Emotion improvement evaluation method and system based on resting-state functional magnetic resonance imaging

Abstract

The application belongs to the technical field of neural regulation and brain image analysis. An emotion improvement evaluation method and system based on resting-state functional magnetic resonance imaging are proposed. The emotion disorder scale after transcranial alternating current stimulation is calculated to obtain a first judgment result and a second judgment result. According to the resting-state functional image data, a third judgment result and a fourth judgment result are obtained. When the first judgment result and the third judgment result are both true, a stop treatment suggestion is generated. When the first judgment result and the fourth judgment result are both true, or the second judgment result and the third judgment result are both true, a consolidation treatment suggestion is generated. When the second judgment result and the fourth judgment result are both true, a continue treatment suggestion is generated. The application overcomes the deficiency of traditional simple dependence on the scale. The brain network index can reflect the improvement degree of the brain function state before and after treatment, support the efficacy evaluation from the physiological level, and improve the evaluation accuracy.

Description

Technical Field

[0001] This invention relates to the field of neural modulation and brain imaging analysis technology, specifically to a method and system for assessing mood improvement based on resting-state functional magnetic resonance imaging. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] Currently, mental disorders are among the most widespread mental illnesses globally. Existing treatments include psychological counseling, drug intervention, and physical therapy. The evaluation of their efficacy mainly relies on subject self-reports (such as subjective scales like the Beck Mood Disorders Scale (BDI)) and diagnoses by clinicians based on the Hamilton Mood Disorders Scale (HDRS). This evaluation model suffers from problems such as strong subjectivity, limited scale sensitivity, poor inter-rater consistency, and difficulty in timely reflecting changes in brain function.

[0004] In recent years, resting-state functional magnetic resonance imaging (rs-fMRI), as a non-invasive method for measuring brain function, can acquire spontaneous BOLD signals in the brain without external task intervention. The low-frequency oscillation characteristics of these signals are closely related to functional connectivity between brain regions. Brain network atlas studies based on rs-fMRI have shown that patients with mental disorders exhibit abnormal functional connectivity in their Default Mode Network (DMN) and Central Executive Network (ECN), such as oversynchronization within the DMN, weakened connectivity within the ECN, and connectivity imbalance between the DMN and ECN. These abnormalities are directly related to decreased introspective thinking, attentional control, and emotion regulation abilities.

[0005] Although some researchers have used extracting functional connectivity indicators to reflect the differences between the mood disorder group and the healthy control group, most of these are offline point-to-point comparative analyses. They lack a process that incorporates the dynamic changes of the brain network at multiple time points during the treatment process. In particular, when non-invasive neuromodulation technologies such as Hi-tACS are beginning to be applied to the intervention of mental disorders, there is still no mature method to quantify the changes in the brain network in real time after each stimulation cycle and compare them with clinical symptoms to guide the next treatment decision. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a method and system for assessing mood improvement based on resting-state functional magnetic resonance imaging (fMRI). By dynamically monitoring changes in brain network indicators using resting-state functional imaging data, it introduces objective neuromarkers for evaluating the efficacy of mood disorders, overcoming the limitations of traditional methods that rely solely on scales. Brain network indicators can reflect the degree of improvement in brain function before and after treatment, supporting efficacy assessment from a physiological perspective and improving assessment accuracy.

[0007] To achieve the above objectives, the present invention adopts the following technical solution:

[0008] In a first aspect, the present invention provides a method for assessing mood improvement based on resting-state functional magnetic resonance imaging.

[0009] A method for assessing mood improvement based on resting-state functional magnetic resonance imaging includes the following procedures:

[0010] The acquired resting-state functional magnetic resonance imaging data are preprocessed to obtain resting-state functional imaging data;

[0011] The mood disorder scale was calculated after transcranial strong alternating current stimulation. If the decrease in the score of the mood disorder scale was greater than or equal to the set decrease threshold, the first judgment result was obtained; otherwise, the second judgment result was obtained.

[0012] Based on resting-state functional imaging data, the first average connectivity strength of the brain default mode network node set and the second average connectivity strength of the brain central executive network node set are obtained; when the first average connectivity strength is less than or equal to the first set threshold and when the second average connectivity strength is greater than or equal to the second set threshold, a third judgment result is obtained; otherwise, a fourth judgment result is obtained.

[0013] When the first and third judgments are both true, a recommendation to stop treatment is generated; when the first and fourth judgments are both true, or when the second and third judgments are both true, a recommendation to consolidate treatment is generated; when the second and fourth judgments are both true, a recommendation to continue treatment is generated.

[0014] In one implementation of the first aspect of the present invention, preprocessing of the acquired resting-state functional magnetic resonance imaging data includes:

[0015] Time correction is performed based on the acquisition sequence of each resting-state functional magnetic resonance imaging (fMRI) image.

[0016] Register all time-corrected functional images to the same reference frame;

[0017] Align the registered functional images with the structural images of the same subject;

[0018] The aligned structural and functional images are simultaneously mapped to the standard brain template space;

[0019] The normalized functional image is smoothed by a Gaussian kernel to remove signal drift and high-frequency noise.

[0020] By removing confounding variables through a multiple linear regression model, resting-state functional image data were finally obtained.

[0021] In one implementation of the first aspect of the present invention, obtaining a first average connectivity strength of the brain's default mode network node set and a second average connectivity strength of the brain's central executive network node set based on resting-state functional imaging data includes:

[0022] Spatial division of resting-state functional imaging data was performed based on standard brain atlas templates. Spatial voxels corresponding to regions of interest in the standard brain atlas templates were extracted, and the mean blood oxygen level dependent signal time series of all voxels in each region of interest was calculated.

[0023] Based on the mean blood oxygenation level dependent signal time series of each region of interest, the correlation coefficient between any two regions of interest is calculated, and Fisher-Z transform is performed on all correlation coefficient values ​​to construct a standardized whole-brain functional connectivity matrix.

[0024] Based on a predefined list of nodes, the node sets belonging to the brain's default mode network and the node sets belonging to the brain's central executive network are identified respectively. Based on the standardized whole-brain functional connectivity matrix, the average Fisher-Z transform values ​​of all node pairs within the brain's default mode network and the brain's central executive network are calculated to obtain the first average connectivity strength and the second average connectivity strength.

[0025] As a further limitation of the first aspect of the present invention, the standardized whole-brain functional connectivity matrix is ​​as follows: , ,in, Represents the number of regions of interest. Representing the The region of interest and the first The connection relationships between regions of interest Representing the The region of interest and the first Correlation coefficients for regions of interest.

[0026] In one implementation of the first aspect of the present invention, the first average connectivity strength is: ,in, The total number of nodes representing the default mode network node set of the brain. Representing the Areas of interest Representing the Areas of interest Representing the The region of interest and the first The connection relationships between regions of interest This represents the set of network nodes representing the brain's default mode.

[0027] In one implementation of the first aspect of the present invention, the second average connectivity strength is: ,in, This represents the total number of nodes in the central executive network of the brain. Representing the Areas of interest Representing the Areas of interest Representing the The region of interest and the first The connection relationships between regions of interest It represents the set of nodes in the central executive network of the brain.

[0028] Secondly, the present invention provides a mood improvement assessment system based on resting-state functional magnetic resonance imaging.

[0029] A mood improvement assessment system based on resting-state functional magnetic resonance imaging, comprising:

[0030] The data preprocessing unit is configured to preprocess the acquired resting-state functional magnetic resonance imaging data to obtain resting-state functional image data.

[0031] The mood disorder assessment unit is configured to: calculate the mood disorder scale after transcranial strong alternating current stimulation; if the decrease in the score of the mood disorder scale is greater than or equal to the set decrease threshold, a first judgment result is obtained; otherwise, a second judgment result is obtained.

[0032] The functional imaging evaluation unit is configured to: obtain the first average connectivity strength of the brain default mode network node set and the second average connectivity strength of the brain central executive network node set based on resting state functional imaging data; obtain a third judgment result when the first average connectivity strength is less than or equal to a first set threshold and when the second average connectivity strength is greater than or equal to the second set threshold; otherwise, obtain a fourth judgment result.

[0033] The mood improvement assessment unit is configured to: generate a recommendation to stop treatment when both the first and third judgment results are true; generate a recommendation to consolidate treatment when both the first and fourth judgment results are true, or when both the second and third judgment results are true; and generate a recommendation to continue treatment when both the second and fourth judgment results are true.

[0034] Thirdly, the present invention provides a computer device, comprising: a processor and a computer-readable storage medium;

[0035] A processor, adapted to execute computer programs;

[0036] A computer-readable storage medium storing a computer program, which, when executed by a processor, implements the mood improvement assessment method based on resting-state functional magnetic resonance imaging, which is the first aspect of the present invention.

[0037] Fourthly, the present invention provides a computer-readable storage medium storing a computer program adapted to be loaded by a processor and to execute the mood improvement assessment method based on resting-state functional magnetic resonance imaging of the first aspect of the present invention.

[0038] Fifthly, the present invention provides a computer program product, which includes a computer program that, when executed by a processor, implements the mood improvement assessment method based on resting-state functional magnetic resonance imaging according to the first aspect of the present invention.

[0039] Compared with the prior art, the beneficial effects of the present invention are:

[0040] This invention utilizes resting-state functional imaging data to dynamically monitor changes in brain network indicators, introducing objective neuromarkers for evaluating the efficacy of treatment for mood disorders. This overcomes the shortcomings of traditional methods that rely solely on scales. Brain network indicators can reflect the degree of improvement in brain function before and after treatment, supporting efficacy assessment from a physiological perspective and improving assessment accuracy. The decrease in the first average connectivity of the default mode network node set reflects a reduction in patients' introspective and rumination symptoms; the increase in the second average connectivity of the central executive network node set reflects an improvement in control and executive functions. Compared to traditional scale scores, these changes provide objective neurobiomarkers. By combining these two indicators, this invention can clearly quantify the neurological effects of treatment.

[0041] This invention combines network indicators with clinical scores to form a dynamic decision-making mechanism, enabling individualized tracking of each patient. If significant improvement in network indicators is observed early in treatment, intervention can be terminated in advance; if indicators do not improve, treatment suggestions can be given promptly to avoid delays. This feedback mechanism improves the efficiency and safety of treatment. The solution of this invention is not only applicable to clinical monitoring but can also be used to study the response mechanism of mood disorders to brain stimulation. Quantified network change data helps to reveal the essence of brain dysfunction during the occurrence of mood disorders, providing a scientific basis for neuromodulation therapy. It realizes the quantification and objectification of efficacy evaluation for mood disorders, helps to improve the accuracy and efficiency of treatment monitoring, and has significant social and clinical value.

[0042] Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0043] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0044] Figure 1 This is a schematic diagram of the implementation path provided in Embodiment 1 of the present invention;

[0045] Figure 2 This is a schematic diagram of the image data preprocessing process provided in Embodiment 1 of the present invention;

[0046] Figure 3 This is a flowchart illustrating the brain network index calculation method provided in Embodiment 1 of the present invention;

[0047] Figure 4 This is a flowchart illustrating the result determination method provided in Embodiment 1 of the present invention;

[0048] Figure 5 This is a schematic diagram of the mood improvement assessment system based on resting-state functional magnetic resonance imaging provided in Embodiment 1 of the present invention;

[0049] Figure 6 This is a schematic diagram of a computer device provided in Embodiment 1 of the present invention. Detailed Implementation

[0050] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0051] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0052] Example 1:

[0053] Current mood improvement assessment strategies often focus on single assessment indicators, failing to integrate brain network indicators with scale indicators for fusion assessment. Therefore, a systematic and standardized mood improvement tracking program is urgently needed, capable of continuously acquiring brain network indicators after multiple Hi-tACS stimulations and combining them with HDRS scores to provide accurate efficacy assessments and individualized treatment recommendations for clinicians. In view of this, this invention proposes a mood improvement assessment method based on resting-state functional magnetic resonance imaging (rs-fMRI), combining the Hamilton Degree Scale and resting-state functional magnetic resonance imaging (rs-fMRI) brain network indicators to achieve dynamic monitoring and assessment of the process of treating mood disorders with Hi-tACS.

[0054] like Figure 1 As shown, patients need to complete at least two resting-state MRI scans within 14 days. After each scan, data preprocessing is required. Then, for each processed data, two sets of indicators are calculated in parallel: HDRS indicators (indicator A) and brain network indicators (indicator B). After the indicators for each scan are calculated, all indicators A and indicators B are summarized and compared. Finally, the results are judged, and treatment recommendations are output in categories.

[0055] More specifically, it includes the following processes:

[0056] S101: Preprocessing of MRI image data.

[0057] like Figure 2 As shown, the DICOM format data obtained from the original scans of the subjects were converted in batches to NIfTI format. In order to eliminate the influence of signal instability in the early stage of scanning, the first 10 frames of images were automatically discarded. Subsequently, the remaining time-series data were preprocessed as follows: (1) Time correction was performed based on the acquisition order of each image to correct the sampling time difference between different layers; (2) Head motion correction was performed to register all time-corrected functional images to the same reference frame, thereby eliminating the small displacement artifacts of the subjects; (3) Structural image registration was performed to accurately align the registered functional images with the high-resolution structural images of the same subjects; (4) Spatial alignment was performed. (5) Spatial smoothing: Gaussian smoothing kernel is applied to the normalized functional images to enhance the signal-to-noise ratio and reduce individual differences. (6) Linear trend removal and bandpass filtering are performed to remove signal drift and high-frequency noise. (7) Multiple confounding variables, including six-degree-of-freedom head motion parameter estimates, whole-brain average signal, cerebrospinal fluid, and white matter signals, are removed by a multiple linear regression model to suppress physiological noise and motion artifacts from interfering with functional signals to the greatest extent, thereby obtaining high-quality resting-state functional imaging data.

[0058] S102: Perform brain network index calculation.

[0059] like Figure 3 As shown, the preprocessed resting-state functional imaging data is spatially divided based on the standard AAL90 brain atlas template. Spatial voxels corresponding to 90 ROIs (Regions of Interest) in the template are extracted, and the average BOLD signal time series of all voxels within each ROI is calculated. Then, the Pearson correlation coefficient between any two ROIs in these 90 ROI time series is calculated, and Fisher-Z transform is applied to all correlation coefficient values ​​to construct a standardized whole-brain functional connectivity matrix. Next, based on a predefined node list, the core node sets belonging to the DMN (Default Mode Network) and the CEN (Central Executive Network) are identified. Then, the average Fisher-Z transform values ​​of all node pairs within these two sub-networks are calculated, i.e., the average functional connectivity strength within the DMN and the average functional connectivity strength within the CEN. Finally, the calculated average functional connectivity strength value within the DMN (i.e., the first average connectivity strength) is used to construct a standardized whole-brain functional connectivity matrix. ) and the CEN internal average functional connectivity strength value (i.e., the second average connectivity strength) The average functional connection strength value within the DMN is compared with a preset threshold. If the average functional connection strength value within the CEN is less than or equal to the threshold, the sub-network is determined to have improved functionality. If the average functional connection strength value within the CEN is greater than or equal to the threshold, the sub-network is determined to have improved functionality. Finally, based on the sub-network functional improvement determination results, if both the DMN and CEN are determined to have improved, the comprehensive evaluation index is recorded as 1B (i.e., the third judgment result). If only one of the two sub-networks is determined to have improved, the comprehensive evaluation index is recorded as 2B (i.e., the fourth judgment result).

[0060] The Default Mode Network (DMN) of this invention is a spontaneously active neural network in the brain during rest (without a specific task). Its core functions are related to self-awareness, recalling the past, envisioning the future, and social cognition. It is mainly composed of the medial prefrontal cortex, posterior cingulate cortex, precuneus, inferior parietal lobule, and bilateral temporal cortex. These areas exhibit high metabolic activity at rest, but their activity is inhibited when performing external tasks requiring focus. The Central Executive Network (ECN) of this invention generally refers to the frontoparietal control network that dynamically interacts with the DMN, and is responsible for working memory, attention allocation, and cognitive control. This network includes the dorsolateral prefrontal cortex and posterior parietal lobe, and is activated during complex tasks, forming an antagonistic pattern of "task-positive activation – resting-negative activation" with the DMN.

[0061] More specifically, it includes the following sub-processes:

[0062] (1) ROI definition and time series extraction.

[0063] Using the AAL90 template, the brain is divided into 90 ROIs (Regions of Interest). The preprocessed BOLD signals of all voxels within each ROI are averaged to obtain a representative time series for that region. :

[0064] (1-1);

[0065] in, Representing the ROI at time point The representative BOLD (Blood Oxygenation Level Dependent Signal). Represents the ROI number. =1,2,…,90; Represents the number of primes within the ROI; Representative voxels At the point of time The signal; Representative voxels Belongs to the One ROI; Represents the total number of time points (number of time points scanned / number of samples).

[0066] (2) Construct the functional connection matrix.

[0067] Calculate the first Sample mean of each ROI on the time axis :

[0068] (1-2);

[0069] Calculate the Pearson correlation coefficient between any two ROIs:

[0070] (1-3);

[0071] in, Representing the The ROI and the Pearson correlation coefficients of ROIs (range [...]) 1,1]), Representing the ROI at time point Representative BOLD, Representing the The sample mean of each ROI on the time axis.

[0072] Perform Fisher-z transform:

[0073] (1-4);

[0074] in, represent The Fisher-z transform results.

[0075] Obtain the weighted undirected functional connectivity matrix :

[0076] (1-5);

[0077] (3) Determine the DMN and CEN sub-network nodes, as shown in Table 1.

[0078] Table 1: DMN and CEN Subnetwork Nodes

[0079]

[0080] (4) Calculation (i.e., first average connectivity strength) and (i.e., the second average connectivity strength).

[0081] First average connectivity strength of the DMN node set :

[0082] (1-6);

[0083] in, A set of nodes representing the DMN; Represents the number of nodes in the DMN; This represents the summation of all non-repeating, unordered node pairs within the DMN subnetwork (i.e., each node pair is calculated only once).

[0084] The second average connectivity strength of the CEN node set:

[0085] (1-7);

[0086] in, The set of nodes representing CEN; This represents the number of nodes in CEN; This represents the summation of all non-repeating, unordered node pairs within the CEN subnetwork (i.e., each node pair is calculated only once).

[0087] (5) Make a judgment on the results, as shown in Table 2.

[0088] Table 2: Result Judgment

[0089]

[0090] If both indicators of the brain network index (indicator B) improve, it is recorded as 1B as complete improvement; otherwise, it is recorded as 2B as incomplete improvement.

[0091] S103: Make the final result determination.

[0092] like Figure 4 As shown, during the 14-day treatment cycle, patients' HDRS scores and resting-state brain network function indicators were collected simultaneously. Changes in HDRS scores were quantitatively assessed. If the score decreased by ≥50% from the baseline after treatment (i.e., the decrease in the score of the mood disorder scale was greater than or equal to the set decrease threshold, i.e., the set decrease threshold was 50%), it was defined as achieving a clinical response and recorded as 1A (i.e., the first judgment result); otherwise, it was recorded as 2A (i.e., the second judgment result). The degree of improvement in brain network function was assessed. If all preset subnetworks showed functional improvement, it was defined as complete improvement and recorded as 1B (i.e., the third judgment result); otherwise, it was recorded as 2B (i.e., the fourth judgment result). Treatment recommendations were automatically generated based on the combined status of the above two types of indicators: when the result was 1A1B, treatment was recommended to be stopped; when the result was 1A2B or 2A1B, consolidation treatment was recommended; and when the result was 2A2B, treatment was recommended to continue. This scheme uses a two-dimensional auxiliary assessment method to achieve dynamic monitoring and evaluation of the Hi-tACS treatment process for mood disorders, as shown in Table 3.

[0093] Table 3: Recommended Solutions

[0094]

[0095] In summary, this invention utilizes resting-state functional imaging data to dynamically monitor changes in brain network indicators, introducing objective neuromarkers for evaluating the efficacy of treatment for mood disorders. This overcomes the shortcomings of traditional methods that rely solely on scales. Brain network indicators can reflect the degree of improvement in brain function before and after treatment, supporting efficacy assessment from a physiological perspective, improving assessment accuracy, and achieving quantification and objectification of efficacy assessment for mood disorders. This contributes to improving the accuracy and efficiency of treatment monitoring and has significant social and clinical value.

[0096] The invention is further described below with reference to a specific implementation method: A patient with a mood disorder receives Hi-tACS treatment, with each cycle consisting of a 7-day stimulation intervention (example parameters: 15mA current, 30 minutes / session, twice / day). Before the start of treatment, a resting-state fMRI scan is performed to obtain baseline data, and the baseline HDRS score is recorded. Subsequently, resting-state fMRI scans are repeated over 14 days, with simultaneous assessment of HDRS. All fMRI scans are performed using 3T MRI, acquired with the subject at rest with eyes closed (e.g., TR=2000 ms, acquisition time 10 minutes).

[0097] All acquired fMRI data were preprocessed according to the above-described preprocessing procedure: the first 10 frames were removed, slice temporal correction and head motion correction were performed, the images were registered to the MNI standard space and smoothed by 8 mm, linear drift was removed and bandpass filtering was applied at 0.01–0.1 Hz; SPM12 / DPABI regression was used to remove noise interference components such as whole-brain average, cerebrospinal fluid and white matter signals, to obtain the purified BOLD time series. After processing, the brain was divided into several ROIs (using AAL templates), and the correlation coefficient matrix of each pair of ROIs was calculated.

[0098] A weighted brain network is constructed based on the functional connectivity matrix: nodes represent Regions of Interest (ROIs), edge weights are the calculated Pearson correlation coefficients between two ROIs, and Fisher-Z transform is applied to all correlation coefficient values ​​to extract indicators. , For example, brain network indicators after the second data collection, ≤ threshold and A score ≥ the threshold indicates complete improvement in brain network indicators, recorded as 1B. Correspondingly, the HDRS score decreased from 25 points before treatment to 12 points at the second measurement, a decrease of more than 50%, recorded as 1A. These indicators, along with the scale scores, collectively suggest a significant improvement in the patient's mood. After each assessment, a decision rule is used to determine whether to continue treatment. In this case, the patient met the 1A1B criteria after the second data collection, therefore, termination of treatment can be recommended. The above implementation method is only used to illustrate the principles and methods of the present invention; in practical applications, parameters and procedures can be appropriately adjusted as needed. Therefore, this invention, by combining Hi-tACS intervention with multi-timepoint fMRI brain network analysis, provides an innovative auxiliary assessment method with a clear implementation plan and feasibility.

[0099] Figure 5 A mood improvement assessment system based on resting-state functional magnetic resonance imaging is shown, comprising:

[0100] The data preprocessing unit 501 is configured to preprocess the acquired resting-state functional magnetic resonance imaging data to obtain resting-state functional image data.

[0101] The mood disorder assessment unit 502 is configured to: calculate the mood disorder scale after transcranial strong alternating current stimulation; if the decrease in the score of the mood disorder scale is greater than or equal to the set decrease threshold, a first judgment result is obtained; otherwise, a second judgment result is obtained.

[0102] The functional imaging evaluation unit 503 is configured to: obtain the first average connectivity strength of the brain default mode network node set and the second average connectivity strength of the brain central executive network node set based on resting state functional imaging data; obtain a third judgment result when the first average connectivity strength is less than or equal to a first set threshold and when the second average connectivity strength is greater than or equal to the second set threshold; otherwise, obtain a fourth judgment result.

[0103] The mood improvement assessment unit 504 is configured to: generate a suggestion to stop treatment when the first judgment result and the third judgment result are both true; generate a suggestion to consolidate treatment when the first judgment result and the fourth judgment result are both true, or when the second judgment result and the third judgment result are both true; and generate a suggestion to continue treatment when the second judgment result and the fourth judgment result are both true.

[0104] It is understood that the aforementioned units can be individually or entirely merged into one or more other units, or some of the units can be further divided into multiple functionally smaller units. This achieves the same operation without affecting the technical effects of the embodiments of the present invention. The aforementioned units are based on logical functional division. In practical applications, the function of one unit can be implemented by multiple units, or the function of multiple units can be implemented by one unit. In other embodiments of the present invention, the system may also include other units. In practical applications, these functions can also be implemented with the assistance of other units, and can be implemented collaboratively by multiple units.

[0105] According to another embodiment of the present invention, the system of this embodiment can be constructed by running a computer program (including program code) capable of performing the steps involved in the corresponding method of the present invention on a general-purpose computing device, such as a computer, which includes processing elements and storage elements such as a central processing unit (CPU), random access memory (RAM), and read-only memory (ROM). The computer program can be recorded on, for example, a computer-readable recording medium, loaded into the aforementioned computing device through the computer-readable recording medium, and run therein.

[0106] Figure 6A computer device is shown, which includes a processor 601, a communication interface 602, and a computer-readable storage medium 603. The processor 601, communication interface 602, and computer-readable storage medium 603 can be connected via a bus or other means.

[0107] The communication interface 602 is used to receive and send data. The computer-readable storage medium 603 can be stored in the memory of the electronic device. The computer-readable storage medium 603 is used to store computer programs, which include program instructions. The processor 601 is used to execute the program instructions stored in the computer-readable storage medium 603.

[0108] The processor 601 is the computing and control core of an electronic device. It is suitable for implementing one or more instructions, specifically for loading and executing one or more instructions to achieve the corresponding method flow or corresponding function.

[0109] Processor 601 is configured to perform the following procedure:

[0110] The acquired resting-state functional magnetic resonance imaging data are preprocessed to obtain resting-state functional imaging data;

[0111] The mood disorder scale was calculated after transcranial strong alternating current stimulation. If the decrease in the score of the mood disorder scale was greater than or equal to the set decrease threshold, the first judgment result was obtained; otherwise, the second judgment result was obtained.

[0112] Based on resting-state functional imaging data, the first average connectivity strength of the brain default mode network node set and the second average connectivity strength of the brain central executive network node set are obtained; when the first average connectivity strength is less than or equal to the first set threshold and when the second average connectivity strength is greater than or equal to the second set threshold, a third judgment result is obtained; otherwise, a fourth judgment result is obtained.

[0113] When the first and third judgments are both true, a recommendation to stop treatment is generated; when the first and fourth judgments are both true, or when the second and third judgments are both true, a recommendation to consolidate treatment is generated; when the second and fourth judgments are both true, a recommendation to continue treatment is generated.

[0114] This invention also provides a computer-readable storage medium, which is a memory device in an electronic device for storing programs and data. It is understood that the computer-readable storage medium here may include both built-in storage media in the electronic device and extended storage media supported by the electronic device. The computer-readable storage medium provides storage space for storing the processing system of the electronic device.

[0115] Furthermore, this storage space also contains one or more instructions suitable for loading and execution by the processor. These instructions can be one or more computer programs (including program code). It should be noted that the computer-readable storage medium here can be high-speed RAM memory or unstable memory, such as at least one disk storage device; optionally, it can also be at least one computer-readable storage medium located remotely from the aforementioned processor.

[0116] In one embodiment, the computer-readable storage medium stores one or more instructions; the processor loads and executes the one or more instructions stored in the computer-readable storage medium to perform the following process:

[0117] The acquired resting-state functional magnetic resonance imaging data are preprocessed to obtain resting-state functional imaging data;

[0118] The mood disorder scale was calculated after transcranial strong alternating current stimulation. If the decrease in the score of the mood disorder scale was greater than or equal to the set decrease threshold, the first judgment result was obtained; otherwise, the second judgment result was obtained.

[0119] Based on resting-state functional imaging data, the first average connectivity strength of the brain default mode network node set and the second average connectivity strength of the brain central executive network node set are obtained; when the first average connectivity strength is less than or equal to the first set threshold and when the second average connectivity strength is greater than or equal to the second set threshold, a third judgment result is obtained; otherwise, a fourth judgment result is obtained.

[0120] When the first and third judgments are both true, a recommendation to stop treatment is generated; when the first and fourth judgments are both true, or when the second and third judgments are both true, a recommendation to consolidate treatment is generated; when the second and fourth judgments are both true, a recommendation to continue treatment is generated.

[0121] The present invention also provides a computer program product or computer program comprising computer instructions stored in a computer-readable storage medium. A processor of an electronic device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the electronic device to perform the following process:

[0122] The acquired resting-state functional magnetic resonance imaging data are preprocessed to obtain resting-state functional imaging data;

[0123] The mood disorder scale was calculated after transcranial strong alternating current stimulation. If the decrease in the score of the mood disorder scale was greater than or equal to the set decrease threshold, the first judgment result was obtained; otherwise, the second judgment result was obtained.

[0124] Based on resting-state functional imaging data, the first average connectivity strength of the brain default mode network node set and the second average connectivity strength of the brain central executive network node set are obtained; when the first average connectivity strength is less than or equal to the first set threshold and when the second average connectivity strength is greater than or equal to the second set threshold, a third judgment result is obtained; otherwise, a fourth judgment result is obtained.

[0125] When the first and third judgments are both true, a recommendation to stop treatment is generated; when the first and fourth judgments are both true, or when the second and third judgments are both true, a recommendation to consolidate treatment is generated; when the second and fourth judgments are both true, a recommendation to continue treatment is generated.

[0126] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed in this invention can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can implement the described functions using different methods for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0127] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the flow or function according to the embodiments of the present invention is generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in or transmitted through a computer-readable storage medium. The computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic cable, digital cable) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data processing device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive), etc.

[0128] Example 2:

[0129] This embodiment proposes a graph-based method for tracking the mood improvement effect of transcranial strong alternating current stimulation (HAC). Patients need to complete at least two resting-state magnetic resonance imaging (fMRI) scans within 14 days. After each scan, data preprocessing is performed. Subsequently, two sets of indicators are calculated in parallel for each processed data: the HDRS indicator (indicator A) and the graph-based network topology indicator (indicator B). After the indicator calculations for each scan are completed, all indicators A and B are summarized and compared. Finally, the results are judged, and treatment suggestions are output in a categorized manner. This invention combines transcranial strong alternating current stimulation (Hi-tACS) with resting-state functional magnetic resonance imaging (rs-fMRI) and uses graph-based analysis to quantitatively evaluate the therapeutic effect of transcranial strong alternating current stimulation on mood improvement.

[0130] Specifically, it includes the following processes:

[0131] The acquired resting-state functional magnetic resonance imaging data are preprocessed to obtain resting-state functional imaging data;

[0132] The mood disorder scale was calculated after transcranial strong alternating current stimulation. When the decrease in the score of the mood disorder scale was greater than or equal to the set decrease threshold, the first judgment result (i.e., 1A) was obtained; otherwise, the second judgment result (i.e., 2A) was obtained.

[0133] Based on resting-state functional imaging data, the global efficiency and clustering coefficient are obtained; when the global efficiency is greater than or equal to the first set threshold and the clustering coefficient is less than or equal to the second set threshold, the third judgment result (i.e., 1B) is obtained; otherwise, the fourth judgment result (i.e., 2B) is obtained.

[0134] When the first and third judgments are both true, a recommendation to stop treatment is generated; when the first and fourth judgments are both true, or when the second and third judgments are both true, a recommendation to consolidate treatment is generated; when the second and fourth judgments are both true, a recommendation to continue treatment is generated.

[0135] In this embodiment, specifically, the DICOM data obtained from the original scans of the subjects are first converted in batches to NIfTI format, and the first 10 frames are automatically discarded to eliminate the influence of signal instability in the early stages of scanning; subsequently, the remaining time-series data are preprocessed as follows:

[0136] (1) Time correction based on the acquisition order of each slice is used to correct the sampling time difference between different layers; (2) Head motion correction is used to register all time-corrected functional images to the same reference frame to eliminate small displacement artifacts of the subject; (3) Structural image registration is used to accurately align the corrected functional images with the high-resolution structural images of the same subject; (4) Spatial normalization is used to map the aligned structural images and functional images to the standard brain template space to ensure spatial consistency across subjects; (5) Spatial smoothing is used to apply Gaussian smoothing kernel to the normalized functional images to enhance the signal-to-noise ratio and reduce individual differences; (6) Linear trend removal and bandpass filtering are used to remove signal drift and high-frequency noise; (7) Multiple confounding variables, including the estimated values ​​of six-degree-of-freedom head motion parameters, the average signal of the whole brain, and the signals of cerebrospinal fluid and white matter, are removed by multiple linear regression model to suppress the interference of physiological noise and motion artifacts on functional signals to the greatest extent, thereby obtaining high-quality resting-state functional imaging data.

[0137] In this implementation, the preprocessed resting-state functional imaging data is spatially partitioned based on the standard AAL90 brain atlas template. Spatial voxels corresponding to 90 ROIs (Regions of Interest) in the template are extracted, and the average BOLD (Blood Oxygenation Level Dependent signal) time series of all voxels within each ROI is calculated. Subsequently, the Pearson correlation coefficient between any two ROIs in these 90 ROI time series is calculated, and Fisher-Z transform is performed on all correlation coefficient values ​​to construct a standardized whole-brain functional connectivity matrix. Next, the top 15% of strong connections (sparseness S=0.15) are retained, and weak connections are removed through sparsification to reduce noise influence. Finally, the global efficiency is calculated. And the clustering coefficient C value; then, the calculated global efficiency The clustering coefficient C value is compared with a preset threshold. If the global efficiency is... If the value is greater than or equal to the threshold, the subnetwork is judged to have improved functionality; if the clustering coefficient C value is less than or equal to the threshold, the subnetwork is judged to have improved functionality; finally, based on the subnetwork functionality improvement judgment results, if the global efficiency... If both the clustering coefficient C and the clustering index C are determined to be improved, the comprehensive evaluation index is recorded as 1B (i.e., the third judgment result). If only one of the two indicators is determined to be improved, the comprehensive evaluation index is recorded as 2B (i.e., the fourth judgment result).

[0138] More specifically, it includes the following sub-processes:

[0139] (1) ROI definition and time series extraction.

[0140] Using the AAL90 template, the brain was divided into 90 regions of interest (ROIs). The preprocessed BOLD signals of all voxels within each ROI were averaged to obtain a representative time series for that region. :

[0141] (2-1);

[0142] in, Representing the The representative BOLD signal of a ROI at time point t; Represents the ROI number, i=1,2,…,90; Represents the number of primes within the ROI; Representative voxels At the point of time The signal; Representative voxels Belongs to the One ROI; Represents the total number of time points (number of time points scanned / number of samples).

[0143] (2) Construct the functional connection matrix.

[0144] Calculate the first Sample mean of each ROI on the time axis :

[0145] (2-2);

[0146] Calculate the Pearson correlation coefficient between any two ROIs:

[0147] (2-3);

[0148] in, Representing the With the Pearson correlation coefficients of ROIs (range [...]) 1,1]), Representing the The representative BOLD signal of a ROI at time point t. Representing the The sample mean of each ROI on the time axis.

[0149] Perform Fisher-z transform:

[0150] (2-4);

[0151] in, represent The Fisher-z transform results.

[0152] The weighted undirected functional connectivity matrix is ​​obtained as follows:

[0153] (2-5);

[0154] (3) Definition of whole brain network.

[0155] Treating the 90 AAL ROIs as network nodes, the edge weights are determined by... Given, form a weighted undirected graph. .

[0156] Calculate the adjacency matrix:

[0157] (2-6);

[0158] The top 15% of strong connections (sparseness S=0.15) are retained, and weak connections are removed through sparsification to reduce the impact of noise.

[0159] (4) Calculate graph theory network topology indices and .

[0160] In a weighted network, if the weights To represent "connection strength", the side length is often set as the reciprocal of the weight:

[0161] (2-7);

[0162] (2-8);

[0163] in, This represents the edge length between nodes.

[0164] Calculate global efficiency :

[0165] (2-9);

[0166] in, This represents the number of nodes (90 in this case). Representative node arrive The shortest path length; It represents the efficiency of network information integration; the higher the efficiency, the more integrated the information.

[0167] Calculate the average clustering coefficient :

[0168] compute nodes How many edges connect to other nodes:

[0169] (2-10);

[0170] Calculation by node The number of triangles with vertices (without repeating):

[0171] (2-11);

[0172] calculate:

[0173] (2-12);

[0174] Calculate the average:

[0175] (2-13);

[0176] in, This represents the elements of a binary adjacency matrix (1 for connected elements, 0 for disconnected elements). ); Representative node Degree (number of neighbors); Represented by nodes The number of triangles with vertices (without repeating any); Representative node The local clustering coefficient; The average clustering coefficient of the entire network. Representing the Each node.

[0177] (5) Result judgment, as shown in Table 1.

[0178] Table 4: Indicator Judgment Criteria

[0179]

[0180] According to Table 1, if the global efficiency If both the clustering coefficient C and the overall clustering coefficient C are deemed to be improved, the comprehensive evaluation index is recorded as 1B (i.e., the third judgment result). If only one of the two indicators is deemed to be improved, the comprehensive evaluation index is recorded as 2B (i.e., the fourth judgment result).

[0181] During the 14-day treatment cycle, patients' HDRS scores and resting-state brain network function indicators were collected simultaneously. Changes in HDRS scores were quantified: a decrease of ≥50% from baseline after treatment was defined as achieving a clinical response and recorded as 1A; otherwise, it was recorded as 2A. The degree of improvement in brain network function was assessed: complete improvement was defined as complete improvement and recorded as 1B; otherwise, it was recorded as 2B. Treatment recommendations were automatically generated based on the combined status of these two types of indicators: treatment was recommended to stop when the result was 1A1B; consolidation treatment was recommended when the result was 1A2B or 2A1B; and treatment was recommended to continue when the result was 2A2B. This invention, through a dual-dimensional auxiliary assessment method, achieved dynamic monitoring and evaluation of the Hi-tACS treatment process for depression, as shown in Table 2.

[0182] Table 5: Correspondence between Indicators and Recommended Solutions

[0183]

[0184] In summary, this invention, by integrating fMRI and graph theory analysis, achieves the quantification and objectification of the efficacy assessment of depression treatment, which helps to improve the accuracy and efficiency of treatment monitoring and has significant social and clinical value.

[0185] The following example illustrates this: A patient with depression undergoes Hi-tACS treatment, with each cycle consisting of a 7-day stimulation intervention (example parameters: 15mA current, 30 minutes / session, twice daily). Before treatment begins, a resting-state fMRI scan is performed to obtain baseline data, and the baseline HDRS score is recorded. Subsequently, resting-state fMRI scans are repeated over 14 days, with simultaneous HDRS assessment. All fMRI scans are performed using 3T MRI, acquired with the subject at rest with eyes closed (e.g., TR=2000ms, acquisition time 10 minutes).

[0186] All acquired fMRI data were preprocessed according to the above-described preprocessing procedure: the first 10 frames were removed, slice temporal correction and head motion correction were performed, the images were registered to the MNI standard space and smoothed by 8 mm, linear drift was removed and bandpass filtering was applied at 0.01–0.1 Hz; regression was used to remove noise interference components such as whole-brain average, cerebrospinal fluid, and white matter signals to obtain the purified BOLD time series. After processing, the brain was divided into several ROIs (using the AAL template), and the correlation coefficient matrix of each pair of ROIs was calculated.

[0187] A weighted brain network is constructed based on the functional connectivity matrix: nodes represent Regions of Interest (ROIs), edge weights are calculated based on the Pearson correlation coefficient between two ROIs, and Fisher-Z transform is applied to all correlation coefficients, retaining the top 15% of strong connections (sparseness S=0.15); sparsification removes weak connections to reduce noise impact. For example, graph theory network topology metrics are calculated after the second data collection. A score ≥ the threshold and C ≤ the threshold indicates complete improvement in brain network indicators, recorded as 1B. Correspondingly, the HDRS score decreased from 25 points before treatment to 12 points at the second measurement, a decrease of more than 50%, recorded as 1A. These indicators, along with the scale score, collectively suggest a significant improvement in the patient's mood. After each assessment, a decision-making rule is used to determine whether to continue treatment. In this example, the patient met the criteria for 1A1B after the second data collection, therefore, discontinuation of treatment can be recommended.

[0188] This embodiment also proposes a graph theory-based transcranial strong alternating current stimulation emotion improvement tracking system, including:

[0189] The data preprocessing unit is configured to preprocess the acquired resting-state functional magnetic resonance imaging data to obtain resting-state functional image data.

[0190] The mood disorder assessment unit is configured to: calculate the mood disorder scale after transcranial strong alternating current stimulation; if the decrease in the score of the mood disorder scale is greater than or equal to the set decrease threshold, a first judgment result is obtained; otherwise, a second judgment result is obtained.

[0191] The graph theory index evaluation unit is configured to: obtain global efficiency and clustering coefficient based on resting state functional image data; obtain a third judgment result when the global efficiency is greater than or equal to a first set threshold and the clustering coefficient is less than or equal to a second set threshold; otherwise, obtain a fourth judgment result.

[0192] The mood improvement tracking unit is configured to: generate a suggestion to stop treatment when the first and third judgment results are both true; generate a suggestion to consolidate treatment when the first and fourth judgment results are both true, or when the second and third judgment results are both true; and generate a suggestion to continue treatment when the second and fourth judgment results are both true.

[0193] This embodiment calculates the global efficiency. Graph theory metrics such as clustering coefficient C can clearly quantify the neurological effects of treatment. The improvement in C reflects the patient's improved attention allocation and cognitive control ability, while the decrease in C indicates the relief of excessive cohesion of local network modularization, which is closely related to the reduction of rumination introspection and emotional fixation. Compared with traditional scale scores, the dynamic changes of these graph theory network topology indicators provide more objective neurobiomarkers.

[0194] This embodiment combines graph theory network topology indicators with clinical scores to form a dynamic decision-making mechanism, enabling individualized tracking of each patient. If significant improvement in graph theory network topology indicators can be observed in the early stages of treatment, intervention can be terminated early; if the indicators do not improve, treatment suggestions can be given in a timely manner to avoid delays. This feedback mechanism improves the efficiency and safety of treatment.

[0195] This embodiment utilizes resting-state fMRI to dynamically monitor changes in graph network topology indicators, introducing objective neuromarkers for evaluating the efficacy of treatment for depression. This overcomes the shortcomings of traditional methods that rely solely on scales. Graph network topology indicators can reflect the degree of improvement in brain function before and after treatment, supporting efficacy assessment from a physiological perspective.

[0196] This invention is not only applicable to clinical monitoring, but also to the study of the response mechanism of depression to brain stimulation. The quantitative graph theory network topology index change data helps to reveal the essence of brain dysfunction in depression and provides a scientific basis for neuromodulation therapy. By integrating fMRI and graph theory analysis, the quantitative and objective assessment of the efficacy of depression treatment is realized, which helps to improve the accuracy and efficiency of treatment monitoring and has significant social and clinical value.

[0197] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An emotion improvement evaluation system based on resting state functional magnetic resonance imaging, characterized in that, The system comprises: a data preprocessing unit configured to preprocess the acquired resting-state functional magnetic resonance imaging data to obtain resting-state functional image data; an emotional disorder evaluation unit configured to calculate an emotional disorder scale after transcranial alternating current stimulation, and obtain a first judgment result when the score of the emotional disorder scale decreases by an amount greater than or equal to a set decrease threshold, or a second judgment result otherwise; a functional image evaluation unit configured to obtain a first average connectivity strength of a brain default mode network node set and a second average connectivity strength of a brain central executive network node set according to the resting-state functional image data; and obtain a third judgment result when the first average connectivity strength is less than or equal to a first set threshold and the second average connectivity strength is greater than or equal to a second set threshold, or a fourth judgment result otherwise. According to the resting-state functional image data, the first average connectivity strength of the brain default mode network node set and the second average connectivity strength of the brain central executive network node set are obtained, comprising: spatially dividing the resting-state functional image data based on a standard brain atlas template, extracting spatial voxels corresponding to regions of interest of the standard brain atlas template, and calculating the average blood oxygen level dependent signal time series of all voxels in each region of interest; based on the average blood oxygen level dependent signal time series of each region of interest, calculating the correlation coefficient between any two regions of interest, and performing Fisher-Z transformation on all correlation coefficient values to construct a standardized whole brain functional connection matrix; according to a predefined node list, respectively identifying a node set belonging to the brain default mode network and a node set belonging to the brain central executive network; and according to the standardized whole brain functional connection matrix, respectively calculating the average of the Fisher-Z transformed values of all node pairs in the brain default mode network and the brain central executive network to obtain the first average connectivity strength and the second average connectivity strength; The standardized whole-brain functional connectivity matrix is , ; in, Represents the number of regions of interest. Representing the The region of interest and the first The connection relationships between regions of interest Representing the The region of interest and the first Correlation coefficient of each region of interest; The first average connectivity strength is: wherein, a total number of nodes representing a set of nodes of a brain default mode network, a first region of interest, a second region of interest, a first region of interest, a second region of interest, a first region of interest, a second region of interest, and a connection relationship between the first region of interest and the second region of interest, a set of nodes of a brain default mode network; the second average connectivity strength is: wherein, a total number of nodes representing the set of nodes of the central executive network of the brain, a first region of interest, a second region of interest, a third region of interest, a fourth region of interest, a connection relationship between the first region of interest and the second region of interest, a connection relationship between the third region of interest and the fourth region of interest, a connection relationship between the first region of interest and the fourth region of interest, the set of nodes of the central executive network of the brain; an emotional improvement evaluation unit configured to generate a stop treatment recommendation when the first judgment result and the third judgment result are both true; generate a consolidation treatment recommendation when the first judgment result and the fourth judgment result are both true, or the second judgment result and the third judgment result are both true; and generate a continue treatment recommendation when the second judgment result and the fourth judgment result are both true.

2. The emotional improvement evaluation system based on resting-state functional magnetic resonance imaging according to claim 1, wherein preprocessing the acquired resting-state functional magnetic resonance imaging data comprises: time correction based on the acquisition order of each resting-state functional magnetic resonance imaging; registering all functional images after time correction to the same reference frame; aligning the registered functional images with the structural images of the same subject; mapping the aligned structural images and functional images to the standard brain template space at the same time; applying a Gaussian smoothing kernel to the normalized functional images and removing signal drift and high-frequency noise; removing confounding variables through a multiple linear regression model to finally obtain the resting-state functional image data. comprising:

3. A computer device, comprising: a processor and a computer readable storage medium; a processor adapted to execute a computer program; ​ A computer readable storage medium, wherein a computer program is stored in the computer readable storage medium, and the computer program, when executed by a processor, implements the emotion improvement evaluation system based on resting state functional magnetic resonance imaging according to any one of claims 1 to 2.

4. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, and the computer program is adapted to be loaded and executed by a processor to implement the emotion improvement evaluation system based on resting state functional magnetic resonance imaging according to any one of claims 1 to 2.

5. A computer program product, characterised in that, The computer program product comprises a computer program, and the computer program, when executed by a processor, implements the emotion improvement evaluation system based on resting state functional magnetic resonance imaging according to any one of claims 1 to 2.

Images