A real-time enhanced personalized regulation system for the treatment of postpartum depression

Abstract

The present invention relates to the technical field of postpartum depression, and discloses a real-time enhanced individualized regulation system for the treatment of postpartum depression, including a data collection module, an fMRI processing and analysis module, and an rTMS regulation module. The fMRI processing and analysis module is used to pre-process the resting fMRI image data, and then construct an individualized partition map and determine the TMS stimulation target after finding the PPD-driven brain area. The rTMS regulation module includes a cognitive reappraisal video training unit and an rTMS enhanced regulation unit. The real-time enhanced individualized regulation system for the treatment of postpartum depression combines cognitive reappraisal and rTMS technologies for the first time, utilizes FCD-STD correlation calculation and individualized partition map construction, and determines the TMS stimulation target considering the individual differences of each patient, which has an accurate emotional relief effect on postpartum depression. At the same time, it combines cognitive reappraisal video training to improve patients' cognitive disorders such as attention, working memory, and negative thinking patterns, and has a long-term and stable emotional relief effect.

Description

Technical Field

[0001] The present invention relates to the technical field of postpartum depression, and in particular to a real-time enhanced individualized regulation system for treating postpartum depression. Background Art

[0002] Maternal mental health problems in the early postpartum period are a global public health issue that seriously affects the physical and mental health of women and infants. Postpartum depression (PPD) is one of the most common mental health disorders.

[0003] Emotion regulation is one of the important aspects of psychological regulation in PPD, which refers to the process of assisting individuals in evaluating, managing, and modifying emotional experiences and expressions through various methods. The most commonly used emotion regulation strategy is cognitive reappraisal, which is a method of cognitive reconstruction. It can correct the potential emotional meaning of the situation that causes the emotion by reinterpreting the meaning or self-relevance of the situation before the emotion fully develops. Cognitive reappraisal enables people to look at things from a more positive and reasonable perspective, thereby reducing negative emotions or enhancing positive emotions. It has been widely used in people facing various challenges and pressures, especially for improving the emotions of patients with depression who often experience negative sadness, self-denial and other emotions;

[0004] At present, the response rate of rTMS technology applied to depression is generally 40%-50%, and the remission rate is generally 20%-30%. At the same time, rTMS also has certain limitations in improving cognitive problems in depression. It has a better effect on improving specific cognitive areas (such as working memory and attention), but for other cognitive problems in patients with depression, such as decision-making difficulties and reduced self-worth, the improvement effect is not obvious and the effect is relatively short-lived. Summary of the invention

[0005] 1. Technical issues to be solved

[0006] In view of the deficiencies of the prior art, the present invention provides a real-time enhanced individualized regulation system for the treatment of postpartum depression, which has the advantages of providing accurate emotional relief for patients with postpartum depression, and solves the above-mentioned technical problems.

[0007] (II) Technical solution

[0008] To achieve the above-mentioned object, the present invention provides the following technical solutions: a real-time enhanced individualized regulation system for the treatment of postpartum depression, comprising a data collection module, an fMRI processing and analysis module, and an rTMS regulation module;

[0009] The data collection module is used to collect resting state fMRI image data (rs-fMRI) and T1 structural image data (sMRI);

[0010] The fMRI processing and analysis module is used to pre-process the resting-state fMRI image data, and then construct an individualized partition map and determine the TMS stimulation target after finding the PPD-driven brain area;

[0011] The specific steps of the fMRI processing and analysis module are as follows:

[0012] A1. Preprocessing: Preprocess the resting-state fMRI image data, including removing the first N time points of the resting-state fMRI image data, and registering all time points to the first time point. Then, register the resting-state fMRI image data to the standard template space, smooth the data, and then perform a 0.01-0.1 Hz filter to regress the signals of motion, white matter, and cerebrospinal fluid.

[0013] A2. Find the PPD driving brain region: calculate the FCD-STD correlation, and select the region with the highest FCD-STD correlation as the driving brain region;

[0014] A3. Construct an individualized partition map and determine the TMS stimulation target: Calculate the similarity of the whole-brain functional connection pattern between any two voxels, obtain the similarity matrix between each voxel in a specific area, and complete the individualized brain area division. Select the sub-area with the strongest functional connection between each sub-area and the sub-prefrontal cortex area as the TMS stimulation target, and finally register the target back to the T1 structural image to guide the rTMS regulation module.

[0015] The rTMS regulation module includes a cognitive reappraisal video training unit and an rTMS enhancement regulation unit;

[0016] The cognitive reappraisal video training unit stores a family relationship picture library and a cognitive reappraisal video library, wherein the family relationship picture library stores pictures of family relationships, wherein family relationships include husband-wife relationships, mother-in-law-and-daughter-in-law relationships, and parent-child relationships, and the videos stored in the cognitive reappraisal video library correspond one-to-one to the pictures in the family relationship picture library;

[0017] The rTMS enhancement control unit is used to stimulate the previously determined TMS stimulation target point with a high-frequency pulse of 10 Hz when the cognitive reappraisal video training unit is called, and the start and end times correspond to the cognitive reappraisal training video.

[0018] As a preferred technical solution of the present invention, the specific process of step A2 is as follows:

[0019] A2.1. Segment the preprocessed resting-state fMRI image data to generate an M×t time series matrix, where M represents the number of regions of interest (ROI) and t represents the scanning time of the subject; select a window with a length of w=30 time points to generate an M×w window time series matrix with a step size of 1 time point, and perform correlation calculation on the two windows to obtain an M×M FC matrix; extract the upper triangle of each FC matrix for vectorization and splicing, and autocorrelate the obtained (M×(M-1) / 2)×(t-w+1) matrix; finally obtain a (t-w+1)×(t-w+1) dynamic functional connectivity matrix to complete the FCD calculation;

[0020] A2.2. For the time series matrix in step A2.1, calculate the standard deviation of each ROI in each window to complete the SW-STD calculation;

[0021] A2.3. Calculate the FCD-STD correlation.

[0022] As a preferred technical solution of the present invention, the expression for calculating the FCD-STD correlation in the fMRI processing and analysis module in step A2.3 is as follows:

[0023]

[0024] Among them, R i represents the FCD-STD correlation of the ith brain region, t represents the scanning time of the subject, w represents the sliding window length, represents the jth column of the average FCD matrix, represents the first-order derivative of the jth column of the average FCD, R(*) represents the rank of the internal element, Represents the average rank of the internal elements. represents the first-order derivative of the jth column of the SW-STD matrix

[0025] As a preferred technical solution of the present invention, the specific steps of step A3 are as follows:

[0026] A3.1. The similarity of the whole-brain functional connectivity pattern between two voxels is obtained by calculating the formula, and the similarity matrix between each voxel is obtained;

[0027] A3.2. Average the similarity matrices of all subjects to obtain a group-level average similarity matrix;

[0028] A3.3. Use the neighbor propagation clustering method to determine the number of sub-areas at each area group level;

[0029] A3.4. The group-level partitioning results in step A3.3 were used as the initial points of individual clustering. The average similarity of each cluster initial point was calculated as the cluster center. The K-means clustering method was used to achieve individualized brain area division. The sub-region with the strongest functional connection between each sub-region and the sub-prefrontal cortex was selected as the TMS stimulation target.

[0030] As a preferred technical solution of the present invention, the calculation expression in step A3.1 is as follows:

[0031]

[0032] Among them, SS within and SS combined Represent the within-group variance and between-group variance, respectively. i and b i represents the functional connectivity value between function maps a and b at position i, m i represents the average value of the connection between function maps a and b at position i, Represents the average connectivity value of all locations in the functional map.

[0033] Compared with the prior art, the present invention provides a real-time enhanced individualized regulation system for the treatment of postpartum depression, which has the following beneficial effects:

[0034] 1. The present invention combines cognitive reappraisal and rTMS for the first time, uses FCD-STD correlation calculation and individualized partition map construction, and determines TMS stimulation targets considering the individual differences of each patient, which has an accurate mood relief effect on postpartum depression. At the same time, combined with cognitive reappraisal video training, it improves patients' cognitive disorders such as attention, working memory, and negative thinking patterns, and has a long-term and stable mood relief effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a schematic diagram of the system framework of the present invention;

[0036] Figure 2 It is a schematic diagram of the process of the present invention.

[0037] Figure 3 Schematic diagram of the process of finding the PPD-driving brain area in step A2. DETAILED DESCRIPTION

[0038] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0039] See also Figure 1 , a real-time enhanced individualized regulation system for the treatment of postpartum depression, including a data collection module, an fMRI processing and analysis module, and an rTMS regulation module;

[0040] The data collection module is used to collect resting-state fMRI image data and T1 structural image data (sMRI);

[0041] The fMRI processing and analysis module is used to pre-process the resting-state fMRI image data, find the PPD-driven brain areas, construct individualized partition maps, and determine the TMS stimulation targets;

[0042] The specific steps of the fMRI processing and analysis module are as follows:

[0043] A1. Preprocessing: Preprocess the resting-state fMRI image data, including removing the first N time points of the resting-state fMRI image data, and registering all time points to the first time point. Then, register the resting-state fMRI image data to the standard template space, smooth the data, and then perform a 0.01-0.1 Hz filter to regress the signals of motion, white matter, and cerebrospinal fluid.

[0044] A2. Find the PPD driving brain region: calculate the FCD-STD correlation, and select the region with the highest FCD-STD correlation as the driving brain region;

[0045] A3. Construct an individualized partition map and determine the TMS stimulation target: Calculate the similarity of the whole-brain functional connection pattern between any two voxels, obtain the similarity matrix between each voxel in a specific area, and complete the individualized brain area division. Select the sub-region with the strongest functional connection (FC) between each sub-region and the subdural prefrontal cortex (DLPFC) as the TMS stimulation target, and finally register the target back to the T1 structural image to guide the rTMS regulation module.

[0046] The rTMS regulation module includes a cognitive reappraisal video training unit and an rTMS enhancement regulation unit;

[0047] The cognitive reappraisal video training unit stores a cognitive reappraisal picture library and a cognitive reappraisal video library. The pictures in the cognitive reappraisal picture library are selected from the International Affective Picture System (IAPS), the Chinese Affective Picture System (CAPS), the Asian Youth Social Acceptance / Rejection Picture Library, and the self-built Family Relationship Picture Library. The videos stored in the cognitive reappraisal video library correspond one-to-one to the pictures in the cognitive reappraisal picture library;

[0048] The family relationship picture library is constructed as follows:

[0049] We screened and purchased the copyrights of 300 pictures related to family relationships from professional photo library websites, including husband-wife relationships, mother-in-law-and-daughter-in-law relationships, and parent-child relationships. We then recruited volunteers to rate the pictures using the following criteria: valence (from 1 = very negative, 5 = neutral, to 9 = very positive) and arousal (from 1 = very relaxed, 5 = neutral, to 9 = very aroused), as well as the rating of vicarious emotional experience of putting oneself in the shoes of a specific person in the picture (from 1 = very sad, 5 = neutral to 9 = very happy). We statistically analyzed all the ratings, classified the pictures, and constructed a picture library that specifically represents family relationships.

[0050] The cognitive reappraisal video library is divided into three sub-libraries according to the conditions: neutral videos, negative videos, and reappraisal negative videos. The videos in each sub-library are converted from the pictures in the picture library. The specific situation when the cognitive reappraisal video library is called is as follows:

[0051] 1) The neutral video sub-library was converted from 100 neutral pictures. When a neutral video was selected for viewing, 10 neutral videos were randomly selected from the neutral video sub-library as a baseline condition without emotion to test the emotional arousal effect of the subsequent experimental treatment;

[0052] 2) There are 200 unclassified negative videos in the negative video sub-library. When choosing a negative video to watch, 20 negative videos are randomly selected from the negative video sub-library to arouse the subject's emotions;

[0053] 3) There are 800 videos in the negative reappraisal sub-library, which are divided into four categories: life emergency events (death, car accident, widowhood, divorce, major disease, etc.), family relationships (husband-wife relationship, mother-in-law-daughter-in-law relationship, parent-child relationship, etc.), work environment (interpersonal relationship, work pressure, career and family conflict), socioeconomic status and social support. Each category consists of 200 videos corresponding to the situation. When choosing to watch a negative reappraisal video, 20 negative videos are randomly selected from the negative video sub-library, and after the video ends, an animation guidance clip is attached to guide the subjects to conduct cognitive reappraisal.

[0054] The rTMS enhancement control unit is used to stimulate the previously determined TMS stimulation target with a high-frequency pulse of 10 Hz when the cognitive reappraisal video training unit is called, and the start and end times correspond to the cognitive reappraisal training video.

[0055] Furthermore, the specific process of step A2 is as follows:

[0056] A2.1. Segment the preprocessed resting-state fMRI image data to generate an M×t time series matrix, where M represents the number of regions of interest (ROI) and t represents the scanning time of the subject; select a window with a length of w=30 time points to generate an M×w window time series matrix with a step size of 1 time point, and perform correlation calculation on the two windows to obtain an M×M FC matrix; extract the upper triangle of each FC matrix for vectorization and splicing, and autocorrelate the obtained (M×(M-1) / 2)×(t-w+1) matrix; finally obtain a (t-w+1)×(t-w+1) dynamic functional connectivity matrix to complete the FCD calculation;

[0057] A2.2. For the time series matrix in step A2.1, calculate the standard deviation of each ROI in each window to complete the SW-STD calculation;

[0058] A2.3. Calculate the FCD-STD correlation.

[0059] Furthermore, the expression for calculating the FCD-STD correlation in the fMRI processing and analysis module in step A2.3 is as follows:

[0060]

[0061] Among them, R i represents the FCD-STD correlation of the ith brain region, t represents the scanning time of the subject, w represents the sliding window length, represents the jth column of the average FCD matrix, represents the first-order derivative of the jth column of the average FCD matrix, R(*) represents the rank of the internal element, Represents the average rank of the internal elements. represents the first-order derivative of the j-th column of the SW-STD matrix.

[0062] Furthermore, the specific steps of step A3 are as follows:

[0063] A3.1. The similarity of the whole-brain functional connectivity pattern between two voxels is obtained by calculating the formula, and the similarity matrix between each voxel is obtained;

[0064] A3.2. Average the similarity matrices of all subjects to obtain a group-level average similarity matrix;

[0065] A3.3. Use the neighbor propagation clustering method to determine the number of sub-areas at each area group level;

[0066] A3.4. The group-level partitioning results in step A3.3 were used as the initial points of individual clustering. The average similarity of each cluster initial point was calculated as the cluster center. The K-means clustering method was used to achieve individualized brain area division. The sub-region with the strongest functional connection between each sub-region and the sub-prefrontal cortex was selected as the TMS stimulation target.

[0067] Furthermore, the calculation expression in step A3.1 is as follows:

[0068]

[0069] Among them, SS within and SS combined Represent the within-group variance and between-group variance, respectively. i and b i represents the functional connectivity value between function maps a and b at position i, m i represents the average value of the connection between function maps a and b at position i, Represents the average connectivity value of all locations in the functional map.

[0070] The specific process of cognitive reassessment of the present invention is as follows:

[0071] 1) The subjects sat in a quiet TMS treatment room, about 60-70 cm away from the monitor, with horizontal and vertical viewing angles less than 6°.

[0072] 2) The cognitive reappraisal training video is about 25 minutes long and is divided into three parts:

[0073] (1) Watching a neutral video for 2 minutes: At the beginning of this stage, a white fixation dot “+” appeared on the black computer screen, followed by a prompt word and a prompt sound “Please watch the video carefully”. After the prompt word disappeared, a blank screen was presented for 1 second, followed by 10 neutral videos, with a 2-second interval between each video.

[0074] (2) Watching the negative video for 4-5 minutes: At the beginning of this stage, a white fixation dot “+” appeared on the black computer screen, followed by the prompt word “Please watch the video carefully and experience the emotional feelings brought by the video” and the prompt sound. After the prompt word disappeared, a blank screen was presented for 1 second, followed by the presentation of the negative event video, with a 2-second interval between each video;

[0075] (3) 16-17 min negative event reappraisal task. At the beginning of this stage, a white fixation dot “+” appeared on the black computer screen, followed by the prompt word and sound “Please watch the video carefully, experience the emotions brought by the video, follow the instructions of the video, and positively reappraise the negative events shown in the video.” After the prompt word disappeared, a blank screen appeared for 1 second, followed by a negative reappraisal video. There was a 3-second interval between the negative event and the reappraisal instructions. During the video presentation, the subjects were required to adjust their emotions as much as possible according to the instructions. The interval between each video was 5 seconds, which made it easier for the subjects to adjust their emotions. There was a 30-second interval between the first and second blocks, and a 1-minute break between the second and third blocks, so that the subjects’ moods could return to the baseline level.

[0076] 3) rTMS setting: When the cognitive reappraisal training video was playing, the rTMS coil was placed on the previously determined TMS stimulation target based on neuronavigation. During the video presentation, a high-frequency pulse of 10 Hz was used to stimulate the corresponding target, and the start and end times corresponded to the cognitive reappraisal training video;

[0077] Table 1 Comparison of basic data of subjects and changes in scores before and after emotional relief

[0078]

[0079] Note: PPD denotes postpartum depression; EPDS scores denote Edinburgh Postpartum Depression Scale; PSQ scores denote Postpartum Support Questionnaire.

[0080] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A real-time enhanced individualized control system for the treatment of postpartum depression, characterized in that: It includes data collection module, fMRI processing and analysis module and rTMS regulation module; The data collection module is used to collect resting state fMRI image data and T1 structural image data; The fMRI processing and analysis module is used to pre-process the resting-state fMRI image data, and then construct an individualized partition map and determine the TMS stimulation target after finding the PPD-driven brain area; The specific steps of the fMRI processing and analysis module are as follows: A1. Preprocessing: Preprocess the resting fMRI image data, including removing the first N time points of the resting fMRI image data, and registering all time points to the first time point. Then, register the resting fMRI image data to the standard template space, smooth the data, and then perform a 0.01-0.1 Hz filter to regress the motion, white matter, and cerebrospinal fluid signals. A2. Find the PPD driving brain region: calculate the FCD-STD correlation, and select the region with the highest FCD-STD correlation as the driving brain region; A3. Construct an individualized partition map and determine the TMS stimulation target: Use the driving brain area as the seed point to calculate the functional connection between each voxel and the whole brain voxel, then calculate the similarity of the whole brain functional connection pattern between any two voxels, obtain the similarity matrix between each voxel in a specific area, and complete the individualized brain area division. Among them, the sub-area with the strongest functional connection with the dorsolateral prefrontal cortex is determined as the TMS stimulation target, and finally the target is aligned back to the T1 structural image to guide the rTMS regulation module; The specific steps of step A3 are as follows: A3.

1. The similarity of the whole-brain functional connectivity pattern between two voxels is obtained by calculating the formula, and the similarity matrix between each voxel is obtained; A3.

2. Average the similarity matrices of all subjects to obtain a group-level average similarity matrix; A3.

3. Use the neighbor propagation clustering method to determine the number of sub-areas at each area group level; A3.4, using the group-level partitioning results in step A3.3 as the initial point of individual clustering, calculating the average similarity of each cluster initial point as the cluster center, using the K-means clustering method to achieve individualized brain region division, and selecting the subregion with the strongest functional connection between each subregion and the subprefrontal cortex area as the TMS stimulation target; The rTMS regulation module includes a cognitive reappraisal video training unit and an rTMS enhancement regulation unit; The cognitive reappraisal video training unit stores a cognitive reappraisal picture library and a cognitive reappraisal video library, wherein the cognitive reappraisal picture library includes a family relationship picture library, in which pictures of family relationships are stored, including husband-wife relationships, mother-in-law-in-law relationships, and parent-child relationships, and the videos stored in the cognitive reappraisal video library correspond one-to-one to the pictures in the cognitive reappraisal picture library; The rTMS enhancement control unit is used to stimulate the previously determined TMS stimulation target point with a high-frequency pulse of 10 Hz when the cognitive reappraisal video training unit is called, and the start and end times correspond to the cognitive reappraisal training video.

2. A real-time enhanced individualized control system for the treatment of postpartum depression according to claim 1, characterized in that: The specific process of step A2 is as follows: A2.

1. Segment the preprocessed resting-state fMRI image data to generate an M×t time series matrix, where M represents the number of regions of interest and t represents the scanning time of the subject; select a window with a length of w=30 time points to generate an M×w window time series matrix with a step size of 1 time point, and perform correlation calculation on the two windows to obtain an M×M FC matrix; extract the upper triangle of each FC matrix for vectorization and concatenation, and autocorrelate the obtained (M×(M-1) / 2)×(t-w+1) matrix; finally obtain a (t-w+1)×(t-w+1) dynamic functional connectivity matrix to complete the FCD calculation; A2.

2. For the time series matrix in step A2.1, calculate the standard deviation of each ROI in each window to complete the SW-STD calculation; A2.

3. Calculate the FCD-STD correlation.

3. A real-time enhanced individualized control system for the treatment of postpartum depression according to claim 2, characterized in that: The expression for calculating the FCD-STD correlation in the fMRI processing and analysis module in step A2.3 is as follows: Among them, R i represents the FCD-STD correlation of the i-th brain region, t represents the scanning time of the subject, w represents the sliding window length, represents the jth column of the average FCD matrix, represents the first-order derivative of the jth column of the average FCD, R(*) represents the rank of the internal element, Represents the average rank of the internal elements. represents the first-order derivative of the j-th column of the SW-STD matrix.

4. A real-time enhanced individualized control system for the treatment of postpartum depression according to claim 1, characterized in that: The calculation expression in step A3.1 is as follows: Among them, SS within and SS combined Represent the within-group variance and between-group variance, respectively. i and b i represents the functional connectivity value between function maps a and b at position i, m i represents the average value of the connection between function maps a and b at position i, Represents the average connectivity value of all locations in the functional map.

Images