Depression Diagnostic System Based on Serum Protein and Brain Functional Imaging Indicators

Abstract

This invention relates to the field of bioscience and technology, and in particular to a model, system, and application based on serum proteins and brain functional indicators. The brain functional indicators include resting-state low-frequency amplitude (ALFF) and local coherence (ReHo) values, and the serum protein concentrations include serum IL-4, IL-6, IL-10, BDNF, and cortisol concentrations. This invention constructs a model using linear discriminant analysis (LDA) that integrates indicators such as ALFF in the right caudate nucleus, ReHo in the left middle occipital gyrus and superior frontal gyrus, and serum IL-4, IL-6, IL-10, BDNF, and cortisol concentrations. This model can be used as a tool for the specific diagnosis of depression, providing a novel approach to diagnosing depression.

Description

Technical Field

[0001] This invention relates to the field of biological sciences, specifically to a diagnostic system for depression based on serum proteins and brain functional imaging indicators. Background Technology

[0002] Depression is a prevalent and disabling disorder worldwide, significantly impacting the mental and physical health of sufferers. It impairs work and learning abilities, severely affecting quality of life and potentially leading to suicide. A 2017 meta-analysis of the Global Burden of Disease study indicated that depression affects approximately 300 million people, making it a leading cause of disability and disease burden globally.

[0003] Accurate diagnosis of depression is crucial for effective treatment. However, the diagnosis of depression still relies primarily on clinical symptomatology. Furthermore, the diverse and nonspecific clinical symptoms of depression lead to a high rate of misdiagnosis, hindering effective treatment. This highlights the limitations of relying solely on symptomatology for diagnosis, emphasizing the importance of introducing objective diagnostic biomarkers. Unfortunately, despite decades of research and the discovery of numerous biomarkers related to the pathogenesis of MDD, truly effective clinical diagnostic biomarkers are still lacking. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention proposes a diagnostic system for depression based on serum proteins and brain functional imaging indicators.

[0005] The objective of this invention can be achieved through the following technical solutions:

[0006] A diagnostic system for depression, comprising:

[0007] The reagent detection module includes kits for detecting BDNF and cortisol protein concentrations, and kits for detecting IL-4, IL-6, and IL-10 concentrations;

[0008] The image detection module is used to acquire the patient's rs-fMRI data;

[0009] The data analysis module calculates the ALFF and ReHo values ​​based on the rs-fMRI data.

[0010] Optionally, the kit for detecting the concentrations of IL-4, IL-6, and IL-10 is a Th1 / Th2 / Th17 subgroup detection kit.

[0011] Optionally, the kit for detecting BDNF and cortisol protein concentration is an ELISA kit.

[0012] Optionally, the kit for detecting BDNF and cortisol protein concentrations includes a washing solution, biotin-labeled anti-human BDNF / cortisol, avidin, a buffer solution, a stop solution, and a chromogenic solution.

[0013] Optionally, the data analysis module is configured to construct machine learning models based on single-dimensional indicators of neuroimaging, single-dimensional indicators of neurophysiology, and a combination of both dimensions using linear discriminant analysis. The two indicators refer to the aforementioned "single-dimensional indicators of neuroimaging" and "single-dimensional indicators of neurophysiology".

[0014] A diagnostic model for depression is proposed, which constructs machine learning models based on single-dimensional indicators from neuroimaging, single-dimensional indicators from neurophysiology, and a combination of both dimensions using linear discriminant analysis.

[0015] Optionally, the neuroimaging dimension metrics include rs-fMRI data.

[0016] Optionally, the neurophysiological dimension indicators include BDNF, cortisol, IL-4, IL-6 and IL-10 concentrations.

[0017] The beneficial effects of this invention are as follows: It is the first discovery that combining the right caudate nucleus ALFF, left middle occipital gyrus and superior frontal gyrus ReHo, and serum BDNF, cortisol, IL-4, IL-6, and IL-10 concentrations can serve as a multifactorial platform for the specific diagnosis of depression, providing a completely new approach to diagnosing depression. Using this multifactorial platform, which includes the above indicators, as an indicator for diagnosing depression, the area under the ROC curve (AUC) is 0.99, with a sensitivity and specificity of 92% and 100%, respectively, and an accuracy as high as 96.3% (5-fold cross-validation). (Validation results) and superior to models using multiple neuroimaging indicators (right caudate nucleus ALFF, left middle occipital gyrus and superior frontal gyrus ReHo) (AUC=0.79, sensitivity=0.70, specificity=0.78, accuracy=74.4%) or models using multiple neurophysiological indicators (serum BDNF, cortisol, IL-4, IL-6 and IL-10 concentrations) (AUC=0.99, sensitivity=0.84, specificity=0.98, accuracy=91.5%), demonstrating very good diagnostic value for depression. Attached Figure Description

[0018] The invention will now be further described with reference to the accompanying drawings.

[0019] Figure 1 The brain regions showing differences in ALFF between the depression group and the healthy control group after GRF correction;

[0020] Figure 2The brain regions showing differences in ReHo between the depression group and the healthy control group after GRF correction are shown in the diagram.

[0021] Figure 3 The graph shows the comparison of serum IL-4 (A), IL-6 (B), IL-10 (C), BDNF (D), and cortisol (E) concentrations between the depression group and the healthy control group.

[0022] Figure 4 A is a graph showing the results of 5-fold cross-validation of the model that uses receiver operating characteristic curve analysis to distinguish between the depression group and the healthy control group by combining linear discriminant analysis with the ALFF value of the right caudate nucleus and the ReHo values ​​of the left middle occipital gyrus and superior frontal gyrus.

[0023] Figure 4 B is a graph showing the results of 5-fold cross-validation using receiver operating characteristic curve analysis to distinguish between the depression group and the healthy control group through a model combining linear discriminant analysis with serum IL-4, IL-6, IL-10, BDNF and cortisol concentrations;

[0024] Figure 4 C is a 5-fold cross-validation result graph showing the differentiation between the depression group and the healthy control group using receiver operating characteristic curve analysis, combined with a model of linear discriminant analysis using the ALFF value of the right caudate nucleus, the ReHo values ​​of the left middle occipital gyrus and superior frontal gyrus, and the concentrations of serum IL-4, IL-6, IL-10, BDNF, and cortisol. Detailed Implementation

[0025] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. Unless otherwise specified, the materials, processing software, etc. used in the following embodiments can be obtained commercially. Experimental methods not specified with specific conditions in the embodiments are generally performed under conventional conditions or according to the conditions recommended by the manufacturer.

[0026] General information, relevant scales, and resting-state functional magnetic resonance imaging (rs-fMRI) data of 63 patients with depression and 81 non-mental illness controls were collected. Serum samples were also collected from 37 patients with depression and 45 non-mental illness controls. Specifically, the diagnosis of depression was made and confirmed by experienced tertiary psychiatrists (chief / associate chief physician, attending physician, and senior resident physician) according to the diagnostic criteria in the Diagnostic and Statistical Manual of Mental Disorders (Fourth Edition) (DSM-IV). The patients with depression were recruited from Zhongda Hospital Affiliated to Southeast University, and the data of healthy controls were obtained from recruited members of the public.

[0027] The data processing, testing, and evaluation included: calculation of resting-state spontaneous brain activity indicators ALFF and ReHo; detection of serum IL-4, IL-6, IL-10, BDNF, and cortisol concentrations; and assessment of disease severity, including the Hamilton Depression Rating Scale-17 (HAMD-17) and the Hamilton Anxiety Rating Scale (HAMA).

[0028] The rs-fMRI data were processed and statistically analyzed using DPABI software to obtain the ALFF and ReHo values. Additionally, software such as Matlab, SPM, and SPSS were also required.

[0029] Before obtaining ALFF and ReHo, rs-fMRI data need to be preprocessed. The specific steps are as follows:

[0030] (1) Convert the format of the obtained image, that is, convert the T1 and BOLD original DICOM files into NIFTI files for subsequent processing;

[0031] (2) The first 10 time point images used for machine calibration were removed, and the remaining 230 time point images were used for further analysis;

[0032] (3) Use layer 31 as a reference layer to perform time correction on the images at the remaining 230 time points;

[0033] (4) Head movement correction: Subjects whose head movements were translated more than 2 mm or rotated more than 2° in any direction (x, y or z) were excluded from the analysis;

[0034] (5) Spatial standardization: T1-weighted anatomical images are segmented into white matter, gray matter and cerebrospinal fluid. Then, the transformation parameters estimated by the unified segmentation algorithm are used to register each subject's T1 image to the MNI space designed by the software. These transformation parameters are applied to the functional images, and the images are resampled with a voxel size of 3mm*3mm*3mm.

[0035] (6) Spatial smoothing: Gaussian smoothing is performed on the data with a half-height and full width of 6mm*6mm*6mm to reduce registration error and increase the normality of the data (this step is not performed in the preprocessing when acquiring ReHo);

[0036] (7) Delinear drift: Remove linear trends that have been generated and accumulated due to various reasons;

[0037] (8) Regression of covariate signals: The recommended Friston-24 model was used as the head motion regression model and was regressed together with white matter and cerebrospinal fluid signals to eliminate the influence of these signals;

[0038] (9) Filtering: Extract low-frequency amplitude signals in the frequency range of 0.01 to 0.08 Hz to reduce the influence of low-frequency and high-frequency oscillation signals.

[0039] The specific steps for performing ALFF and ReHo analysis using DPABI software are as follows:

[0040] (1) Calculate the low-frequency amplitude in the range of 0.01-0.08Hz:

[0041] First, the time signal is converted into a frequency domain power spectrum using a Fourier transform (FFT) algorithm. The average amplitude of the square root of the power spectrum in the range of 0.01-0.08 Hz is then calculated, which is the ALFF value. Next, the ALFF value of each voxel is normalized to a Z value (zALFF) to eliminate the differences in the overall ALFF level of the whole brain between individuals.

[0042] (2) Calculate local consistency (ReHo):

[0043] Kendall's coefficient concordance (KCC) was used to measure the consistency of a given voxel with its 26 nearest surrounding voxels in time series (Zang Y, Jiang T, Lu Y, et al. Regional homogeneity approach to fMRI data analysis[J]. Neuroimage, 2004, 22(1):394-400). The KCC of each voxel throughout the entire brain was calculated to construct the ReHo map for each subject. The ReHo map was then standardized by dividing the KCC of each voxel by the average ReHo of the entire brain. Finally, the data were spatially smoothed using a 6mm full-width isotropic Gaussian kernel.

[0044] (3) Extract ALFF and ReHo values:

[0045] The ROI signal extraction function of DPABI software was used to extract the ALFF and ReHo values ​​of brain regions that still showed differences between the depressed and healthy control groups after GRF correction, for subsequent necessary statistical analysis.

[0046] Serum BDNF and cortisol protein concentrations were detected using an ELISA kit, while serum IL-4, IL-6, and IL-10 were detected using a Th1 / Th2 / Th17 subset detection kit. Details of the kits used are as follows: BDNF and cortisol kit: R&D Systems, Mutlukent Mah, Arda Sk, USA; Th1 / Th2 / Th17 subset detection kit: Jiangxi Saiji Biotechnology Co., Ltd., which requires flow cytometry for detection.

[0047] The BDNF and cortisol kits provide the necessary, but not all, reagents required for detecting BDNF and cortisol protein concentrations: a 96-well microplate, two vials of human BDNF / cortisol protein standards, one 25ml vial of 20x wash buffer, two vials of biotin-labeled anti-human BDNF / cortisol, one 200µL vial of avidin, one 12ml vial of buffer, one 8ml vial of stop solution, one 30ml vial of chromogenic solution A, and one 15ml vial of stop solution. Additional reagents and materials required include: a microplate reader, pipettes, pipette tubes, graduated cylinders, absorbent paper, distilled or deionized water, and data analysis and plotting software.

[0048] Prepare all necessary items before the testing begins. The specific testing steps are as follows:

[0049] (1) Determine the number of antibody-coated ELISA plate wells required for this test. Each sample, standard and blank should be replicated.

[0050] (2) Sample addition: After serially diluting the BDNF / cortisol protein standard with diluent A, add 50 μL of each standard at the corresponding concentration to a row of wells pre-coated with BDNF / cortisol antibodies. Set up one blank control well (containing only diluent A, with no sample added, i.e., standard concentration of 0). Add 50 μL of sample (serum samples from 37 controls with depression and 45 controls without mental illness) to the remaining wells and mix gently. Incubate at 37°C for 45 minutes.

[0051] (3) Solution preparation: Dilute the 20-fold concentrated washing solution with distilled water 20 times for later use;

[0052] (4) Washing: Pour out the liquid in the wells of the microplate, shake dry, then fill each well with washing buffer, let stand for 30 seconds and then discard, repeat washing 4 times;

[0053] (5) Add biotin-labeled anti-human BDNF / cortisol: Add 50 μL of biotin-labeled anti-human BDNF / cortisol to all wells of the microplate and incubate at 37°C for 30 minutes.

[0054] (6) Cleaning: Repeat step 4;

[0055] (7) Add avidin: Add 50 μL of avidin to all wells of the microplate and mix gently. Incubate at 37°C for 15 minutes;

[0056] (8) Cleaning: Repeat step 4;

[0057] (9) Color development: Add 50 μL of color development solution A and 50 μL of color development solution B to all wells of the microplate and incubate at 37°C for 15 minutes.

[0058] (10) Stop the reaction: Add 50 μL of stop solution to all wells of the microplate to stop the reaction (the color immediately changes from blue to yellow);

[0059] (11) Analysis: With the blank well as zero, the optical density (OD value) was measured at 450 nm using an enzyme-linked immunosorbent assay (ELISA) instrument within 15 minutes after adding the stop solution. The BICC1 standard protein was serially diluted to a known concentration, and a standard curve was plotted after measuring the OD value. The BDNF / cortisol content in the sample was calculated based on the standard curve.

[0060] The Th1 / Th2 / Th17 subgroup detection kit requires, but is not exhaustive, the following reagents for detecting IL-4, IL-6, and IL-10 concentrations: 1 vial of 2.4 ml capture microsphere mixture (composed of polystyrene and 7 capture microspheres with different fluorescence intensities, the surfaces of which are coated with specific antibodies against human IL-2 / IL-4 / IL-6 / IL-10 / IL-17A / tumor necrosis factor α / interferon γ), 1 vial of 10 ng lyophilized recombinant human protein powder (quantitative standard), 1 vial of 2 ml PE-labeled detection antibody (fluorescent detection reagent), 1 vial of 0.75 ml polystyrene magnetic microspheres (calibration microspheres), 1 vial of 0.25 ml phycoerythrin fluorescein-labeled antibody control (calibration solution A), 1 vial of 0.25 ml fluorescein isothiocyanate-labeled antibody control (calibration solution B), 1 vial of 15 ml sample diluent, and 1 vial of 5 ml microsphere buffer. In addition, the following additional reagents and materials are required: flow cytometer, pipettes, pipette tubes, graduated cylinders, absorbent paper, PBS solution, distilled or deionized water, data analysis and plotting software, etc.

[0061] Prepare all necessary items before the testing begins. The specific testing steps are as follows:

[0062] 1. Preparation before the experiment

[0063] Items to bring: 1M PBS solution.

[0064] 2. Reagent preparation and sample addition

[0065] (1) Calculate the required number of experimental subjects n [n = number of samples + 10 standards + 1 negative control]. In this example, n = (37 + 45) + 10 + 1 = 93

[0066] (2) Open the quantitative standard, transfer the standard to a centrifuge tube, and label the tube as the highest concentration;

[0067] (3) Resuspend the standard in 2 mL of sample diluent and let it stand at room temperature for 15 minutes;

[0068] (4) Gently mix the standard with a pipette tip, avoiding violent shaking; take 9 experimental sample tubes and label them as 1:2, 1:4, 1:8, 1:16, 1:32, 1:64, 1:128, and 1:256 respectively, and add 300 μL of sample diluent to each tube;

[0069] (5) Take 300uL of liquid from the highest concentration standard tube into the 1:2 tube, mix by pipetting, take 300uL of liquid from the 1:2 tube into the 1:4 tube, mix by pipetting, and so on, until the 1:256 tube.

[0070] (6) Centrifuge the microsphere capture mixture at 200g for 5 minutes using a low-speed centrifuge, carefully aspirate the supernatant, add the same volume of microsphere buffer as the aspirated supernatant, vortex to mix thoroughly, and incubate in the dark for 30 minutes.

[0071] (7) Vortex mix the microsphere mixture and add 25 μL to each experimental tube;

[0072] (8) Add 25 μL of the serially diluted standard to the standard tube; as shown in the table below:

[0073]

[0074] (9) Add 25uL of the test sample to each sample tube;

[0075] (10) Add 25 μL of fluorescence detection reagent to all experimental tubes;

[0076] (11) After all the experimental tubes were thoroughly mixed by vortexing, they were incubated at room temperature in the dark for 2.5 hours.

[0077] (12) Add 1 mL of PBS solution to each experimental tube, centrifuge at 200 g for 5 minutes, and carefully aspirate the supernatant;

[0078] (13) Add 100 μL of PBS solution to each tube and let it stand until detection.

[0079] 3. Fluorescence detection

[0080] After each experimental tube is vortexed for 3-5 seconds, fluorescence detection is performed sequentially on a calibrated flow cytometer in the order of standard tube, negative control tube, and sample tube.

[0081] Modeling and testing using the classification learner toolkit in MATLAB 2021a (MathWorks, Natick, MA):

[0082] The ALFF or ReHo values ​​of brain regions that still showed differences between the depressed group and the healthy control group after GRF correction, as well as the serum BDNF, cortisol, IL-4, IL-6, and IL-10 concentrations, were used as input features. Linear discriminant analysis (LDA) was used to construct machine learning models based on single-dimensional indicators of neuroimaging, single-dimensional indicators of neurophysiology, and a combination of both dimensions. The ability of the models to distinguish between the depressed group and the healthy control group was tested, and the predictive ability and stability of the models were verified by 5-fold cross-validation.

[0083] The demographic and clinical characteristics of all subjects are shown in Table 1. The demographic and clinical characteristics of subjects with both imaging and serum protein data are shown in Table 2. The ROC curve analysis results after 5-fold cross-validation, used to differentiate between depression and non-psychiatric controls based on single imaging indicators, single serum protein concentrations, combined LDA models of single neuroimaging dimensions, combined LDA models of single neurophysiological dimensions, and combined LDA models of two dimensions, are shown in Table 3. Differences in ALFF and ReHo brain regions between depressed patients and non-psychiatric controls are shown in Table 3. Figure 1 and Figure 2 The serum concentrations of BDNF, cortisol, IL-4, IL-6, and IL-10 in controls with depression and non-mental illness are shown in the table below. Figure 3 The ROC curve analysis of the multi-dimensional, multi-indicator combined LDA model for diagnosing depression, after 5-fold cross-validation, is shown in [reference needed]. Figure 4 .

[0084] Table 1. Demographic and clinical characteristics of all participants.

[0085]

[0086] Note: a Independent samples t-test; b Chi-square test.

[0087] Table 2. Demographic and clinical characteristics of subjects with both imaging and serum protein data.

[0088]

[0089]

[0090] Note: a Independent samples t-test; b Chi-square test.

[0091] Table 3. ROC curve analysis results of LDA models based on multiple indicators across different dimensions and single indicators for diagnosing depression, after 5-fold cross-validation.

[0092]

[0093] Note: LDA 影像 The model is an LDA model that combines the ALFF value of the right caudate nucleus, the ReHo values ​​of the left middle occipital gyrus and superior frontal gyrus; LDA 蛋白 The model is an LDA model combining serum IL-4, IL-6, IL-10, BDNF, and cortisol; LDA 影像蛋白 The model is an LDA model that combines all the above-mentioned imaging and serum protein indicators. LDA, Linear Discriminant Analysis; AUC, Area Under the Curve; ALFF, Low Frequency Amplitude; ReHo, Local Consistency; IL-4, Interleukin-4; IL-6, Interleukin-6; IL-10, Interleukin-10; BDNF, Brain-Derived Neurotrophic Factor.

[0094] in Figure 1 As can be seen, after GRF correction, the right caudate nucleus ALFF was significantly elevated in the depression group compared with the non-mental disease control group.

[0095] Figure 2 The results showed that, after GRF correction, the ReHo values ​​of the left middle occipital gyrus and superior frontal gyrus were reduced in the depression group compared with the non-mental illness control group.

[0096] Figure 3 A indicates that serum IL-4 concentration was significantly lower compared to non-mental disease controls; Figure 3 B indicates that serum IL-6 concentration was significantly increased compared with non-mental disease controls; Figure 3 D indicates that serum BDNF concentration was significantly lower compared to non-mental disease controls; Figure 3 E indicates that serum cortisol concentration was significantly increased compared to non-mental disease controls.

[0097] Figure 4 A indicates that the LDA model based on imaging features (ALFF value of the right caudate nucleus, ReHo values ​​of the left middle occipital gyrus and superior frontal gyrus) showed a 5-fold cross-validation area under the ROC curve (AUC) of 0.79, a sensitivity of 70%, a specificity of 78%, and an accuracy of 74.4% in diagnosing depression. Figure 4B indicates that the LDA model based on multiple serum protein levels (IL-4, IL-6, IL-10, BDNF, and cortisol) for diagnosing depression, after 5-fold cross-validation, has an area under the ROC curve (AUC) of 0.99, a sensitivity of 84%, a specificity of 98%, and an accuracy of 91.5%. Figure 4 C indicates that the area under the ROC curve (AUC) of the multi-index LDA model based on combined imaging and protein analysis for diagnosing depression, after 5-fold cross-validation, is 0.99, with a sensitivity of 92% and a specificity of 100%, and an accuracy as high as 96.3%. This demonstrates excellent diagnostic value for depression and is significantly superior to LDA models combining single-dimensional indices. Table 3 shows the AUC, sensitivity, specificity, and accuracy for each case. The area under the ROC curve ranges from 1.0 to 0.5. When AUC > 0.5, the closer the AUC is to 1, the better the diagnostic effect. AUC between 0.5 and 0.7 indicates low accuracy, AUC between 0.7 and 0.9 indicates some accuracy, and AUC above 0.9 indicates high accuracy. An AUC of 0.5 indicates that the diagnostic method is completely ineffective and has no diagnostic value. An AUC < 0.5 is unrealistic and rarely occurs in practice. Furthermore, sensitivity and specificity are also important indicators for evaluating diagnostic biomarkers. Studies have shown that clinically useful biomarkers or tests for the accurate diagnosis and classification of diseases have a sensitivity and specificity of at least 80% (Schneider B, Prvulovic D. Novel biomarkers in major depression. Curr Opin Psychiatry. 2013. 26(1): 47-53.).

[0098] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0099] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention.

Claims

1. A diagnostic system for depression, characterized in that, include: The reagent detection module consists of kits for detecting BDNF and cortisol protein concentrations, and kits for detecting IL-4, IL-6, and IL-10 concentrations. The image detection module is used to acquire the patient's rs-fMRI data; The data analysis module calculates the ALFF and ReHo values ​​based on the rs-fMRI data. The data analysis module is configured to construct machine learning models based on a single neuroimaging dimension indicator, a single neurophysiological dimension indicator, and a combination of both dimensions using linear discriminant analysis. The neuroimaging dimension indicator is the ALFF and ReHo values ​​calculated from rs-fMRI data. The single neurophysiological dimension indicator is the concentration of BDNF, cortisol, IL-4, IL-6, and IL-10.

2. The depression diagnostic system according to claim 1, characterized in that, The kit for detecting the concentrations of IL-4, IL-6, and IL-10 is a Th1 / Th2 / Th17 subgroup detection kit.

3. The depression diagnostic system according to claim 1, characterized in that, The kit for detecting BDNF and cortisol protein concentrations is an ELISA kit.

4. The depression diagnostic system according to claim 1, characterized in that, The kit for detecting BDNF and cortisol protein concentrations includes washing solution, biotin-labeled anti-human BDNF / cortisol, avidin, buffer solution, stop solution, and chromogenic solution.

5. A diagnostic model for depression, comprising constructing machine learning models based on a single neuroimaging dimension indicator, a single neurophysiological dimension indicator, and a combination of both dimensions using linear discriminant analysis; wherein the neuroimaging dimension indicator is the ALFF and ReHo values ​​calculated from rs-fMRI data; and the single neurophysiological dimension indicator is the concentrations of BDNF, cortisol, IL-4, IL-6, and IL-10.

Images