Disclosure of Invention
It is an object of the present invention to overcome the above-mentioned drawbacks of the prior art by providing a method, device, processor and computer readable storage medium thereof for achieving TMS treatment for mood disorders and anxiety disorders based on mood induction.
To achieve the above object, the method, apparatus, processor and computer readable storage medium thereof for achieving TMS treatment for mood disorders and anxiety disorders based on mood induction of the present invention are as follows:
The control treatment method for mood disorder and anxiety disorder based on emotion induction in TMS treatment is mainly characterized by comprising the following steps:
(1) Clinical evaluation of the mental and psychological of the patient is carried out by using basic emotion element materials and complex emotion stimulation element materials before treatment;
(2) Performing brain magnetic resonance examination on the patient based on the emotion-induced state;
(3) Performing priority ranking on the selected individualized targets according to clinical evaluation of the patient and abnormal degree of emotion-induced lower brain functional activities;
(4) Setting a TMS treatment scheme under the emotion induction scene according to the target point priority ranking sequence;
(5) Performing treatment evaluation in a short-term period, and performing long-term prediction of a future residual treatment track by using a treatment effect prediction model;
(6) And feeding back the treatment track to a local database after treatment is finished, and carrying out clinical evaluation after treatment.
Preferably, the individualizing target in the step (3) includes frontal lobe under negative emotion, frontal lobe under positive emotion, parietal lobe and temporal parietal region, and the prioritization is performed according to the following manner:
(3.1) performing head movement detection and removal, noise signal removal, time sequence high-pass filtering, image space registration and linear regression analysis processing on the acquired brain function images under the emotion induction state, so as to obtain brain activation mapping diagrams under various emotion situations;
(3.2) extracting activation map data of brain regions related to emotion processing of a subject through a standard brain partition map, wherein the brain regions at least comprise a below-knee anterior cingulate gyrus, an inner forehead lobe, a prefrontal lobe, a dorsolateral forehead lobe and a temporal vertex joint region and an auxiliary exercise region;
(3.3) obtaining activation map data of the brain regions of the healthy reference population according to the flow, establishing a normal model for each brain region and the average activation level under each situation, and fitting probability distribution parameters of the normal model, including but not limited to center position and standard deviation;
(3.4) for each brain region of the individual patient, the degree of deviation Z is calculated as follows, in comparison with the distribution of the healthy reference group obtained in step (3.3):
Z=[x-mu]/std;
Wherein x is the activation value of the brain region of the patient, mu is the activation mean value of the brain region of the healthy control, and std is the activation standard deviation of the brain region of the healthy control;
(3.5) summarizing the degree of deviation Z of the activation level of all brain regions of the individual patients under all situations, sorting the absolute values of the degree of deviation Z from large to small, recording the sorting and the corresponding brain regions and activation situations, and thus completing the prioritization processing.
Preferably, the step (3) further comprises the following calculation process of the personalized brain function regulation target point:
a. Acquiring situation time sequence data of a target brain region, functional connection matrix data of the target brain region and situation activation map data of the target brain region based on brain functional magnetic resonance scanning processing results of a patient;
b. Respectively carrying out space linear change on the situation time sequence data of the target brain region and the functional connection matrix data of the target brain region to obtain a situation time sequence template of a reference group and a functional connection matrix template of the reference group, and carrying out comprehensive space linear transformation on the two templates and the situation activation map data of the target brain region together to obtain a situation activation map aligned with the reference group;
c. The method comprises the steps of referring to a group normal model, obtaining the activation deviation degree of a target brain region in various situations, sequencing the target brain region according to the deviation degree from large to small, and recording the situations corresponding to the corresponding deviation degree;
d. Taking 1 to 3 target brain areas as target points, judging whether the selected target points are more than 3cm away from the scalp or not, if so, calculating the functional connection of the target points and the cortex area, selecting a stimulable position with the maximum forward functional connection value as a substitute target point, otherwise, directly recording the positions of the target points, the directions deviating from the reference group and the situation attribute with the maximum deviation value.
Preferably, said step (3.3) builds said constant model as follows:
(3.3.1) extracting a mask of the target brain region from the standard map;
(3.3.2) expanding 5 voxels for each directional edge on the mask;
(3.3.3) fMRI data applied to each patient, obtaining individual data of the target brain region;
(3.3.4) performing procruste transformation on the time sequence of the target brain region to obtain an average space-time signal;
(3.3.5) calculating the correlation coefficient of the time sequence of each voxel in the target brain region and other voxels outside the target brain region, and obtaining a correlation matrix according to the correlation coefficient;
(3.3.6) calculating a related brain activation map under each emotional situation condition by adopting a linear regression model;
(3.3.7) carrying out procruste transformation on the related matrix to obtain an average related matrix;
(3.3.8) respectively obtaining a space-time signal template of the target brain region and a functional connection matrix template of the target brain region, and calculating an individual transformation matrix for time sequence, an individual transformation matrix for functional connection and corresponding weighted projection transformation;
and (3.3.9) performing activation map projection processing according to the obtained brain activation map, so as to obtain a normal mode model of the activation degree of the target brain region.
Preferably, the step (4) specifically includes the following steps:
(4.1) selecting at least 3 segments of stimulating materials with the same contextual attribute tag for the current patient in a mood scenario material library for mood induction;
(4.2) selecting a stimulation target point associated with the current stimulation material, and selecting a corresponding iTBS stimulation mode, a 10Hz rTMS stimulation mode or a 1Hz rTMS mode for TMS regulation treatment;
(4.3) after the stimulation task of one stage is finished, timely judging the correct and wrong of the mathematical calculation formula by keys to perform the distraction task test treatment, immediately watching the next situation stimulation material after the distraction task is finished, and repeating the operations;
(4.4) according to the number of selected targets, brain modulation treatments of all stimulation targets are arranged in the same day time and are performed at intervals such that the emotional stimuli received before each TMS modulation have the same properties to give an accurate TMS treatment regimen.
Preferably, the step (5) specifically includes the following steps:
(5.1) mid-term HAMD scoring or PHQ9 self-scoring when 1/4, 1/2 of treatment is completed, respectively;
(5.2) taking the treatment base line as an origin, estimating the mid-1/4 and 1/2 phases as path intermediate points, drawing a treatment effect expected range according to the improvement rate of the estimated mid-phase treatment effect of the linear mixed model, and projecting the treatment effect expected range to the expected treatment effect according to the 95% CI range;
(5.3) if the range of 50% reduction rate is <5% according to the treatment starting origin, 1/4 and 1/2 HAMD reduction rate projection to the prediction range of the final treatment effect, judging that the treatment benefit is not good, removing the first-ranked target point from the treatment scheme, and replacing the first-ranked target point in the alternative target point;
And (5.4) archiving the treatment track after the treatment is finished, and optimizing the parameter estimation of the mid-term treatment effect estimated by the linear mixed model.
The device for realizing the control treatment of mood disorder and anxiety disorder based on emotion induction in TMS treatment is mainly characterized in that the device comprises:
a processor configured to execute computer-executable instructions;
and a memory storing one or more computer executable instructions which, when executed by the processor, implement the steps of implementing a control treatment method for mood disorders and anxiety disorders based on emotion induction in the TMS therapy described above.
The processor for realizing the control treatment of mood disorders and anxiety disorders based on emotion induction in TMS treatment is mainly characterized in that the processor is configured to execute computer executable instructions, and the computer executable instructions realize the steps of the control treatment method for realizing mood disorders and anxiety disorders based on emotion induction in TMS treatment when being executed by the processor.
The computer readable storage medium is mainly characterized in that the computer program is stored thereon, and the computer program can be executed by a processor to realize the steps of the control treatment method for mood disorders and anxiety disorders based on emotion induction in the TMS treatment.
The method, the device, the processor and the computer readable storage medium for realizing TMS treatment based on mood disorder and anxiety disorder can effectively display that a patient can intuitively see obvious differences between the activities of different brain areas and health control groups under different situations through the scene stimulus library and the situation brain function images, the directions of the differences are related to the situations, the effectiveness of situation stimulus materials and the effectiveness of the situation brain function images on the selection of individual target regulation targets are effectively illustrated, and the method has obvious application value.
Detailed Description
In order to more clearly describe the technical contents of the present invention, a further description will be made below in connection with specific embodiments.
Before describing in detail embodiments that are in accordance with the present invention, it should be observed that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
Referring to fig. 1, the method for implementing control treatment for mood disorders and anxiety disorders based on emotion induction in TMS treatment includes the following steps:
(1) Clinical evaluation of the mental and psychological of the patient is carried out by using basic emotion element materials and complex emotion stimulation element materials before treatment;
(2) Performing brain magnetic resonance examination on the patient based on the emotion-induced state;
(3) Performing priority ranking on the selected individualized targets according to clinical evaluation of the patient and abnormal degree of emotion-induced lower brain functional activities;
(4) Setting a TMS treatment scheme under the emotion induction scene according to the target point priority ranking sequence;
(5) Performing treatment evaluation in a short-term period, and performing long-term prediction of a future residual treatment track by using a treatment effect prediction model;
(6) And feeding back the treatment track to a local database after treatment is finished, and carrying out clinical evaluation after treatment.
As a preferred embodiment of the present invention, the individualizing targets in the step (3) include frontal lobe under negative emotion, frontal lobe under positive emotion, parietal lobe and temporal parietal region, and the prioritization is performed according to the following manner:
(3.1) performing head movement detection and removal, noise signal removal, time sequence high-pass filtering, image space registration and linear regression analysis processing on the acquired brain function images under the emotion induction state, so as to obtain brain activation mapping diagrams under various emotion situations;
(3.2) extracting activation map data of brain regions related to emotion processing of a subject through a standard brain partition map, wherein the brain regions at least comprise a below-knee anterior cingulate gyrus, an inner forehead lobe, a prefrontal lobe, a dorsolateral forehead lobe and a temporal vertex joint region and an auxiliary exercise region;
(3.3) obtaining activation map data of the brain regions of the healthy reference population according to the flow, establishing a normal model for each brain region and the average activation level under each situation, and fitting probability distribution parameters of the normal model, including but not limited to center position and standard deviation;
(3.4) for each brain region of the individual patient, the degree of deviation Z is calculated as follows, in comparison with the distribution of the healthy reference group obtained in step (3.3):
Z=[x-mu]/std;
Wherein x is the activation value of the brain region of the patient, mu is the activation mean value of the brain region of the healthy control, and std is the activation standard deviation of the brain region of the healthy control;
(3.5) summarizing the degree of deviation Z of the activation level of all brain regions of the individual patients under all situations, sorting the absolute values of the degree of deviation Z from large to small, recording the sorting and the corresponding brain regions and activation situations, and thus completing the prioritization processing.
As a preferred embodiment of the present invention, the step (3) further includes performing calculation processing of the personalized brain function control target in the following manner:
a. Acquiring situation time sequence data of a target brain region, functional connection matrix data of the target brain region and situation activation map data of the target brain region based on brain functional magnetic resonance scanning processing results of a patient;
b. Respectively carrying out space linear change on the situation time sequence data of the target brain region and the functional connection matrix data of the target brain region to obtain a situation time sequence template of a reference group and a functional connection matrix template of the reference group, and carrying out comprehensive space linear transformation on the two templates and the situation activation map data of the target brain region together to obtain a situation activation map aligned with the reference group;
c. The method comprises the steps of referring to a group normal model, obtaining the activation deviation degree of a target brain region in various situations, sequencing the target brain region according to the deviation degree from large to small, and recording the situations corresponding to the corresponding deviation degree;
d. Taking 1 to 3 target brain areas as target points, judging whether the selected target points are more than 3cm away from the scalp or not, if so, calculating the functional connection of the target points and the cortex area, selecting a stimulable position with the maximum forward functional connection value as a substitute target point, otherwise, directly recording the positions of the target points, the directions deviating from the reference group and the situation attribute with the maximum deviation value.
As a preferred embodiment of the present invention, the step (3.3) builds the normal mode model as follows:
(3.3.1) extracting a mask of the target brain region from the standard map;
(3.3.2) expanding 5 voxels for each directional edge on the mask;
(3.3.3) fMRI data applied to each patient, obtaining individual data of the target brain region;
(3.3.4) performing procruste transformation on the time sequence of the target brain region to obtain an average space-time signal;
(3.3.5) calculating the correlation coefficient of the time sequence of each voxel in the target brain region and other voxels outside the target brain region, and obtaining a correlation matrix according to the correlation coefficient;
(3.3.6) calculating a related brain activation map under each emotional situation condition by adopting a linear regression model;
(3.3.7) carrying out procruste transformation on the related matrix to obtain an average related matrix;
(3.3.8) respectively obtaining a space-time signal template of the target brain region and a functional connection matrix template of the target brain region, and calculating an individual transformation matrix for time sequence, an individual transformation matrix for functional connection and corresponding weighted projection transformation;
and (3.3.9) performing activation map projection processing according to the obtained brain activation map, so as to obtain a normal mode model of the activation degree of the target brain region.
As a preferred embodiment of the present invention, the step (4) specifically includes the following steps:
(4.1) selecting at least 3 segments of stimulating materials with the same contextual attribute tag for the current patient in a mood scenario material library for mood induction;
(4.2) selecting a stimulation target point associated with the current stimulation material, and selecting a corresponding iTBS stimulation mode, a 10Hz rTMS stimulation mode or a 1Hz rTMS mode for TMS regulation treatment;
(4.3) after the stimulation task of one stage is finished, timely judging the correct and wrong of the mathematical calculation formula by keys to perform the distraction task test treatment, immediately watching the next situation stimulation material after the distraction task is finished, and repeating the operations;
(4.4) according to the number of selected targets, brain modulation treatments of all stimulation targets are arranged in the same day time and are performed at intervals such that the emotional stimuli received before each TMS modulation have the same properties to give an accurate TMS treatment regimen.
As a preferred embodiment of the present invention, the step (5) specifically includes the steps of:
(5.1) mid-term HAMD scoring or PHQ9 self-scoring when 1/4, 1/2 of treatment is completed, respectively;
(5.2) taking the treatment base line as an origin, estimating the mid-1/4 and 1/2 phases as path intermediate points, drawing a treatment effect expected range according to the improvement rate of the estimated mid-phase treatment effect of the linear mixed model, and projecting the treatment effect expected range to the expected treatment effect according to the 95% CI range;
(5.3) if the range of 50% reduction rate is <5% according to the treatment starting origin, 1/4 and 1/2 HAMD reduction rate projection to the prediction range of the final treatment effect, judging that the treatment benefit is not good, removing the first-ranked target point from the treatment scheme, and replacing the first-ranked target point in the alternative target point;
And (5.4) archiving the treatment track after the treatment is finished, and optimizing the parameter estimation of the mid-term treatment effect estimated by the linear mixed model.
In one embodiment of the invention, the overall process flow of the method for effecting TMS treatment for mood and anxiety disorders based on mood induction is as follows:
1. clinical assessment of the mental mind of the patient;
2. performing magnetic resonance examination of the brain, including functional scans in structural, resting functions, and various emotional contexts;
3. data by functional data processing under emotion induction, individual targets of frontal lobe (sgACC, MPFC, DLPFC, SMA), parietal lobe (Precuneus, IPL) and temporal-top combined region (TPJ) under negative and positive emotion were selected. The targets are prioritized according to the clinical evaluation of the patient and the abnormal degree of emotion-induced lower brain functional activity.
And 3a, processing brain function images under emotion induction by using PhiPipe multi-mode nerve image processing systems (Hu et al, 2022 and Human Brain Mapping) published by the inventor team, wherein main steps comprise head movement detection and removal, noise signal removal, time sequence high-pass filtering, image space registration, linear regression analysis and the like, so as to obtain brain activation maps of various emotion situations.
Extracting activation map data of brain regions related to emotion processing of a subject through a standard brain partition map, wherein the brain regions at least comprise a below-knee anterior cingulate gyrus, an inner forehead lobe, a prefrontal lobe, an outer dorsiflexion forehead lobe and a temporal vertex joint region and an auxiliary exercise region.
Obtaining activation map data of the brain regions of a healthy reference group (at least 50 people) according to the procedures of the steps 2, 3a and 3b, establishing a normal model for each brain region and the average activation level under each situation, and fitting probability distribution parameters of the normal model, including but not limited to a central position and a standard deviation.
For each brain region of the patient individual, the mean activation level in each situation is compared with the distribution of the healthy reference group obtained in 3c, and the degree of deviation is calculated as Z= [ x-mu ]/std, where x is the patient brain region activation value, mu is the healthy control brain region activation mean, and std is the healthy control brain region activation standard deviation.
3E, summarizing the degree of deviation Z of the activation level in all brain regions of the individual patient, in all situations, the absolute value of Z is ordered from big to small. The ordering and corresponding brain regions and activation scenarios are recorded.
4. And formulating a TMS treatment scheme under the emotion induction scene according to the target point priority ordering condition, wherein the TMS treatment scheme comprises the step of screening a matched optimization scheme from a treatment scheme library. In TMS treatment, the brain regions and corresponding situations of a plurality of top ranks in the ranking table obtained in 3e are selected as intervention targets and corresponding situations. In the intervention, firstly, selecting materials with the same situation attribute as the corresponding situation of the current target point from the emotion situation material library, enabling a patient to see, hear or experience, and then stimulating the corresponding target point by using a TMS stimulation system. The above process may be repeated multiple times with a rest time between each and before the next contextual stimulus begins, the use of a distraction task reduces the previous emotional experience. Each repetition may select a different brain region and its corresponding contextual attributes from the ranking table obtained in 3 e. The course of treatment is 10 days or less, and a brief symptom assessment is performed daily.
5. Mental and psychological assessment was performed after 1/4 and 1/2 of the course of treatment was completed. And according to the feedback data, predicting the future residual treatment track by using a treatment effect prediction model, and if the predicted improvement amplitude is lower than 50%, sequencing and adjusting the treatment scheme in order according to the target point priority.
6. And after treatment is finished, feeding back the treatment track to the system database, and performing iterative optimization on the personalized target spot priority ranking algorithm.
In practical application, the technical scheme is divided into basic emotion and complex emotion stimulation for the design of the situation material. The specific process flow is as follows:
1. Elemental material of basic emotion contains 3 types, no obvious emotion, sad emotion, and happy emotion, respectively. Each length of material was about 30s. And (5) carrying out finishing numbering on the screened materials and storing basic data. In order to simulate the psychological process of emotion fluctuation and emotion conversion in actual life, on the basis of basic emotion element materials, the element materials are combined into an overall stimulating material according to the average value and variance, social attribute, application range and the like of each dimension evaluation of different emotion materials in the following sequence: 1-2-3-4-1-2-3-4, wherein the numbers represent 1-no clear emotion, 2-negative emotion, 3-positive emotion, 4-distraction task, respectively. The distraction task material is to determine if a calculation equation is correct within a finite time.
2. Complex emotional materials employ multiple segments of video between 30 seconds and 2 minutes and 30 seconds long, each segment of video (elemental material) telling a social episode with mood swings and storyline inversions. Unlike basic emotional element materials, complex emotional materials are composed of different human stories, each material has relative integrity and coherence inside, has advantages in emotion arousal and content inclusion, and can involve more complex emotional content such as homonymies, emotions, anxiety, pain and the like. The emotional materials library is composed of multiple versions of the above contextual materials, each of which is associated with an emotional valence and arousal degree score, and an event attribute score. A database is established to manage the material segments and their corresponding scores and attributes.
As shown in fig. 2, the method for calculating the individual brain function regulation target under the emotion situation specifically comprises the following processing steps:
1. According to the video library emotion-induced video material and the second-by-second negative emotion-neutral emotion-positive emotion evaluation label thereof, a general linear model Y=BX+A+E of individual level is established, wherein Y is a real signal of each brain region, X is a vector generated by convolving negative-neutral-positive three emotion coordinates defined according to an emotion response curve with a blood oxygen dynamic function, E is a calculation error, A is an intercept, represents a baseline activation level, B is a regression coefficient vector, represents the linear relation magnitude of X and Y, represents the brain activation level under each emotion situation, and can be represented as [ B1, B2, B3 and b4. ]. Wherein b1 is a passive emotional response parameter, b2 is a neutral emotional response parameter, b3 is a positive emotional response parameter, and b4 and later coefficients are contributions of control variables. Let b1-b2 be the passive mood brain activation mode and b3-b2 be the active mood brain activation mode. A single sample T test judges that b1-b2 or b3-b2 is not 0 obviously as the brain activation standard.
2. And constructing emotion situation brain function activation normal model reference models of more than 50 healthy groups, and describing brain function activation characteristics and group variation levels of the healthy groups. Target brain areas involved in the following steps include, but are not limited to, anterior wedge lobes, anterior below knee closure, medial forehead lobes, auxiliary movement areas, visual cortex, and right temporal-top combination.
In practical application, as shown in fig. 3, the specific method for constructing the emotion situation brain function activation normal mode reference model in the technical scheme is as follows:
1.1. The mask used to mark the target brain region is extracted from the standard brain atlas.
1.2 On the mask, 5 voxels are inflated for each direction.
1.3. For fMRI data of each subject, after a preprocessing step (including at least steps of head motion correction, noise signal rejection, and registration with a standard brain map), using the mask generated in step 1.2, a time-series signal of the target brain region is extracted, represented by a matrix Di, including T rows (the number of time points), V columns (the number of voxels), and i represents the serial number of the subject.
1.4. For Di and Dj (i=2n, j=2n+1, n=0.. The floor (nsubj/2), where floor (nsubj/2) is a downward integer of 1/2 of the number of subjects), the following calculations are made (where steps a-d are part of the procruste transformation):
a) Calculating Di ', wherein each column dik ' = dik-mean (dik), dj ', wherein each column djk ' = djk-mean (djk), k=1, 2, V, each column of Di ' represents a normalized representation of the time series for each voxel of subject i, with a time series mean of 0.
B) Singular value decomposition is calculated, USVT = Di'(Dj')T, wherein U and V are orthogonal matrices, S is a diagonal matrix, and S is a eigenvalue of a projection matrix of data of the subject i to j.
C) R=uvT is calculated, R represents the rotational transformation of the data from subject i to subject j.
D) C=tr (S)/tr ((Di ')T Di') is calculated, where tr represents the trace of the matrix and c represents the scaling factor in the spatial transformation of subject i to subject j.
E) Dij= cRDj' +Tby1mean(di)T is calculated, wherein 1Tby1 is a T-dimensional column vector with all elements being 1, and Dij represents data obtained by performing procruste spatial transformation on a data matrix of a subject i to a data matrix of the subject j.
F) Exchanging Di and Dj, and repeating the steps a-e to obtain Dji.
G) And averaging Dij and Dji to obtain Di+j.
1.5. Step 1.4 is iteratively performed for each two subjects Di, dj calculating the obtained plurality of di+j until a unique di+j is obtained, and the last obtained unique di+j is denoted DTEMPLATE. DTEMPALTE is the average data matrix of all subjects after the procruste transformation.
1.6. Taking time series signals Di and DTEMPLATE of a target brain region of each subject as input of the step 1.4, performing calculation of 1.4a to 1.4d to obtain ci and Ri, wherein ci and Ri respectively represent scaling and selection transformation after the data matrix of the subject i is subjected to procruste transformation to DTEMPLATE.
1.7. For a subject i, calculating a correlation matrix of an average time sequence of the target brain region and other brain regions to obtain a correlation coefficient matrix Ei, wherein the correlation coefficient matrix comprises P rows and V columns, P is the number of other unexpected brain regions of the target brain region, and V is the voxel number of the target brain region.
1.8. With Ei and Ej (i=2n, j=2n+1, n=0..floor (nsubj/2), wherein floor (nsubj/2) is a downward integer of 1/2 of the number of subjects) replaces Di and Dj in step 1.4, and all calculation steps of step 1.4 are performed.
1.9. For each two subjects Ei, ej, calculating the obtained plurality of Di+j, iteratively executing step 1.4 until a unique Ei+j is obtained, and marking the unique Ei+j obtained finally as ETEMPLATE, wherein ETEMPLATE represents the average data matrix of the correlation matrices of all the subjects after the correlation matrices are transformed by each other procruste.
1.10. Taking time series signals Ei and ETEMPLATE of a target brain area of each subject as input of the step 1.4, performing calculation of 1.4a to 1.4d to obtain ci 'and Ri', wherein ci and Ri respectively represent scaling and selection transformation after procruste transformation of a functional connection matrix of the subject i to ETEMPLATE.
1.11. For each subject's data, calculate;Where i represents subject number, lamda is constant, 0< lamda <1, cci and RRi represent scaled and rotated transforms after weighted averaging of transforms of subject i to DTEMPLATE and ETEMPALTE, respectively.
1.12. For the data of each subject, calculating a brain activation graph related to each emotion situation condition by adopting a linear regression model, and recording as ACTi, wherein i represents a subject serial number, ACTi is a matrix of m rows and V columns, m is the situation number, and V is the voxel number of a target brain region.
1.13. For each subject's data, ACTi' was calculated, with each column: ACTik '= ACTik-mean (ACTik), k=1, 2,; ACTi' is the normalized data of the activation map data of subject i after the normalization between contexts, and the average value of each column is 0.
1.14. For each subject's data, calculate. ACTi "is an activation map matrix of subject i transformed by RRi and cci procruste.
1.15. For each row with ACTi ", the mean and standard deviation of the corresponding column of ACTi" for all subjects are calculated and the parameter beta_v (mu, sigma, skew, kurtosis) of the distribution curve is fitted, where V represents the voxel number in the target brain region, v=1..v, i.e. the parameter set is fitted separately for each voxel V.
2. And (3) comparing the activation degree of the target brain region of the patient under various situations with the healthy population distribution curve of the corresponding target brain region obtained in the step (1) to obtain a lower degree value. The method comprises the following specific steps:
2.1 After the target brain region of the patient is subjected to the preprocessing step (at least including the steps of head motion correction, noise signal elimination and registration with a standard brain map), the mask generated in the step 1.2 is used for extracting time sequence signals of the target brain region, the time sequence signals are represented by a matrix D, T rows (time point number) and V columns (voxel number) are included as input variables, the calculation step 1.6 is executed, c and R are obtained, and the data D of the patient is respectively represented to carry out the scaling and rotation transformation of procruste transformation to the data template DTEMPLATE of the reference group population.
2.2 Calculating a correlation matrix of the average time sequence of the target brain region and other brain regions to obtain a correlation coefficient matrix E, wherein the correlation coefficient matrix E comprises P rows and V columns, P is the number of other unexpected brain regions of the target brain region, and V is the voxel number of the target brain region. Taking E as an input variable, performing a calculation step 1.10 to obtain c ', R', which respectively represent scaling and rotation transformation of patient data E into procruste transformation of the data templates ETEMPLATE of the reference group population.
2.3 Using the c, R, c ', R' obtained in the calculation steps 1.6 and 1.10 as input, a calculation step 1.11 is performed to obtain cc, RR, which is the scaled and rotated transformation after the procruste transformation of the integrated D and E, respectively.
2.4 And calculating a brain activation graph related to each emotion situation condition of the patient by adopting a linear regression model, and recording the brain activation graph as ACT, wherein m is the situation number, and V is the voxel number of the target brain region. As input variables, calculation steps 1.13 and 1.14 are performed to obtain ACT.
2.5 In combination with the distribution parameter beta_v (mu, sigma, skew, kurtosis) obtained at 1.15, the percentile of each voxel in the ACT "with respect to beta_v (v means corresponding to a voxel in the ACT") is calculated, the deviation value= |50-percentile|.
2.6 The deviation values of the target brain regions under various situations obtained by the calculation are ordered, the 1-3 target brain regions arranged in the front are used as individual stimulation targets, and the directions (higher and lower) of the deviation reference population and the situation attribute of the maximum deviation value of the brain regions are stored.
2.7 And (3) locating the stimulation target point, namely, determining the stimulation target point as a non-cortical stimulation target point when the distance between the stimulation target point and the scalp exceeds 3cm or is not suitable for TMS stimulation, calculating the functional connection between the original target point and the cerebral cortex area by using the situation stimulation time sequence data, and selecting a stimulable position with the maximum forward functional connection value as an indirect stimulation target point of the corresponding original target point.
As shown in fig. 4, the personalized TMS treatment regimen in the emotional context is specifically as follows:
1. And selecting at least 3 segments of stimulating materials with the same situation attribute label from the emotion material library according to the individuation brain regulation target determined by the method and the situation attribute label corresponding to the individuation brain regulation target.
2. In the brain regulation implementation stage, firstly, a patient watches the situation video material for 1min, and after watching is finished, the transcranial magnetic stimulation is immediately started to be repeated, wherein the target point is a target point which is related to the stimulation material and is determined by the calculation. The modes of stimulation included iTBS modes (a set of 350 Hz pulses, stimulated at 5Hz frequency for 2 min), 10Hz rTMS mode, and 1Hz rTMS mode. The specific stimulation mode is determined according to whether the activity of the stimulation target point of the patient is lower or higher than that of the reference population, and the stimulation duration and the stimulation time sequence can be customized according to the requirements of different stimulation modes.
3. At the end of the stimulation of one phase (or after the subject has been at rest), a distraction task is performed for 2min. The distraction task is to judge the accuracy of the mathematical calculation formula in time through keys. After the distraction task is completed, the next situational stimulus material is then viewed and the above is repeated.
4. Brain modulation of all stimulation targets is arranged in the same day and at intervals according to the number of selected targets and is given contextual video induction of the respective context prior to stimulation as described in step 2. The library of contextual materials is such that the emotional stimuli received before each TMS modulation have the same properties, but are not simply repeated.
As shown in fig. 5, the method for optimizing the personalized treatment scheme based on the treatment response short-term evaluation and the treatment response track in the technical scheme specifically comprises the following steps:
1. treatment completed 1/4, 1/2 were scored for metaphase HAMD or PHQ9 self-scores, respectively.
2. And taking the treatment base line as an origin, estimating the mid-term treatment effect as a path midpoint by 1/4 and 1/2 of the mid-term treatment effect, drawing a treatment effect expected range according to the improvement rate of the estimated mid-term treatment effect of the linear mixed model, and projecting the expected treatment effect according to the 95% CI range.
3. If the range of 50% reduction rate is <5% from the treatment start origin, 1/4 and 1/2 HAMD reduction rate projection to the final treatment effect prediction range, the treatment benefit is judged to be poor, the first-ranked target point is removed in the treatment scheme and replaced by the highest-ranked target point in the alternative targets, for example, the 1 st, 2 nd and 3 rd targets are adopted for initial treatment, and the 2 nd, 3 rd and 4 th targets are adopted when the treatment benefit is poor.
4. And after the treatment is finished, archiving the treatment track, and optimizing the parameter estimation of the linear mixed model for estimating the mid-term curative effect.
The device for realizing control treatment on mood disorder and anxiety disorder based on emotion induction in TMS treatment, wherein the device comprises:
a processor configured to execute computer-executable instructions;
and a memory storing one or more computer executable instructions which, when executed by the processor, implement the steps of implementing a control treatment method for mood disorders and anxiety disorders based on emotion induction in the TMS therapy described above.
The processor for realizing the control treatment of mood disorders and anxiety disorders based on emotion induction in TMS treatment is configured to execute computer executable instructions, and when the computer executable instructions are executed by the processor, the steps of the control treatment method for realizing mood disorders and anxiety disorders based on emotion induction in TMS treatment are realized.
The computer readable storage medium has stored thereon a computer program executable by a processor to perform the steps of implementing a control treatment method for mood disorders and anxiety disorders based on emotion induction in TMS therapy as described above.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing specific logical functions or steps of the process, and further implementations are included within the scope of the preferred embodiment of the present invention in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the present invention.
It is to be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution device.
Those of ordinary skill in the art will appreciate that all or a portion of the steps carried out in the method of the above-described embodiments may be implemented by a program to instruct related hardware, and the program may be stored in a computer readable storage medium, where the program when executed includes one or a combination of the steps of the method embodiments.
The above-mentioned storage medium may be a read-only memory, a magnetic disk or an optical disk, or the like.
In the description of the present specification, reference to the terms "one embodiment," "some embodiments," "examples," "specific examples," or "embodiments," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiments or examples. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
While embodiments of the present invention have been shown and described above, it will be understood that the above embodiments are illustrative and not to be construed as limiting the invention, and that variations, modifications, alternatives and variations may be made to the above embodiments by one of ordinary skill in the art within the scope of the invention.
The method, the device, the processor and the computer readable storage medium for realizing TMS treatment based on mood disorder and anxiety disorder can effectively display that a patient can intuitively see obvious differences between the activities of different brain areas and health control groups under different situations through the scene stimulus library and the situation brain function images, the directions of the differences are related to the situations, the effectiveness of situation stimulus materials and the effectiveness of the situation brain function images on the selection of individual target regulation targets are effectively illustrated, and the method has obvious application value.
In this specification, the invention has been described with reference to specific embodiments thereof. It will be apparent that various modifications and variations can be made without departing from the spirit and scope of the invention. The specification and drawings are, accordingly, to be regarded in an illustrative rather than a restrictive sense.