the average FA skeleton. Apply the transformation matrix and skeleton projection vector obtained from the DTI-FA analysis to register the MK mapping obtained after the DKI processing to the same space as the FA skeleton. Use the randomization tool in FSL to perform statistical analysis with 5000 permutations of the preset number, apply threshold-free cluster enhancement TFCE, and use family-wise error FWE correction to correct the results of multiple comparisons (p < 0.05). Include age, gender, and years of education as covariates in the design matrix. Use the Johns Hopkins University (JHU) ICBM-DTI-81 white matter label atlas available in FSL to identify white matter bundles. For atlas-based analysis, extract the parameter values of 48 white matter bundles from the JHU ICBM-DTI-81 atlas.

[0031] S6. Perform data analysis, and perform grouped analysis, correlation analysis, and mediation analysis on the data processed in steps S1 - S5. The specific operations are as follows: Perform grouped analysis, use GraphPad Prism 9 (GraphPad, CA, USA) to analyze the differences in multiple parameters, evaluate normality using the Kolmogorov-Smirnov test, perform an independent two-sample t-test on Gaussian distribution variables, perform a Mann-Whitney U test on non-Gaussian distribution continuous variables, and use the chi-square test to evaluate categorical variables. Statistical significance is defined as p < 0.05.

[0032] Perform correlation analysis, include diffusion metrics and clinical characteristics in Pearson or Spearman correlation analysis to explore their relationships with age, gender, and years of education as covariates.

[0033] Mediation analysis was performed in the PROCESS macro of SPSSAU (https: / / spssau.com) according to the guidelines of Baron and Kenny. The significance threshold was set at p < 0.05. The bootstrapping method with 1,000 resamplings was used to analyze the indirect effects. The bias-corrected 95% confidence interval (CI) was obtained by this method to assess the statistical significance of the indirect effect. If the interval did not contain zero, the indirect effect was considered significant.

[0034] Second embodiment This embodiment provides a system for analyzing emotion-related diseases of TED patients based on MRI images. The system in this embodiment is used to perform the method for analyzing emotion-related diseases of TED patients based on MRI images provided in the first embodiment. The system in this embodiment includes: Questionnaire assessment data set: All subjects were assessed with questionnaires, and the severity of depression and anxiety of the subjects were assessed through neuropsychological assessment; MRI scanning equipment: using a scanner to perform a whole-brain MRI scan on the subject in step S1 to obtain MRI image data of the subject; Data preprocessing unit: performing data preprocessing, preprocessing the MRI image data obtained in step S2 to obtain preprocessed MRI data; Data post-processing unit: performing data post-processing including DTI processing and DKI processing on the pre-processed MRI data; TBSS processing unit: performing TBSS analysis on the data after the DTI processing and the DKI processing, including FA map registration, skeleton generation, projection image, MK map registration, statistical analysis, correction and inclusion of covariates, using the atlas to identify white matter tracts and extract parameter values; Data analysis unit: performs data analysis, and performs grouping analysis, correlation analysis, and mediation analysis on the data processed in steps S1-S5.

[0035] Third embodiment This embodiment provides a storage medium, which is used to store a computer program. The computer program in this embodiment includes the system for analyzing emotion-related diseases in TED patients based on MRI images provided in the second embodiment and / or the computer program in this embodiment, when executed by a processor, is used to implement the method for analyzing emotion-related diseases in TED patients based on MRI images provided in the first embodiment.

[0036] Fourth embodiment This embodiment provides an electronic device, the electronic device comprising: A memory for storing a computer program, the computer program including the system for analyzing mood-related diseases of TED patients based on MRI images as described in the second embodiment; A processor for executing the computer program stored in the memory, and when the computer program is executed by the processor, it is used to implement the method for analyzing mood-related diseases of TED patients based on MRI images as described in the first embodiment.

[0037] Experimental example This experimental example was approved by the Ethics Committee of Shanghai Ninth People's Hospital and Shanghai Jiao Tong University School of Medicine (No.: SH9H-2022-T229-2).

[0038] All subjects provided written informed consent before participation.

[0039] According to the clinical guidelines proposed by the European Group on Graves' Orbitopathy (EUGOGO) and the Ophthalmic Plastic and Orbit Disease Subcommittee of the Chinese Medical Association Ophthalmology Society, a total of 24 TED patients and 27 well-matched healthy controls (HC) were enrolled from October 2022 to October 2023.

[0040] According to the conclusion proposed by the Thyroid Subcommittee of the Chinese Medical Association Endocrinology Society, only active TED patients were included in the study. There is evidence that active TED patients experience more severe anxiety and depressive symptoms. The enrollment situation of all participants is as Figure 1 shown. The exclusion criteria are as follows: 1) Signs and past medical history of various eye diseases or history of eye surgery; 2) Obvious abnormalities in the brain parenchyma; 3) History of systemic diseases or mental disorders or congenital and acquired diseases; 4) Drug dependence; 5) MRI contraindications or insufficient image quality.

[0041] For TED patients, several clinical and laboratory characteristics were evaluated, including best corrected visual acuity (BCVA), thyroid-stimulating hormone (TSH), free triiodothyronine (fT3), free thyroxine (fT4), and thyroid-stimulating hormone receptor antibody (TRAb). Normal range of thyroid function: TSH: 0.56 - 5.91 μIU / ml; fT3: 3.1 - 6.8 pmol / L; fT4: 12.0 - 22.0 pmol / L; TRAb: 0.00 - 1.75 IU / L. For the HC group, BCVA was also detected. The mean values of binocular data were calculated for analysis.

[0042] This experimental example refers to the method for analyzing mood-related diseases of TED patients based on MRI images provided in the first embodiment, and the following results and discussions are obtained.

[0043] I. Results 1.1 Demographic and clinical characteristics There were no significant differences in gender distribution, age, years of education, or BCVA between the TED group and the HC group. However, there were significant differences in the HARS score (p < 0.0001) and HDRS score (p = 0.012) (see Table 1), indicating that TED patients had mild anxiety compared with normal people and might be at risk of depression. Continuous variables were expressed as mean (± standard deviation) or median (interquartile range). Categorical variables were expressed as counts. TED: patient group with thyroid eye disease; HC: healthy control group; BCVA: best corrected visual acuity; TSH: thyroid stimulating hormone; fT3: free triiodothyronine; fT4: free thyroxine; TRAb: TSH receptor antibody; HARS: Hamilton Anxiety Scale; HDRS: Hamilton Depression Scale.

[0044] 1.2. Diffusion indices Before applying TFCE multiple comparison correction, compared with the HC group, the TED group showed extensive DTI-FA alterations ( Figure 2 a). However, there were no significant differences after correction. In contrast, after TFCE correction, the DKI-MK values of the TED patient group were significantly lower than those of the HC group. In the analysis based on TBSS and DKI-MK values, the brain regions with significant differences between the two groups, such as Figure 2 shown in b and summarized in Table 2 (pFWE < 0.05, TFCE correction). This indicates that DKI may have higher sensitivity than DTI in detecting subtle white matter changes related to TED.

[0045] 1.3. Correlation analysis between diffusion indices and clinical characteristics The thyroid function indices, questionnaire assessment results, and DKI-MK values of the TED group were included in the correlation analysis, with age, gender, and years of education as covariates. The results were as Figure 3 shown.

[0046] Among the thyroid function indices, the left ATR MK value was significantly positively correlated with fT3 (r = 0.413, p = 0.045) and fT4 (r = 0.539, p = 0.007); the left ALIC MK value was significantly positively correlated with fT4 (r = 0.503, p = 0.012); the left SCR MK value was significantly positively correlated with fT3 (r = 0.495, p = 0.014); In the questionnaire assessment, the MK values of the left SLF and HDRS (r = -0.455, p = 0.026), the MK values of the left SCR and HDRS (r = -0.452, p = 0.027), and the MK values of the left CT and HDRS (r = -0.468, p = 0.021) were all significantly negatively correlated. This indicates that the more severe the white matter damage in TED patients, the higher the depression level. The integrity of the left corona radiata and corpus callosum in patients with major depressive disorder is impaired, which further supports this conclusion. The changes in white matter tracts may disrupt the neural circuits involved in emotion regulation and cognitive function, leading to mood disorders in TED patients.

[0047] 1.4, Mediation analysis In the correlation analysis, a significant correlation was found between the DKI-MK value of the left SCR and the fT3 and HDRS scores. There is a potential link between the thyroid function and the depression level in TED patients. Therefore, mediation analysis was performed to determine whether the left SCR mediates the relationship between fT3 (independent variable) and HDRS score (dependent variable).

[0048] fT3 indirectly affects HDRS significantly through the left SCR, with an effect size of -0.803. The effect of fT3 on the left SCR is significant but small, with an effect size of 0.009. When fT3 is controlled, the effect of the left SCR on HDRS is significantly negative, with an effect size of -88.576. The direct effect of fT3 on HDRS is not significant. In addition, the total effect of fT3 on HDRS is also not significant (the results are as Figure 4 shown and summarized in Table 3). This indicates that lower fT3 and fT4 levels (the fT3 levels of the included TED patients tend to be at the lower end of the normal range, while the fT4 levels are mostly below the normal range) may be related to more obvious changes in the white matter microstructure of TED patients, which in turn is associated with more obvious self-reported depressive symptoms.

[0049] II. Discussion White matter changes mediate the relationship between thyroid function and depression in thyroid eye disease, as Figure 5As shown, thyroid dysfunction can trigger mood disorders in TED patients by impairing the integrity of white matter neural pathways (such as disrupting neural circuits involved in emotion regulation). However, the relationship among mood disorders, thyroid function, and brain changes (especially changes in white matter tracts) remains a limited area of research in TED. DTI and DKI are powerful tools for studying white matter microstructure. For the first time in this invention, DTI and DKI parameters are combined, and TBSS analysis is used to compare the white matter integrity between the TED patient group and the HC group. Although the preliminary analysis of DTI-FA indicates widespread white matter changes in TED patients, these differences are not significant after multiple comparison corrections. In contrast, DKI-MK analysis shows that even after TFCE correction, multiple white matter tracts in TED patients are significantly reduced. This suggests that DKI may have higher sensitivity than DTI in detecting subtle white matter changes related to TED. In addition, the MK value within the left white matter tract in the brain is significantly correlated with lower thyroid hormone levels and higher depression scores. Importantly, mediation analysis shows that changes in white matter integrity completely mediate the relationship between thyroid function and depressive symptoms in TED patients. These findings emphasize the potential role of white matter changes as a neural pathway through which thyroid dysfunction may affect the mood state of TED patients.

[0050] The increased sensitivity of DKI can be attributed to its ability to more accurately simulate the complexity of water diffusion in biological tissues. DTI assumes unrestricted Gaussian diffusion along a specific gradient direction, which makes its parameters dependent on the b-value and ignores subtle microstructural abnormalities. In contrast, DKI provides diffusion parameter estimates independent of the b-value by quantifying non-Gaussian diffusion, thus allowing for a more accurate approximation of the diffusion-weighted signal attenuation. The conclusion in 1.2 also reflects this advantage, as DKI detected significant white matter abnormalities that were not found after DTI correction. After TFCE correction, the DKI-MK values in TED patients showed that the significantly reduced white matter bundles were mainly located in the left hemisphere, including the left SLF, corona radiata bundles (ACR, SCR, PCR), thalamic radiation bundles (PTR, ATR), ALIC, CT, and the body of the corpus callosum. MK is a valuable indicator of tissue microstructural complexity, reflecting factors such as the density, orientation, and organization of cell membranes, axonal sheaths, and myelin layers. A decrease in MK usually indicates a decrease in microstructural complexity, suggesting the loss or degradation of these structural components, such as a decrease in cell density or impaired myelin and axonal integrity. The functions of the white matter bundles found in TED are components of various cognitive and emotional processes. The left SLF connects the frontal and parietal lobes and is involved in language, speech motor planning, and syntactic processing, playing a crucial role in emotion regulation and speech-related tasks. ACR, SCR, and PCR are involved in attentional mechanisms, including focusing, maintaining, stabilizing, and shifting attention. PTR is related to visual short-term memory ability, while ATR is related to cognitive processes such as processing speed and executive function. ALIC is a key target for surgical treatment of refractory mental illnesses such as bipolar disorder and generalized anxiety disorder. On the other hand, CT is responsible for voluntary motor control. Previous studies on the white matter of the brain in TED have mainly focused on the visual pathway. However, other neuroimaging studies have revealed alterations in brain regions related to working memory deficits, cognitive function, movement, and attention in TED patients. These extensive alterations suggest that TED has a more widespread impact on brain functions beyond the visual pathway, affecting regions related to emotion and cognition.

[0051] An important finding of the present invention is the lateralization of white matter changes in the left hemisphere. Lateralization refers to the specialization of certain functions or cognitive processes in one hemisphere of the brain over the other. It is well known that the two hemispheres of the brain process emotional information differently. The left hemisphere is more closely associated with positive emotions such as happiness and optimism, while the right hemisphere is more involved in processing negative emotions, including anxiety, sadness, and fear. The present invention found that the MK value of the white matter tracts in the left hemisphere decreased significantly, while after multiple comparison corrections, similar changes in the right hemisphere were not obvious, indicating that the significant decrease in the MK value within the left white matter tracts of TED patients may be related to the weakening of positive emotions. This conclusion is consistent with the results in 1.1: TED patients have relatively mild levels of anxiety and depression (median HARS score of 9.50 and median HDRS score of 6.00; among them, an HARS score ≥ 7 indicates possible anxiety, and an HDRS score around 8 indicates possible depression). In addition, correlation analysis further supports this conclusion (the significant decrease in the MK value within the left white matter tracts of TED patients may be related to the weakening of positive emotions), indicating that the more severe the white matter damage in TED patients, the higher the level of depression. The integrity of the left corona radiata and the CC body is impaired in patients with major depressive disorder, which further supports this conclusion. Changes in white matter tracts may disrupt neural circuits involved in emotion regulation and cognitive function, leading to mood disorders in TED patients.

[0052] It is worth noting that the relationship between thyroid function and emotional state remains controversial, which is crucial for understanding the multi-organ accumulation of autoimmune thyroid diseases such as TED. Some studies have shown that thyroid hormone levels can affect emotions and cause depressive symptoms. As reported by Maes et al., even within the standard range, patients with major depressive disorder usually show a decrease in basal TSH levels and an increase in fT4. In addition, higher serum fT3 levels are significantly correlated with clinical improvement in patients with depression. The test results in 1.4 of the present invention are consistent with these findings. The DKI-MK values of white matter tract changes in TED patients are positively correlated with fT3 and fT4 and negatively correlated with depressive scores. This indicates that lower fT3 and fT4 levels (the fT3 levels of the included TED patients tend to be at the lower end of the normal range, while the fT4 levels are mostly below the normal range) may be related to more obvious changes in the white matter microstructure of TED patients, which in turn is related to more obvious self-reported depressive symptoms.

[0053] Interestingly, the MK value of the left SCR was positively correlated with fT3 and negatively correlated with the HDRS score. The results of the mediation analysis further showed that the left SCR played a complete mediating role in the relationship between fT3 level and HDRS score. Specifically, although the direct effect of fT3 on the severity of depression was not significant, its indirect effect through altering white matter microstructure was statistically significant. This indicates that the alteration of white matter integrity, especially the change in the left SCR, might be the basis for the effect of fT3 on depressive symptoms in TED patients. These findings suggest that in TED, the effect of thyroid function on depressive symptoms might be mediated by alterations in brain microstructure, highlighting the potential role of the left SCR as a key neural substrate in this process. Additionally, this supports the view that the central nervous involvement in autoimmune thyroid diseases extends beyond peripheral thyroid dysfunction, contributing to the understanding of neuropsychiatric manifestations in these diseases).

[0054] III. Conclusion The present invention provides new research methods and insights into the complex relationship among mood disorders, thyroid function, and white matter tract alterations in TED patients. The research results of the present invention show that DKI exhibits higher sensitivity than DTI in detecting subtle changes in white matter integrity, especially in the left hemisphere. These changes are related to mood disorders and thyroid function in TED patients. In addition, white matter abnormalities might be an important mediator in the relationship between thyroid function and mood disorders in TED patients. Future studies with larger sample sizes, detailed subgroup analyses, and longitudinal designs are needed to further clarify the relationship among brain alterations, mood disorders, and thyroid function in TED patients.

[0055] The specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the above specific embodiments, and those skilled in the art can make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. Without conflict, the embodiments of the present application and the features in the embodiments can be combined with each other arbitrarily.

Claims

1. A method for analyzing emotion-related diseases in TED patients based on MRI images, characterized in that: The following steps are involved: S1. All subjects were evaluated by questionnaires and the severity of depression and anxiety of the subjects were assessed by neuropsychological assessment; S2, using a scanner to perform a whole-brain MRI scan on the subject in step S1 to obtain MRI image data of the subject; S3, performing data preprocessing, preprocessing the MRI image data obtained in step S2 to obtain preprocessed MRI data; S4, performing data post-processing including DTI processing and DKI processing on the pre-processed MRI data; S5, performing TBSS analysis on the data after the DTI processing and the DKI processing, including FA map registration, skeleton generation, projection image, MK map registration, statistical analysis, correction and inclusion of covariates, using the atlas to identify white matter tracts and extract parameter values; S6. Perform data analysis, and perform grouping analysis, correlation analysis, and mediation analysis on the data processed in steps S1-S5.

2. The method for analyzing emotion-related diseases in TED patients based on MRI images according to claim 1, characterized in that: In S1, all subjects were evaluated by questionnaires, and the severity of depression and anxiety of the subjects were evaluated by neuropsychological assessment, specifically: The severity of depression is measured using the Hamilton Depression Rating Scale (HDRS), which includes a number of depression assessment sub-items that score the severity of depression respectively, and a total depression score is calculated after scoring the depression assessment sub-items respectively, the higher the total depression score, the higher the severity of depression; The anxiety severity level is measured using the Hamilton Anxiety Rating Scale (HARS), which includes several anxiety assessment sub-items that score the anxiety severity separately. After scoring the anxiety assessment sub-items separately, the total anxiety score is calculated. The higher the total anxiety score, the higher the anxiety severity.

3. The method for analyzing emotion-related diseases in TED patients based on MRI images according to claim 1, characterized in that: In step S2, a whole-brain MRI scan is performed on the subject in step S1 using a scanner to obtain MRI image data of the subject, specifically: The whole-brain MRI scan is performed using a scanner equipped with a phased array head coil with a preset number of channels, and auxiliary tools including foam padding and earplugs are used during the scan to reduce head movement and noise interference during the scan; When performing the whole-brain MRI scan, high-resolution sagittal T1-weighted structural scan and whole-brain diffusion imaging are performed in sequence. At the same time, diffusion shells with different b values ​​are set during the whole-brain diffusion imaging, and data of a corresponding number of diffusion-weighted directions are acquired, as well as a b0 image encoded with a specific phase for distortion correction.

4. The method for analyzing emotion-related diseases in TED patients based on MRI images according to claim 1, characterized in that: In step S3 the preprocessing flow includes denoising, Gibbs ringing artifact removal, eddy current and motion correction, B0 field inhomogeneity correction and brain extraction.

5. According to the method for analyzing emotion-related diseases of TED patients based on MRI images according to claim 1, in step S4, the pre-processed MRI data is subjected to data post-processing including DTI processing and DKI processing, specifically: DTI processing was performed, and the diffusion tensor was fitted using the error of FSL to generate the FA map; DKI processing was performed, the diffusion kurtosis tensor was estimated using the diffusion kurtosis estimator, and the MK map was generated using the constrained linear least squares quadratic programming algorithm.

6. The method for analyzing emotion-related diseases in TED patients based on MRI images according to claim 1, characterized in that: In step S5, TBSS analysis is performed on the data after the DTI processing and the DKI processing, including FA map registration, skeleton generation, projection image, MK map registration, statistical analysis, correction and inclusion of covariates, and the atlas is used to identify white matter tracts and extract parameter values, specifically: To perform the TBSS analysis, all FA images obtained after the DTI processing are aligned with a preset template through nonlinear registration; creating an average FA image based on the aligned FA maps, and thinning the image to generate an average FA skeleton representing all bundle centers common to the group; projecting the aligned FA images of each of the test patients onto the average FA skeleton; Using the transformation matrix and skeleton projection vectors obtained from the DTI-FA analysis, the MK map obtained after the DKI processing is registered to the same space as the FA skeleton; The randomization tool of FSL was used to perform statistical analysis with a preset number of permutations, and the threshold-free clustering enhanced TFCE was applied, and the results of multiple comparisons were corrected using the family-wise error (FWE) correction; Age, gender, and years of education were included as covariates in the design matrix; White matter tracts were identified using the white matter labeling atlas available in FSL; To perform atlas-based analysis, parameter values ​​of a preset number of white matter tracts are extracted from the white matter label atlas.

7. The method for analyzing emotion-related diseases in TED patients based on MRI images according to claim 1, characterized in that: In step S6, data analysis is performed, and the processed data in steps S1-S5 are subjected to group analysis, correlation analysis and mediation analysis, specifically: Group analysis was performed, and multi-parameter differences were analyzed using GraphPad Prism 9. Normality was assessed by Kolmogorov-Smirnov test, independent two-sample t test was performed for Gaussian distribution variables, Mann-Whitney U test was performed for non-Gaussian distribution continuous variables, and chi-square test was performed for categorical variables. P < 0.05 was used as the statistical significance standard; Correlation analysis was performed, and diffusion indicators and clinical characteristics were included in Pearson or Spearman correlation analysis to explore their relationship with age, gender, and years of education as covariates of no interest; Mediation analysis was performed in the PROCESS macro of SPSSAU according to the guidelines of Baron and Kenny, with a significance threshold of p < 0.

05. The bootstrapping method with 1000 resamples was used to estimate the indirect effect, and the bias-corrected 95% confidence interval was obtained by this method to evaluate the statistical significance of the indirect effect. If the interval did not contain zero, the indirect effect was considered significant.

8. A system for analyzing emotion-related diseases of TED patients based on MRI images, for executing the method for analyzing emotion-related diseases of TED patients based on MRI images as described in any one of claims 1 to 7, characterized in that: The system comprises: Questionnaire assessment data set: All subjects were assessed with questionnaires, and the severity of depression and anxiety of the subjects were assessed through neuropsychological assessment; MRI scanning equipment: using a scanner to perform a whole-brain MRI scan on the subject in step S1 to obtain MRI image data of the subject; Data preprocessing unit: performing data preprocessing, preprocessing the MRI image data obtained in step S2 to obtain preprocessed MRI data; Data post-processing unit: performing data post-processing including DTI processing and DKI processing on the pre-processed MRI data; TBSS processing unit: performing TBSS analysis on the data after the DTI processing and the DKI processing, including FA map registration, skeleton generation, projection image, MK map registration, statistical analysis, correction and inclusion of covariates, using the atlas to identify white matter tracts and extract parameter values; Data analysis unit: performs data analysis, and performs grouping analysis, correlation analysis, and mediation analysis on the data processed in steps S1-S5.

9. A storage medium, characterized in that: The storage medium is used to store a computer program, which includes the system for analyzing emotion-related diseases in TED patients based on MRI images as described in claim 8 and / or the method for analyzing emotion-related diseases in TED patients based on MRI images as described in any one of claims 1-7 when the computer program is executed by a processor.

10. An electronic device, characterized in that: The electronic device comprises: A memory for storing a computer program, wherein the computer program includes the system for analyzing emotion-related diseases of TED patients based on MRI images according to claim 8; A processor, wherein the processor is used to execute the computer program stored in the memory, wherein the computer program, when executed by the processor, implements the method for analyzing emotion-related diseases in TED patients based on MRI images as described in any one of claims 1 to 7.

Images