A transcranial magnetic stimulation target point positioning and quality monitoring processing method and system

Abstract

The application discloses a transcranial magnetic stimulation target point positioning and quality monitoring processing method and system, and belongs to the technical field of intelligent medical treatment, which comprises the following steps: extracting emotion situation-specific activation data of each brain region in brain function magnetic resonance data of a target individual and a healthy reference group; generating individualized target point coordinates through a quantitative deviation index based on the extracted data; judging situation induction quality, and statistically analyzing habituation decay trend of induction effect; dynamically scheduling and processing emotion induction video materials according to statistical detection results; jointly modeling based on self-evaluation humanoid scale valence score and visual analog scale depression symptom score data, and outputting symptom change trajectory and early non-response warning data; calculating related target point function data after regulation and control; and updating regulation and control parameters based on the obtained data. The application adopts the above method and system, realizes transcranial magnetic stimulation target point positioning and quality monitoring in an emotion situation induction state, and improves induction quality.

Description

Technical Field

[0001] This invention relates to the field of smart medical technology, and in particular to a method and system for transcranial magnetic stimulation target localization and quality monitoring. Background Technology

[0002] Transcranial magnetic stimulation (TMS) is a common neuromodulation technique for treating mood disorders, but its efficacy is generally limited. Mood disorders and anxiety disorders are emotion-dependent mood disorders, and conventional TMS treatments are performed in a resting state, neglecting the patient's neural responses under different emotional states. Studies have shown that identifying individualized treatment targets in specific emotional situations and combining them with situational induction for TMS treatment can achieve better results.

[0003] Currently, TMS (Transcranial Magnetic Resonance Imaging) methods combining situational evoked responses with individualized target localization have seen some development. Existing methods use functional magnetic resonance imaging (fMRI) of the brain under emotional evoked states to extract activation data from various brain regions in specific emotional situations, comparing this data with the mean and standard deviation of activation in a healthy control group to calculate the degree of individual deviation as a basis for target priority ranking. However, these methods still have shortcomings in the following aspects: existing target quantification indicators only reflect the degree to which individual activation values ​​deviate from the healthy mean, failing to comprehensively consider the functional importance of each brain region in normal emotional processing, leading to a discrepancy between target priority ranking and actual regulatory value; existing methods lack a continuous data monitoring mechanism for the quality of situational evoked responses, and the attenuation of evoked effects caused by repeated use of emotional materials during multiple regulation processes cannot be identified and intervened in a timely manner; furthermore, existing methods lack means to continuously track symptom changes during regulation, making it difficult to achieve real-time data feedback on the treatment effect. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for transcranial magnetic stimulation target localization and quality monitoring to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides a method for transcranial magnetic stimulation target localization and quality monitoring, comprising the following steps: S1. Based on functional magnetic resonance imaging data of the brain under emotional situation induced state, extract emotional situation-specific activation data of target individuals and healthy reference groups in each brain region; S2. Determine the activation effect of each brain region based on data from a healthy reference group; construct the activation distribution of the healthy population, calculate the individual activation percentile of each brain region of the target individual, and generate individualized target coordinates by quantifying deviation indicators. S3. Collect self-assessment valence scores of the human scale after the situation is induced during the control process, make real-time judgment on the effectiveness of the situation inducement, and regularly conduct statistical detection on the habituation decay trend of the inducement effect. S4. Dynamically schedule and process the video materials that induce emotions based on statistical detection results; S5. Collect depression symptom scores from the Visual Analogue Scale one by one, and combine them with the valence data from the Self-Assessment Human Scale to perform joint modeling, and output symptom change trajectory and early non-response warning data. S6. Obtain brain functional magnetic resonance imaging data after the modulation is completed. Based on the individualized target coordinates in step S2, calculate the target functional data and the corresponding clinical scale reduction after the emotional situation induces the state. Calculate the Spearman correlation coefficient between the target functional data and the corresponding clinical scale reduction. S7. Archive the target function data, corresponding clinical scale reductions, Spearman correlation coefficients, induced quality data, and symptom trajectory data to the local database, and update the regulation parameters.

[0006] Preferably, the process for extracting emotion context-specific activation data includes: Obtain subjective valence ratings for situational videos from several reference raters, mark continuous time periods with a valence consensus rate of no less than 60% as effective emotional situation intervals, and convolve these ratings with the blood oxygen dynamics response function to generate emotional situation regression factors. Substitute the emotional situation regression factor into the general linear model to calculate the activation contrast parameter between the emotional situation and the neutral situation. Based on the activation comparison parameters, activation parameter maps of the target individual and several healthy reference individuals in each brain region of interest were extracted to construct the activation distribution of the healthy group.

[0007] Preferably, step S2 specifically includes: S21. For the activation comparison parameters of the healthy control group in each brain region, calculate the effect size relative to the null hypothesis to construct a whole-brain effect size map. ; S22. Calculate the percentile of the target individual in the healthy reference distribution based on the activation value of each brain region. P The calculation formula is: ; in, For the target individual in the brain region j activation value, For the first k A healthy reference individual in the brain region j activation value, For the total number of healthy reference individuals, It is an indicator function; S23, based on the calculation and P， Calculate each brain region of the target individual jThe quantization deviation value Q is given by the formula: ; in, It is a smoothing factor; S24. Map the Q values ​​of each brain region of the target individual to voxel space to generate a whole-brain quantitative deviation map Q-map. Extract the three-dimensional coordinates of the brain region with the highest Q value within the preset target search mask range, and use them as the coordinates of the individualized transcranial magnetic stimulation target.

[0008] Preferably, step S3 specifically includes: S31. After each situational evoked process, collect the valence score of the Self-Assessment Manifestation Scale (SAM). , Use a scale of 1 to 5, where 1 represents completely positive and 5 represents completely negative; ≤2 is used as the effective threshold for situational induction. Output a valid flag when the value is ≤2. If the value is ≥3, the output is marked as insufficient and written to the induced quality log; S32, using the regulatory sequence number during the regulation process as the independent variable. As the dependent variable, each cumulative N The Spearman rank correlation test was performed on the second adjustment to calculate the correlation coefficient. and significance Value, and calculate the previous period's average. Compared with the later mean ; S33. Set habitual alert trigger conditions. A habitual alert will be triggered when any one of the trigger conditions is met. The trigger condition is: Condition A. >0 and <0.05; Condition B, is and ≥3.

[0009] Preferably, the mean of the first period is the mean of the first 1 / 5 of the sequence, and the mean of the second period is the mean of the last 1 / 5 of the sequence.

[0010] Preferably, step S4 specifically includes: after the habituation warning is triggered, video materials are retrieved from the material library according to the set priority order, wherein the first priority is video clips that have not been used in the current treatment course; the second priority is video clips with the lowest usage frequency and the largest distance from recently used clips in the pre-built content similarity matrix; the scheduling record is written to the scheduling log, and the statistical detection of step S32 is re-executed in the next detection cycle after the material is replaced.

[0011] Preferably, step S5 specifically includes: S51. Obtain the symptom score of the Visual Analogue Scale (VAS) before the start of each modulation sequence. , A continuous score from 0 to 100; S52. Perform linear regression on the entire VAS score sequence to calculate the slope of symptom improvement. ; S53. SAM valence score according to Turn to a positive direction, with The Spearman correlation coefficient was calculated for the sequences as a synergistic data indicator of context-induced quality and symptom changes; S54. When the mean VAS value in the current period decreases by less than the preset threshold from the baseline, output early warning data for no response.

[0012] Preferably, step S6 specifically includes: S61. Obtain T1 functional magnetic resonance imaging data of the brain at the second time point after the end of the modulation, and use the same general linear model parameters as in step S1 to extract the activation comparison parameter map of the target individual at the T1 time point in each brain region of interest. S62. Based on the individualized target brain regions obtained in step S2 The activation values ​​of the target individual in this brain region at time T1 are compared with the activation distribution matrix of the healthy group that has been constructed, and the percentile at time T1 is calculated using the same formula as in step S22. ; S63, Based on the percentile of time T0 and percentile at time point T1 Calculate the percentage change of target function ; S64. Obtain the standardized clinical symptom scores of the target individual at time points T0 and T1, and calculate the reduced components of the clinical scale. ; S65. After completing the full T0 to T1 data for multiple target individuals, use the percentage change in target function for each individual. Reduce the number of independent variables and clinical scales. Calculate the Spearman rank correlation coefficient for t is the dependent variable.

[0013] This invention also provides a transcranial magnetic stimulation target localization and quality monitoring system, comprising: The data extraction module is used to extract emotional context-specific activation data of each brain region from the T0 and T1 time points of the target individual and the functional magnetic resonance imaging data of the healthy reference group. The target localization module calculates the activation effect of each brain region and the individual activation percentile at time T0 of the target individual based on the activation data of the healthy reference group, and generates individualized target coordinates by quantifying deviation indicators. The evoked quality monitoring module uses the self-assessment human valence scale to make real-time effectiveness judgments on the quality of situational evoked emotions during the regulation process, regularly conducts statistical detection on the habitual decay trend of the evoked effect, and dynamically schedules and processes the emotional evoked video materials based on the statistical detection results. The symptom tracking module jointly models the self-assessment human figure scale valence score and the visual analog scale depression symptom score data, and outputs the symptom change trajectory and early non-response warning data. The target data calculation module calculates the target functional data, the corresponding clinical scale reduction, and the Spearman correlation coefficient between the target functional data and the corresponding clinical scale reduction based on the relevant data of the target individual at time points T0 and T1. The parameter iteration module updates the regulation parameters based on target function data, corresponding clinical scale reductions, Spearman correlation coefficients, induced quality data, and symptom trajectory data.

[0014] Therefore, the present invention employs the above-described transcranial magnetic stimulation target localization and quality monitoring method and system, which has the following beneficial effects: (1) A target quantification and localization method based on the activation effect of emotional situation and the individual percentile ratio is provided, which overcomes the shortcomings of existing target priority ranking indicators that fail to comprehensively consider the functional importance of brain regions, and realizes more clinically targeted individualized TMS target localization.

[0015] (2) A method for successive monitoring of situation-induced quality and statistical detection of habituation decay based on SAM valence score sequence is provided, which fills the gap in the existing scheme for lack of data monitoring mechanism for the decay of induced effect during multiple control processes, and maintains the induced quality throughout the process through video material dynamic scheduling algorithm.

[0016] (3) A symptom trajectory analysis method based on joint modeling of VAS symptom score sequence and SAM score sequence is provided to realize the sequential data tracking of symptom changes during the regulation process and the output of early warning data of no response.

[0017] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating an embodiment of the transcranial magnetic stimulation target localization and quality monitoring method of the present invention. Figure 2 This is a flowchart illustrating the target localization process according to an embodiment of the present invention. Figure 3 This is a system architecture diagram of an embodiment of the transcranial magnetic stimulation target localization and quality monitoring processing system of the present invention. Detailed Implementation

[0019] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0020] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0021] Example like Figure 1 As shown, the present invention provides a method and system for transcranial magnetic stimulation target localization and quality monitoring, comprising the following steps: S1. Based on the functional magnetic resonance imaging data of the brain under emotional situation induced state, extract the emotional situation-specific activation data of the target individual and the healthy reference group in each brain region.

[0022] The process of extracting emotion context-specific activation data includes: Obtain subjective valence ratings for the situational videos from no fewer than 20 reference raters, mark the continuous time period with a valence consensus ratio of no less than 60% as the effective emotional situation interval, and convolve it with the blood oxygen dynamics response function to generate an emotional situation regression factor; Substitute the emotional situation regression factor into the general linear model to calculate the activation contrast parameter between the emotional situation and the neutral situation. Based on the activation comparison parameters, activation parameter maps of the target individual and no fewer than 60 healthy reference individuals in each brain region of interest were extracted, and the activation distribution of the healthy group was constructed.

[0023] S2. Based on activation data from a healthy reference group, calculate the activation effect size and individual activation percentile for each brain region in the target individual, and generate individualized target coordinates by quantifying deviation indicators. For example... Figure 2 As shown, it specifically includes: S21. For the activation comparison parameters of the healthy control group in each brain region, calculate the effect size relative to the null hypothesis to construct a whole-brain effect size map. .

[0024] S22. Calculate the percentile of the target individual in the healthy reference distribution based on the activation value of each brain region. P The calculation formula is: ; in, For the target individual in the brain region j activation value, For the first k A healthy reference individual in the brain region j activation value, For the total number of healthy reference individuals, This is an indicator function.

[0025] S23, based on the calculation and P， Calculate each brain region of the target individual j The quantization deviation value Q is given by the formula: ; in, As a smoothing factor, it takes the value 0.1, which stably restricts the range of values ​​of the denominator to [0.1, 1.1].

[0026] S24. Map the Q values ​​of each brain region of the target individual to voxel space to generate a whole-brain quantitative deviation map Q-map. Extract the three-dimensional coordinates of the brain region with the highest Q value within the preset target search mask range, and use them as the coordinates of the individualized transcranial magnetic stimulation target.

[0027] S3. Collect self-assessment valence scores on the human scale after each scenario-induced intervention during the intervention process. Perform real-time effectiveness assessments of the scenario-induced quality and periodically statistically monitor the habituation decline trend of the induced effect. Specifically, this includes: S31. After each situational evoked process, collect the valence score of the Self-Assessment Manifestation Scale (SAM). , Use a scale of 1 to 5, where 1 represents completely positive and 5 represents completely negative; ≤2 is used as the effective threshold for situational induction. Output a valid flag when the value is ≤2. If the value is ≥3, an insufficient output flag is displayed and written to the induced quality log.

[0028] S32, using the regulatory sequence number during the regulation process as the independent variable. As the dependent variable, each cumulative N The Spearman rank correlation test was performed on the second adjustment to calculate the correlation coefficient. and significance Value, and calculate the previous period's average. Compared with the later mean The mean of the early stage is the mean of the first 1 / 5 of the sequence, and the mean of the later stage is the mean of the last 1 / 5 of the sequence.

[0029] S33. Set habitual alert trigger conditions. A habitual alert will be triggered when any one of the trigger conditions is met. The trigger condition is: Condition A. >0 and <0.05; Condition B, is and ≥3.

[0030] S4. Dynamically schedule and process the emotion-inducing video materials based on the statistical detection results. Specifically, when a habituation warning is triggered, video materials are retrieved from the material library according to the set priority order. The first priority is video clips that have not been used in the current treatment course; the second priority is video clips with the lowest usage frequency and the largest distance from recently used clips in the pre-built content similarity matrix. The scheduling record is written to the scheduling log, and the statistical detection in step S32 is re-executed in the next detection cycle after the material is changed.

[0031] S5. Collect depressive symptom scores from the Visual Analogue Scale (VAS) sequentially, and perform joint modeling with the valence data from the Self-Assessment Mannequin Scale (SAMS) to output symptom change trajectories and early warning data for non-response. Specifically, this includes: S51. Obtain the symptom score of the Visual Analogue Scale (VAS) before the start of each modulation sequence. , A continuous score from 0 to 100.

[0032] S52. Perform linear regression on the entire VAS score sequence to calculate the slope of symptom improvement. .

[0033] S53. SAM valence score according to Turn to a positive direction, with Spearman correlation coefficients were calculated for the sequences as a synergistic data indicator of context-induced quality and symptom changes.

[0034] S54. When the mean VAS value in the current period decreases by less than the preset threshold from the baseline, output early warning data for no response.

[0035] S6. Acquire functional magnetic resonance imaging (fMRI) data of the brain after the modulation is completed. Based on the individualized target coordinates in step S2, calculate the target functional data and the corresponding clinical scale reduction after the emotional situation-induced state, and calculate the Spearman correlation coefficient between the target functional data and the corresponding clinical scale reduction. Specifically, this includes: S61. Obtain T1 functional magnetic resonance imaging data of the brain at the second time point after the end of the modulation, and use the same general linear model parameters as in step S1 to extract the activation comparison parameter map of the target individual at T1 time point in each brain region of interest.

[0036] S62. Based on the individualized target brain regions obtained in step S2 The activation values ​​of the target individual in this brain region at time T1 are compared with the activation distribution matrix of the healthy group that has been constructed, and the percentile at time T1 is calculated using the same formula as in step S22. The formula is: .

[0037] S63, Based on the percentile of time T0 and percentile at time point T1 Calculate the percentage change of target function The formula is: ; in, >0 indicates that the functional percentile of the target point regresses towards the median (0.5) of the healthy reference group after regulation, that is, functional normalization occurs; A value <0 indicates that the target functional percentile has deviated further from the healthy reference level.

[0038] S64. Obtain the standardized clinical symptom scores of the target individual at time points T0 and T1, and calculate the reduced components of the clinical scale. The formula is: ; In the formula, This represents the standardized clinical symptom score at time point T0. This represents the standardized clinical symptom score at time point T1. A score >0 indicates a decrease in symptom score after regulation, meaning that symptoms have improved.

[0039] S65. After completing the full T0 to T1 data for multiple target individuals, use the percentage change in target function for each individual. Reduce the number of independent variables and clinical scales. Calculate the Spearman rank correlation coefficient for the dependent variable. The formula is: .

[0040] Among them, if <0 indicates that the direction of target function regression is consistent with the direction of symptom improvement, suggesting that regulation produced a clinically significant functional normalization effect under situational pre-activation conditions; if A value >0 indicates that the target function value has rebounded but the symptom improvement is not significant, and a "brain-behavior decoupling" data pattern has appeared. A decoupling warning mark is output, suggesting that the quality of situational induction or target localization parameters need to be further optimized.

[0041] S7. Archive the target function data, corresponding clinical scale reductions, Spearman correlation coefficients, induced quality data, and symptom trajectory data to the local database, and update the regulation parameters in the subsequent regulation process.

[0042] After the regulation is completed, each individual , , Values ​​and corresponding Q-map parameters ( Distribution, target brain regions P The value distribution), the full-process SAM valence score statistics (early and late mean, number of habituation detection triggers), and symptom trajectory data are uniformly written into the system database. The cumulative data is used to update the regulation parameters, including the optimal estimation range of the smoothing factor ε; and the reference distribution of the VAS improvement threshold for early non-response.

[0043] To verify the effectiveness of the method of this invention, data from 30 patients with mood disorders and 82 healthy controls were collected. Activation contrast parameters of each brain region under positive conditions were extracted from natural stimulation functional magnetic resonance imaging (fMRI) data, and the quantitative deviation value Q for each patient was calculated. The results showed a significant positive correlation between the maximum quantitative deviation value under positive conditions and HAMA anxiety symptoms. r =0.482, p =0.008, and showed a positive trend with HAMD depressive symptoms ( r =0.273, p =0.152). This indicates that the larger the target Q value of an individual patient in a positive situation, that is, the more intense the conflict that "areas that should be strongly activated in healthy individuals are weakly activated in patients," the more severe their anxiety and depression symptoms. This demonstrates that this method can effectively locate neuromodulation targets related to symptoms.

[0044] Z-scores and Q-scores capture different dimensions of clinical information: Z-scores reflect the statistical distance of an individual's activation from the healthy mean; Q-scores reflect the degree of functional conflict in individuals regarding insufficient activation of functionally important brain regions in healthy populations. These two types of indicators point to different clinical constructs and are not interchangeable in target localization applications. The Q-score method provides objective, quantitative evidence related to emotion regulation that Z-scores cannot independently provide.

[0045] Regarding the effectiveness of situation-induced quality monitoring and habituation detection, data from the aforementioned patient cohort undergoing situation-induced TMS modulation were used for validation. In the 244 scoring records throughout the course, the mean SAM valence in the situation-induced effective group was 2.46±0.61 and 2.89±0.43, respectively, both significantly lower than the neutral baseline of 3 (p<0.001 and p=0.009). In one group, the mean valence increased from 2.73±0.46 in the early stage of treatment to 3.00±0.43 in the later stage; the dual-condition habituation detection algorithm of this invention identified this decline and triggered an alert. In the other group, after dynamic scheduling of video materials, the mean valence remained significantly lower than the neutral level throughout the course (2.33±0.52 in the early stage and 2.66±0.64 in the later stage), demonstrating that the dynamic scheduling strategy of this invention can effectively delay the habituation process and maintain the overall induced quality.

[0046] like Figure 3 As shown, the present invention also provides a transcranial magnetic stimulation target localization and quality monitoring system, comprising: The data extraction module is used to extract emotional context-specific activation data of each brain region from the T0 and T1 time points of the target individual and the functional magnetic resonance imaging data of the healthy reference group. The target localization module calculates the activation effect of each brain region and the individual activation percentile at time T0 of the target individual based on the activation data of the healthy reference group, and generates individualized target coordinates by quantifying deviation indicators. The evoked quality monitoring module uses the self-assessment human valence scale to make real-time effectiveness judgments on the quality of situational evoked emotions during the regulation process, regularly conducts statistical detection on the habitual decay trend of the evoked effect, and dynamically schedules and processes the emotional evoked video materials based on the statistical detection results. The symptom tracking module jointly models the self-assessment human figure scale valence score and the visual analog scale depression symptom score data, and outputs the symptom change trajectory and early non-response warning data. The target data calculation module calculates the target functional data, the corresponding clinical scale reduction, and the Spearman correlation coefficient between the target functional data and the corresponding clinical scale reduction based on the relevant data of the target individual at time points T0 and T1. The parameter iteration module updates the regulation parameters based on target function data, corresponding clinical scale reductions, Spearman correlation coefficients, induced quality data, and symptom trajectory data.

[0047] The present invention also provides an electronic device, including a memory and a processor. The memory stores a computer program, and when the processor calls the computer program in the memory, it implements a transcranial magnetic stimulation target localization and quality monitoring processing method.

[0048] The present invention also provides a storage medium storing computer-executable instructions, which, when loaded and executed by a processor, implement a method for transcranial magnetic stimulation target localization and quality monitoring.

[0049] Therefore, this invention employs the aforementioned transcranial magnetic stimulation (TMS) target localization and quality monitoring method and system, overcoming the shortcomings of existing target priority ranking indicators that fail to comprehensively consider the functional importance of brain regions, thus achieving more clinically targeted individualized TMS target localization. Through successive monitoring of situational evoked quality and statistical detection of habituation decay based on SAM valence score sequences, it fills the gap in existing schemes regarding the lack of data monitoring mechanisms for evoked effect decay during multiple modulation processes, and maintains the overall evoked quality through a dynamic video material scheduling algorithm. Simultaneously, through symptom trajectory analysis based on joint modeling of VAS symptom score sequences and SAM score sequences, it achieves sequential data tracking of symptom changes during modulation and early warning data output for non-response.

[0050] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for transcranial magnetic stimulation target localization and quality monitoring, characterized in that, Includes the following steps: S1. Based on functional magnetic resonance imaging data of the brain under emotional situation induced state, extract emotional situation-specific activation data of target individuals and healthy reference groups in each brain region; S2. Determine the activation effect of each brain region based on data from a healthy reference group; construct the activation distribution of the healthy population, calculate the individual activation percentile of each brain region of the target individual, and generate individualized target coordinates by quantifying deviation indicators. S3. Collect self-assessment valence scores of the human scale after the situation is induced during the control process, make real-time judgment on the effectiveness of the situation inducement, and regularly conduct statistical detection on the habituation decay trend of the inducement effect. S4. Dynamically schedule and process the video materials that induce emotions based on statistical detection results; S5. Collect depression symptom scores from the Visual Analogue Scale one by one, and combine them with the valence data from the Self-Assessment Human Scale to perform joint modeling, and output symptom change trajectory and early non-response warning data. S6. Obtain brain functional magnetic resonance imaging data after the modulation is completed. Based on the individualized target coordinates in step S2, calculate the target functional data and the corresponding clinical scale reduction after the emotional situation induces the state. Calculate the Spearman correlation coefficient between the target functional data and the corresponding clinical scale reduction. S7. Archive the target function data, corresponding clinical scale reductions, Spearman correlation coefficients, induced quality data, and symptom trajectory data to the local database, and update the regulation parameters.

2. The method for transcranial magnetic stimulation target localization and quality monitoring according to claim 1, characterized in that, The process of extracting emotion context-specific activation data includes: Obtain subjective valence ratings for situational videos from several reference raters, mark continuous time periods with a valence consensus rate of no less than 60% as effective emotional situation intervals, and convolve these ratings with the blood oxygen dynamics response function to generate emotional situation regression factors. Substitute the emotional situation regression factor into the general linear model to calculate the activation contrast parameter between the emotional situation and the neutral situation. Based on the activation comparison parameters, activation parameter maps of the target individual and several healthy reference individuals in each brain region of interest were extracted to construct the activation distribution of the healthy group.

3. The method for transcranial magnetic stimulation target localization and quality monitoring according to claim 2, characterized in that, Step S2 specifically includes: S21. For the activation comparison parameters of the healthy control group in each brain region, calculate the effect size relative to the null hypothesis to construct a whole-brain effect size map. ; S22. Calculate the percentile of the target individual in the healthy reference distribution based on the activation value of each brain region. P The calculation formula is: ； in, For the target individual in the brain region j activation value, For the first k A healthy reference individual in the brain region j activation value, For the total number of healthy reference individuals, It is an indicator function; S23, based on the calculation and P， Calculate each brain region of the target individual j The quantization deviation value Q is given by the formula: ； in, It is a smoothing factor; S24. Map the Q values ​​of each brain region of the target individual to voxel space to generate a whole-brain quantitative deviation map Q-map. Extract the three-dimensional coordinates of the brain region with the highest Q value within the preset target search mask range, and use them as the coordinates of the individualized transcranial magnetic stimulation target.

4. The method for transcranial magnetic stimulation target localization and quality monitoring according to claim 3, characterized in that, Step S3 specifically includes: S31. After each situational evoked process, collect the valence score of the Self-Assessment Manifestation Scale (SAM). , Use a scale of 1 to 5, where 1 represents completely positive and 5 represents completely negative; ≤2 is used as the effective threshold for situational induction. Output a valid flag when the value is ≤2. If the value is ≥3, the output is marked as insufficient and written to the induced quality log; S32, using the regulatory sequence number during the regulation process as the independent variable. As the dependent variable, each cumulative N The Spearman rank correlation test was performed on the second adjustment to calculate the correlation coefficient. and significance Value, and calculate the previous period's average. Compared with the later mean ; S33. Set habitual alert trigger conditions. A habitual alert will be triggered when any one of the trigger conditions is met. The trigger condition is: Condition A. >0 and <0.05; Condition B, is and ≥3.

5. The method for transcranial magnetic stimulation target localization and quality monitoring according to claim 4, characterized in that: The mean of the first period is the mean of the first 1 / 5 of the sequence, and the mean of the second period is the mean of the last 1 / 5 of the sequence.

6. The method for transcranial magnetic stimulation target localization and quality monitoring according to claim 5, characterized in that, Step S4 specifically includes: after the habituation warning is triggered, video materials are retrieved from the material library according to the set priority order. The first priority is video clips that have not been used in the current treatment course; the second priority is video clips with the lowest usage frequency and the largest distance from recently used clips in the pre-built content similarity matrix; the scheduling record is written to the scheduling log, and the statistical detection of step S32 is re-executed in the next detection cycle after the material is replaced.

7. The method for transcranial magnetic stimulation target localization and quality monitoring according to claim 6, characterized in that, Step S5 is as follows: S51. Obtain the symptom score of the Visual Analogue Scale (VAS) before the start of each modulation sequence. , A continuous rating from 0 to 100; S52. Perform linear regression on the entire VAS score sequence to calculate the slope of symptom improvement. ; S53. SAM valence score according to Turn to a positive direction, with The Spearman correlation coefficient was calculated for the sequences as a synergistic data indicator of context-induced quality and symptom changes; S54. When the mean VAS value in the current period decreases by less than the preset threshold from the baseline, output early warning data for no response.

8. The method for transcranial magnetic stimulation target localization and quality monitoring according to claim 7, characterized in that, Step S6 specifically includes: S61. Obtain T1 functional magnetic resonance imaging data of the brain at the second time point after the end of the modulation, and use the same general linear model parameters as in step S1 to extract the activation comparison parameter map of the target individual at the T1 time point in each brain region of interest. S62. Based on the individualized target brain regions obtained in step S2 The activation values ​​of the target individual in this brain region at time T1 are compared with the activation distribution matrix of the healthy group that has been constructed, and the percentile at time T1 is calculated using the same formula as in step S22. ; S63, Based on the percentile of time T0 and percentile at time point T1 Calculate the percentage change of target function ; S64. Obtain the standardized clinical symptom scores of the target individual at time points T0 and T1, and calculate the reduced components of the clinical scale. ; S65. After completing the full T0 to T1 data for multiple target individuals, use the percentage change in target function for each individual. Reduce the number of independent variables and clinical scales. Calculate the Spearman rank correlation coefficient for t is the dependent variable.

9. A transcranial magnetic stimulation (TMS) target localization and quality monitoring system, used to execute a TMS target localization and quality monitoring method as described in any one of claims 1-8, characterized in that, include: The data extraction module is used to extract emotional context-specific activation data of each brain region from the T0 and T1 time points of the target individual and the functional magnetic resonance imaging data of the healthy reference group. The target localization module calculates the activation effect of each brain region and the individual activation percentile at time T0 of the target individual based on the activation data of the healthy reference group, and generates individualized target coordinates by quantifying deviation indicators. The evoked quality monitoring module uses the self-assessment human valence scale to make real-time effectiveness judgments on the quality of situational evoked emotions during the regulation process, regularly conducts statistical detection on the habitual decay trend of the evoked effect, and dynamically schedules and processes the emotional evoked video materials based on the statistical detection results. The symptom tracking module jointly models the self-assessment human figure scale valence score and the visual analog scale depression symptom score data, and outputs the symptom change trajectory and early non-response warning data. The target data calculation module calculates the target functional data, the corresponding clinical scale reduction, and the Spearman correlation coefficient between the target functional data and the corresponding clinical scale reduction based on the relevant data of the target individual at time points T0 and T1. The parameter iteration module updates the regulation parameters based on target function data, corresponding clinical scale reductions, Spearman correlation coefficients, induced quality data, and symptom trajectory data.

Images