Psychological health assessment method based on VR

Abstract

The invention relates to the technical field of psychological assessment, in particular to a psychological health assessment method based on VR (virtual reality), which comprises the following steps: based on virtual scene event nodes, analyzing event types, interaction states and priorities, screening adaptive nodes, adjusting an acquisition window, judging amplitude variation of brain waves, identifying segmentation moments, and combining extreme values and response intervals. And matching a high-risk reference, and classifying psychological risk indexes. According to the method, through dynamic mapping of user interaction behaviors and virtual situation events, whole-course closed-loop management of data streams is achieved, an acquisition window is adjusted in a linkage mode according to multi-source parameters, signal segmentation is defined in real time according to actual brain wave motion and a scene cognitive state, brain wave extreme values and behavior response time sequences are synchronously extracted, and coupling analysis is conducted on multi-dimensional features; dynamic grouping of psychological risks is jointly set based on multiple index intervals, timeliness guarantee is obtained for data in a high-frequency interaction scene, and an evaluation result can reflect continuous evolution of psychological states of a user in a complex environment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of psychological assessment, and in particular to a psychological health assessment method based on VR. BACKGROUND

[0002] The psychological assessment technology mainly includes the collection and analysis technology for the psychological health state of an individual, covering psychological state detection, psychological physiological signal collection, psychological scale assessment, and psychological state discrimination based on physiological and behavioral data. The traditional psychological health assessment method refers to collecting individual subjective information by filling in a psychological scale or artificial interview, or obtaining basic physiological signals through a simple physiological index collection instrument, and then analyzing them in combination with standard psychological assessment tools. The data collection and basic analysis of the psychological state of the evaluated person are usually completed by means of a psychological scale paper and pen questionnaire, a simple electroencephalogram collection instrument, and behavior observation and recording.

[0003] In the traditional psychological assessment method, the data mainly comes from static collection or subjective filling, the on-site environment is single, the interactive feedback is delayed, there is a lack of adaptability to the joint change of multiple types of parameters, the physiological and psychological state is collected in isolation, it is difficult to effectively reflect the dynamic correlation in the interactive process, the grouping judgment is based on a single time point or itemized evaluation, it is difficult to adapt to the complexity of the task and the variability of the situation, the risk discrimination is easy to form a static label, the information is not updated in time, there is a fragmentation between different evaluation dimensions, and the psychological risk under complex emotional response and scene change is difficult to timely stratify and reveal. SUMMARY

[0004] The purpose of the present application is to solve the problems existing in the prior art, and to provide a psychological health assessment method based on VR.

[0005] In order to achieve the above-mentioned purpose, the present application adopts the following technical scheme: a psychological health assessment method based on VR, comprising the following steps: S1: based on the event type trigger node in the virtual scene, analyzing whether the event type is consistent with the preset type, judging the user interaction state and the specification requirement, combining the event priority sorting, comparing each node item by item, screening the nodes that meet the adaptation standard, and obtaining an event adaptation judgment sequence; S2: based on the event adaptation judgment sequence, analyzing the event visual complexity, comparing the relevance of the interactive behavior click frequency and the event priority, calculating the adjustment start and end interval of each node collection window, and obtaining a collection window adjustment section; S3: based on the collection window adjustment section, judging the instantaneous amplitude change of the brain waves of each channel, comparing the average amplitude of the current interval and the previous interval, analyzing the brain wave trend in the difference interval, identifying the segmentation time, and obtaining a brain wave segmentation feature set; S4: Based on the set of electroencephalogram segment features, each segment of the electroencephalogram signal is judged, the cognitive task node is analyzed, the maximum and minimum electroencephalogram amplitudes are used to form an extreme value sequence, which is compared with the user behavior response time, the corresponding interval correlation is calculated, and an extreme value response feature interval is obtained.

[0006] The application improves that the event adaptation judgment sequence includes node validity, node influence degree, and node hierarchical information, the acquisition window adjustment section includes window division mode, window identification, and window related events, the set of electroencephalogram segment features includes segment type, segment stability, and segment range index, and the extreme value response feature interval includes response coupling coefficient, extreme value interval distribution, and behavior synchronization characteristics.

[0007] The application improves that the acquisition step of the event adaptation judgment sequence is specifically: S111: Based on the context event trigger node in the virtual scene, the type and preset standard of the context event in the virtual scene are analyzed, the matching relationship between each event type and the standard type is compared, whether the behavior parameter of the user in the interaction process is consistent with the specification is judged, the node meeting the standard is screened, and a node adaptation comparison sequence is obtained; S112: According to the node adaptation comparison sequence, the relative position difference of each node in the sorting process is calculated in combination with the event priority, the node content of focus is screened, a node screening and sorting sequence is obtained; S113: According to the node screening and sorting sequence, the validity, influence degree and hierarchical attribute of each node are compared, the node adaptation difference is obtained, and the adaptation of each node to the reference is judged, and an event adaptation judgment sequence is obtained.

[0008] The application improves that the acquisition step of the acquisition window adjustment section is specifically: S211: Based on the event adaptation judgment sequence, the color, shape, dynamic element and visual level feature in the scene are counted for the visual performance of each node, the visual load condition of the node in the scene is judged in combination with the node influence degree and hierarchical relationship, and an event visual load index is obtained; S212: Based on the event visual load index, the relationship between the interaction behavior click condition in the event node and the node criticality is analyzed, the response distribution of the interaction operation in the difference node is calculated, and the data segment with outstanding correlation is screened, and an interaction coupling distribution sequence is obtained; S213: According to the interaction coupling distribution sequence, the visual load and interaction distribution change of each node are combined, the interval that the acquisition window of each event node needs to be extended or reduced is judged, the start and end limits of the acquisition time period are optimized, and an acquisition window adjustment section is obtained.

[0009] The application improves that the acquisition step of the set of electroencephalogram segment features is specifically: S311: Based on the adjustment section of the acquisition window, the continuous change of the multi-channel brain wave instantaneous amplitude is analyzed, the amplitude change of each channel in the difference section is compared, the amplitude change sequence between the multi-channel data is constructed, the fluctuation between each section is judged, and amplitude change data is obtained; S312: Based on the amplitude change data, the difference of the continuous change section is analyzed, the change fluctuation sequence is obtained, the section range where the mutation occurs in the multi-channel data is judged, and the amplitude mutation section is obtained. S313: According to the amplitude mutation section, the multi-channel data thereof is screened, the amplitude extreme feature in the section is optimized, the segmented type and amplitude stability in each section are counted, and a brain electrical segmentation feature set is obtained.

[0010] The application improves that the extreme value response feature interval obtaining step is specifically: S411: Based on the brain electrical segmentation feature set, the amplitude feature and fluctuation performance of each segment are analyzed, the extreme value amplitude and the minimum amplitude of each cognitive branch task stage are compared, the amplitude difference situation is calculated, and the associated section is screened, and an extreme value amplitude difference section is obtained. S412: Based on the extreme value amplitude difference section, the time distribution of the user behavior response time is compared, the time sequence of the extreme value fluctuation and the response behavior is judged through the corresponding relationship of the task stage, the combination with a close corresponding relationship is identified, and an extreme value response coupling combination is obtained. S413: Based on the extreme value response coupling combination, the coupling performance of the amplitude change and the behavior interval in each group is analyzed, the key section of the association is screened, and an extreme value response feature interval is obtained.

[0011] The application improves that the step further comprises: S5: Based on the extreme value response feature interval, the joint state of the extreme value sequence interval length and the behavior response interval is analyzed, the matching of the high-risk section benchmark and the delay time benchmark is compared, the high-risk group is screened, the rest is summarized as a low-risk group, and a psychological risk classification index is obtained. The psychological risk classification index comprises a risk level, a grouping label and a risk interval boundary.

[0012] The application improves that the psychological risk classification index obtaining step is specifically: S511: Based on the extreme value response feature interval, the amplitude change of the extreme value sequence interval and the behavior response time interval of the user is analyzed, the change trend of each corresponding node is compared, and the difference performance under all nodes is combed, the data node with consistent difference characteristics is identified, and an extreme value behavior difference node is obtained. S512: Based on the extreme value behavior difference node, the matching of each node with the high-risk continuous section reference and the delay time reference on the feature distribution is judged, the node meeting the high-risk judgment standard is identified, and the high-risk distribution interval is obtained. S513: Based on the high-risk distribution interval, comparison is made with the risk classification basis, the distribution range and grouping condition are analyzed, the data meeting the high-risk grouping is screened, the remaining data is classified into the low-risk grouping, and the psychological risk classification index is obtained.

[0013] Compared with the prior art, the advantages and positive effects of the present application are that: In the present application, through dynamic mapping of user interaction behavior and virtual scenario events, full-cycle management of data flow is realized, the acquisition window is adjusted according to multi-source parameter linkage, the signal segmentation is defined in real time according to actual brain electrical wave fluctuation and scene cognitive state, the brain electrical extreme value and behavior response time sequence are extracted synchronously, multi-dimensional features are coupled for analysis, the dynamic grouping of psychological risk is based on the joint setting of multiple index intervals, the data obtains timeliness guarantee in high-frequency interaction scene, the evaluation result can reflect the continuous evolution of the user's psychological state in complex environment, the system supports asynchronous feedback and scene customization, and is suitable for different users and various psychological state discrimination scenes. BRIEF DESCRIPTION OF DRAWINGS

[0014] Figure 1 The main step flowchart of the present application is shown in the figure. Figure 2 The acquisition flowchart of the event adaptation judgment sequence in the present application is shown in the figure. Figure 3 The acquisition flowchart of the acquisition window adjustment section in the present application is shown in the figure. Figure 4 The acquisition flowchart of the brain electrical segmentation feature set in the present application is shown in the figure. Figure 5 The acquisition flowchart of the extreme value response feature interval in the present application is shown in the figure. Figure 6 The acquisition flowchart of the psychological risk classification index in the present application is shown in the figure. DETAILED DESCRIPTION

[0015] In order to make the purpose, technical scheme and advantages of the present application clearer, the present application is further described in detail below in combination with the figures and examples. It should be understood that the specific examples described herein are only used to explain the present application, and are not used to limit the present application.

[0016] In the description of the present application, it should be understood that the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application. In addition, in the description of the present application, the meaning of "a plurality of" is two or more, unless otherwise explicitly and specifically limited.

[0017] Embodiments Please refer to Figure 1 The present application provides a technical solution: a mental health assessment method based on VR, comprising the following steps: S1: Based on the situation event trigger node in the virtual scene, analyze whether the scene event type corresponds to the preset event type, judge whether the user interaction state parameter is consistent with the standard state requirement, combine the event priority, sort according to the event influence order, compare each node adaptability item by item, filter the nodes that meet the discrimination standard, and get the event adaptation judgment sequence; S2: Based on the event adaptation judgment sequence, analyze the event visual complexity, compare the relevance of the interaction behavior click frequency and the event priority, calculate the adjustment interval of the start and end time of each event node acquisition window, and get the acquisition window adjustment section; S3: Based on the acquisition window adjustment section, judge the continuous change of the instantaneous amplitude parameter of each channel brain wave, compare the statistical difference between the current interval average amplitude and the last interval average amplitude, analyze the amplitude variation trend of the brain wave in the difference interval, identify the moment when the amplitude change quantity meets the segmentation criterion, and get the brain wave segmentation feature set; S4: Based on the brain wave segmentation feature set, judge the interval range of each brain wave signal segmentation, analyze the cognitive branch task node, form an extreme value sequence by the maximum amplitude and the minimum amplitude of the brain wave, and compare with the user behavior response time, calculate the relevance of the corresponding interval, and get the extreme value response feature interval; S5: Based on the extreme value response feature interval, analyze the joint state of the extreme value sequence continuous interval length and the behavior response time interval, compare the matching degree of the joint state with the high risk continuous section benchmark and the delay time benchmark, filter the group belonging to the high risk standard, and the rest data is classified as low risk group, and get the psychological risk classification index.

[0018] The event adaptation determination sequence includes node validity, node influence degree, and node hierarchical information, and the collection window adjustment section includes window division mode, window identifier, and window associated event. The electroencephalogram segment feature set includes segment type, segment stability, and segment range index. The extreme value response feature interval includes response coupling coefficient, extreme value interval distribution, and behavior synchronization characteristic. The psychological risk classification index includes risk level, grouping label, and risk interval boundary.

[0019] In S1, the scene event type refers to a specific interactive event category in a VR virtual scene task. For example, a specific task or scenario trigger node in modules such as "psychological detection", "relaxation training", "attention training", etc. The preset event type refers to an event category standard defined in the background of a psychological assessment scheme or software system, i.e., a list of event types built-in as a reference for discrimination. The user interaction state parameter refers to the state quantitative data of the user when operating or participating in the VR scene. For example, whether a certain click, drag, selection, movement, gaze, voice response, etc. interactive action is completed, as well as the duration and frequency of the action. The standard state refers to the qualified / target state specified by the system for each event node. For example, "must complete a certain step of interactive action", "continuous attention time length meets the requirement", "operation process is not interrupted", etc. The node adaptability refers to the degree of agreement between the current state of the user and the system preset standard in the actual user participation process of each scenario event node. It is used to determine whether the user's performance at this node meets the standard. The discrimination standard refers to a set of rules for filtering nodes, generally set by combining event type, interaction state, priority, etc. according to logical judgment, and is the basis for determining whether the node "meets" the assessment conditions.

[0020] In S2, the event visual complexity refers to the richness of the visual elements (such as color, dynamic, element quantity, visual interference, etc.) of the virtual scene corresponding to the event node, which is used to reflect the load and interference level of the task in the visual sense. The interactive behavior refers to the specific operation completed by the user in the virtual scene, such as button click, object drag, gaze focus, head rotation, etc. The relevance refers to the correlation or coupling degree comparison between the frequency / pattern of interactive behavior and the priority of the event, to analyze whether the interactive activity level matches the importance of the event. The collection window refers to the specific time period (start and end time) for collecting electroencephalogram signal data, i.e., the time range for automatically positioning and collecting electroencephalogram before and after the event node occurs. The adjustment interval refers to the length and position of the collection window dynamically adjusted through the above analysis, which is the optimization and positioning result of the default collection window.

[0021] In S3, each channel refers to the signal path corresponding to different collection electrodes on the electroencephalogram device, such as multi-channel electroencephalogram wave collection in the forehead, parietal lobe, etc.; the current interval / previous interval refers to the electroencephalogram data time period divided according to the collection window, the "current interval" refers to the time period being analyzed, and the "previous interval" refers to the previous time period that has been analyzed and completed; the statistical difference refers to the numerical change between the electroencephalogram signals (such as mean, variance, extreme value, etc.) of the two intervals, which is used to judge the fluctuation of the signal; the amplitude variation trend refers to the rising, falling or stable trend of the electroencephalogram signal when it changes with time, reflecting the dynamic changes of the psychological or physiological state of the subject; the time point that meets the segmentation criterion refers to finding the time point where the signal changes significantly or mutates through amplitude variation trend judgment, which is used as the starting point of data segmentation.

[0022] In S4, the cognitive branch task node refers to a specific cognitive task (such as attention test, reaction time test, stress response) designed in the psychological course, corresponding to the key task steps or nodes in the scene; the extreme value sequence refers to the ordered set of maximum amplitude and minimum amplitude extracted in each segment of electroencephalogram signal, reflecting the main fluctuation interval of the signal in the segmentation interval; the user behavior response time refers to the time taken by the user to complete the specified operation (such as clicking, moving, etc.) in the virtual scene task, reflecting the cognitive and psychological response speed; the corresponding interval correlation refers to the synchronization or correlation analysis of the electroencephalogram extreme value fluctuation and the behavior response time in the same time period, testing the coupling degree between the changes of the two.

[0023] In S5, the joint state refers to the state jointly presented by the duration segment of the extreme value sequence and the interval of the behavior response time, which is used as the joint determination basis; the high-risk duration segment reference and the delay time reference refer to the electroencephalogram extreme value duration standard and the behavior response time delay standard used to define the "high-risk" criterion, which can be set by historical data and industry standards; the high-risk standard grouping refers to the user grouping recognized by the system as high-risk; the low-risk grouping refers to other groupings that do not meet the high-risk standard (generally ordinary or alert levels).

[0024] By adopting a data acquisition architecture based on the cooperation of the XR electroencephalogram feedback system kit and the VR all-in-one machine device, including: When participating in the psychological health assessment process, the user wears a portable electroencephalogram sensor (such as an XR brain-computer device integrated with a TGAM chip) to collect multi-channel electroencephalogram signals of the user in different brain regions (such as the forehead and parietal lobe) in real time; at the same time, the user wears and operates a VR all-in-one machine device (such as PICO Neo3), loads a pre-set virtual scene psychological course task in the device, and the scene contains multiple module tasks such as attention test, relaxation, stress simulation, and psychological venting.

[0025] A plurality of situational event trigger nodes are preset inside the VR scene, and the nodes are configured by scripts to define the conditions for triggering events at different time points or after the user completes different interactive actions (such as gaze, click, movement, and voice response); when the user interacts with the scene in the VR virtual environment, for example, gazes at a red ball that appears quickly for more than 1.5 seconds in the “attention focus test”, the node validity determination is triggered; according to the node determination result combined with the user operation data, the collection action of the electroencephalogram signal is triggered within the corresponding time period, and the XR brain machine continuously sends the user's brain multi-channel electrical signal data stream.

[0026] Based on the priority of the scene task node, the visual load, the user interaction density, and the dynamic analysis of the multi-channel electroencephalogram transient amplitude change, the collection window adjustment section is generated for different nodes, which is used to dynamically determine how long the electroencephalogram signal is recorded before and after the node occurs, for example, the default 10-second window is extended to 15 seconds, or shortened to 7 seconds, to ensure that the collected data accurately corresponds to the current psychological stimulation task.

[0027] The XR brain machine equipment worn by the user samples the electroencephalogram waves of different channels such as the forehead and the parietal lobe within the above-mentioned window to form multi-channel parallel raw time series data; after preliminary filtering processing on the XR brain machine side, the data is transmitted to the VR all-in-one machine through Bluetooth or WiFi wireless mode, or directly transmitted to the backend server for real-time recording.

[0028] The timestamps of all behavior operations (click, drag, gaze duration, voice input) performed by the user in the VR scene are recorded synchronously, and the user behavior time points are compared with the electroencephalogram value segment occurrence time in the subsequent process to achieve accurate alignment of the behavior data and the electroencephalogram data on the time axis, which can be used to derive psychological risk classification indicators.

[0029] Please refer to Figure 2 The acquisition steps of the event adaptation determination sequence are as follows: S111: Based on the situational event trigger nodes in the virtual scene, analyze the types and preset standards of situational events in the virtual scene, compare the matching relationship between each event type and the standard type, determine whether the behavior parameters of the user in the interaction process meet the specifications, filter the nodes that meet the standards, and obtain the node adaptation comparison sequence; The scene task script configuration list is called, the type, number, position coordinates and expected standard value of user behavior of each node record are read in turn, then the event type standard list preset in the psychological evaluation background is opened, the type of the node is compared with the types listed in the list one by one, if the type of the node exists in the same item in the list, it is determined that the node type matches and is valid, then the node is analyzed, otherwise it is directly skipped, then the behavior parameter value generated by the user during interaction is called for the valid node, such as reading the first line of sight focus delay time after the red sphere appears, the total gaze time and the operation completion identifier, comparing the first gaze delay of the user with the maximum delay allowed by the node, and if it is less than or equal to, it is determined to be qualified, comparing the total gaze time of the user with the minimum gaze time required by the node, and if it is greater than or equal to, it is also considered to be qualified, and then checking whether the node operation step is finally completed, if all three items are satisfied, the node record is recorded as a standard matching node, otherwise it is determined to be unqualified, for example, node 1 requires a maximum allowed delay of 1 second and a minimum gaze of 1.5 seconds, the user's data on this node is 0.8 seconds delay and 1.7 seconds gaze and completes the task, all of which are qualified, node 1 is selected for the next step, if the user's delay in node 2 reaches 1.3 seconds and exceeds the standard, it is directly determined to be unqualified, and the subsequent values are not compared, by reading, comparing and judging the behavior of all nodes in turn, the node numbers of all behavior data meeting the node standard are collected into the adaptability comparison sequence.

[0030] S112: According to the node adaptability comparison sequence, combined with the event priority, the relative position difference of each node in the sorting process is calculated, the key node content is screened, and the node screening and sorting sequence is obtained; The priority value configured in the scene task script of each node in the sequence is recorded in turn, the priority is usually an integer from 1 to 5, indicating the importance of the node in the psychological assessment process, and the priority of each node is compared with all other nodes in the sequence in turn by direct numerical comparison, if the priority of the current node is higher than that of another node, a higher count is accumulated once, after comparing each node, the number of times each node is higher than other nodes in the sequence is counted, for example, after comparing node 1 and node 4, node 1 accumulates a higher count once because the priority of node 1 is 4, which is higher than the priority of node 4, which is 3, and after comparing with node 6, node 6 is not increased because the priority of node 6 is 5, which is higher than the priority of node 1, through this process, node 1 accumulates two times higher than two nodes, and node 6 accumulates five times higher than the opponent in all comparisons, then the nodes are sorted in descending order according to the higher count to form a priority sorting list, and then compared with a screening reference value, for example, set the reference value to 2, this reference value is derived from the average value of the priority distribution of nodes in the past similar virtual scene psychological training, which is about 2.1, and is adjusted to 2 for screening more representative nodes, so that node 6 accumulates five times higher than the reference value and is included in the focus, node 1 accumulates two times equal to the reference value and is also included, and node 4 accumulates only one time and is excluded, the obtained focus node screening sorting sequence contains node 6 and node 1, which is used for subsequent acquisition window dynamic adjustment and electroencephalogram segmentation operation stage.

[0031] S113: According to the node screening sorting sequence, compare the effectiveness, influence degree and hierarchical attribute of each node, use the formula: ; Get node adaptation difference , and judge the adaptation of each node to the reference value to get the event adaptation judgment sequence, wherein, represents the total number of nodes, represents the effectiveness parameter of the th node, represents the hierarchical information parameter of the th node, represents the influence degree parameter of the th node, represents the corresponding parameter of the th node in the node adaptation comparison sequence, represents the corresponding parameter of the th node in the node screening sorting sequence.

[0032] The node adaptation difference refers to, in a virtual scene task, for each situational event node, comparing the node's effectiveness, hierarchical attribute and influence degree with the system preset standard after reflecting the difference between the current state of the node and the requirement, used to describe the deviation of a single node from the specification requirements in the whole process, and expressed as a data index that can be used for sorting, screening and further judgment.

[0033] The normalized node effectiveness parameter is obtained The node hierarchical information parameter is obtained , , The node influence degree parameter is obtained The parameter of the corresponding node in the node adaptation comparison sequence is called , , The parameter of the corresponding node in the node screening and sorting sequence is called The result of the formula calculation on the above parameters is as follows: , , The result of the formula calculation on the above parameters is as follows: The result of the formula calculation on the above parameters is as follows: , , The result of the formula calculation on the above parameters is as follows: The result of the formula calculation on the above parameters is as follows: , , The result of the formula calculation on the above parameters is as follows: The result of the formula calculation on the above parameters is as follows: The result of the formula calculation on the above parameters is as follows: ; The result of the formula calculation on the above parameters is as follows: ; The result of the formula calculation on the above parameters is as follows: ; The result of the formula calculation on the above parameters is as follows: .

[0034] The result of the formula calculation on the above parameters is as follows: The result of the formula calculation on the above parameters is as follows: The result of the formula calculation on the above parameters is as follows: ; The result of the formula calculation on the above parameters is as follows: ; The result of the formula calculation on the above parameters is as follows: ; The result of the formula calculation on the above parameters is as follows: ; After calculating the square root, it becomes .

[0035] After substituting into the formula, we get ; Combined with node adaptation differences The calculation results are used to determine whether the preset adaptation reference range is [missing information]. to Between, the result obtained in this calculation If the result falls within this range, it indicates that the analyzed node has strong adaptability after comprehensively considering multiple parameters such as effectiveness, hierarchical information, and degree of impact. This result is also used as a component of the event adaptation judgment sequence and is passed to subsequent stages. The formula integrates three highly structured upper-level indicators into a unified differential data structure and uses the square root normalization process to uniformly measure the underlying sorting and comparison items, so that the entire event adaptation judgment sequence can achieve consistency measurement across task scenarios and difference comparison between nodes, thus completing the acquisition of the event adaptation judgment sequence.

[0036] Please see Figure 3 The specific steps for obtaining the adjustment section of the acquisition window are as follows: S211: Based on the event adaptation judgment sequence, for the visual performance of each node, the color, shape, dynamic elements and visual hierarchy features in the scene are statistically analyzed, and combined with the degree of node influence and hierarchical relationship, the visual load of the node in the scene is judged to obtain the event visual load index. For each node, read its visual representation parameters pre-configured in the scene file, record the number of colors, the number of geometric shapes, whether it contains dynamic animation elements, and the hierarchical depth number of the node in the entire scene rendering sequence, then read the node 1 instance for the color number 6, the shape number 4, the dynamic element flag is yes, the hierarchical number is 5, compare the color number with the scene general reference interval [0, 3], [4, 6], [7, 10] in turn to determine that the node color number belongs to the second class, that is, the medium range, then compare the shape number with the interval [0, 2], [3, 5], [6, 8] to determine the medium level, the dynamic element flag is directly judged to be a binary state, because node 1 has dynamic elements, it is considered as 1, then check the hierarchical number with the preset interval [1, 3], [4, 6] to compare the high level segment, the value is summarized as the node visual parameter state of medium color, medium shape, dynamic, high level, then combine the node influence degree field, such as node 1 is set to influence degree 4 (interval is 1 to 5 integer value, 5 is the strongest) and hierarchical relationship is 5, perform value addition to obtain the initial value of the temporary visual load index as 19, then compare with the preset visual load division reference value 20, because 19 is slightly lower than the reference value 20, it is determined as a medium load node, if node 2 visual parameter statistics is color 9, shape 7, dynamic, level 5 and influence degree 5, the cumulative value is 26 which is greater than the reference value 20, it is determined as a high load node, through reading the number of node graphic elements, judging whether there is animation, comparing the hierarchical number, and then adding the preset influence value of the node and comparing with the reference value, the visual load level of each node is formed, after the calculation of all nodes, a group of node visual load indexes are gradually formed.

[0037] S212: Based on the event visual load index, analyze the relationship between the interaction behavior click situation and the node key in the event node, calculate the response distribution of the interaction operation in the difference node, and filter the data segments with outstanding relevance, to obtain the interaction coupling distribution sequence; The correlation analysis is performed on the number of clicks recorded in the interaction behavior log of each node and the criticality value of the node. First, the number of clicks is read from the record table one by one. For example, the total number of clicks of node 1 is 12, and the node criticality is configured as 4. The number of clicks is multiplied by the criticality value to get 48. Then, the number of clicks of node 2 is 22, and the criticality is 5. The two values are multiplied to get 110. Then, the node product values are compared one by one. According to the intervals [0, 50], [51, 100], and [101, 150], node 1 is classified as a medium node and node 2 is classified as a high node. Then, it is determined whether the click distribution of node 1 in the interaction log is concentrated in a certain time period of the task. For example, it is determined that the click distribution of node 1 is concentrated in the 10th to 20th second period, which accounts for 75% of the total clicks. It is determined that the click distribution of node 2 is evenly distributed in the 5th to 25th second period. Through the record of the interaction operation density of each node in the time period and the cumulative result of the criticality value, the preset threshold 80 is used to determine the prominent interaction density. If the node product value exceeds 80 and the main concentration distribution ratio exceeds 60%, it is considered as a node data segment with significant correlation. The product value of node 2 is 110, and the main time period interaction ratio is 65%, which meets the conditions and is filtered out. The product value of node 1 is 48, which does not exceed the threshold and is not filtered out. In this way, by reading the click record, performing direct numerical multiplication, comparing the distribution ratio, and comparing with the fixed threshold, all node interaction segments with significant correlation are obtained and combined to form an interaction coupling distribution sequence.

[0038] S213: According to the interaction coupling distribution sequence, the visual load and interaction distribution change of each node are combined to determine the interval in which the collection window of each event node needs to be extended or reduced, optimize the start and end limits of the collection time period, and obtain the collection window adjustment section. The visual load level and interaction distribution record of each node are read again. In this process, it is first determined whether the node is determined to be high visual load and has high interaction density in the interaction coupling sequence. If so, the node collection window time needs to be extended. The extension ratio is determined according to the fixed window adjustment coefficient 1.5, which is extended from the default 10-second collection window to 15 seconds. Then, if the visual load of the node is medium but the interaction coupling value is low, i.e., the product is less than 50, the collection window is reduced to 0.7 times the original, i.e., 7 seconds. In the judgment process, the node attributes are read one by one to determine whether the two conditions are met at the same time, i.e., high load and high coupling or medium-low load and low coupling. Then, the time adjustment of the collection window is performed. By reading two types of index values, performing double-condition judgment, and directly modifying the start and end times of the collection window, the time range of the collection window of all nodes is gradually extended or reduced. After completion, the adjusted start time and end time of all nodes are recorded in the collection window adjustment section result table.

[0039] Please refer to Figure 4 The acquisition steps of the electroencephalogram segment feature set are as follows: S311: Based on the adjustment of the collection window section, the continuous change of the multi-channel brain wave instantaneous amplitude is analyzed, the amplitude change of each channel in the difference section is compared, the amplitude change sequence between the multi-channel data is constructed, the fluctuation between each section is judged, and the amplitude change data is obtained; Firstly, the multi-channel brain electrical original data in each node corresponding to the collection window is read in sequence, each window is divided into multiple time periods, the brain electrical sampling value in each time period is arranged in millisecond sequence with millivolt as the unit, the signal sequence is extracted separately for different channels during reading, for example, the time sequence extracted by the forehead channel has 1200 points, and the time sequence extracted by the top of the head channel has 1180 points. Then, the instantaneous amplitude difference between adjacent two sampling points in each time sequence is calculated, the amplitude change list of each channel in the window is obtained by continuous calculation, the average amplitude change rate of each channel is formed by accumulating and averaging the instantaneous amplitude difference value, then the average amplitude change rates of the forehead channel and the top of the head channel are directly compared, if the average amplitude of the forehead channel is 1.3 millivolt and the average amplitude of the top of the head channel is 0.7 millivolt, it is recorded that the change of the forehead channel in the window is higher, according to the preset amplitude change interval [0, 0.5] as stable, [0.6, 1.0] as moderate fluctuation, [1.1, 1.5] as obvious fluctuation, 1.3 is judged as obvious fluctuation, after all channels are processed in sequence, a multi-channel amplitude change sequence is summarized in each window, and the sequence is compared with the average amplitude change rate of the same channel in the previous collection window, if the change amplitude increases by more than 0.5 millivolt, the window is marked as a difference section with obvious change in the data structure, for example, if the forehead channel of the second collection window of node 1 changes from 0.6 millivolt to 1.3 millivolt, it is recorded as a mutation section, by reading the original brain electrical sequence in each collection window in sequence, calculating the instantaneous amplitude difference point by point, and comparing the average amplitude change of the same channel across windows, the amplitude change data of each section of each node is obtained.

[0040] S312: Based on the amplitude change data, the difference of the continuous change section is analyzed, and the formula is: ; Obtaining change fluctuation sequence , judging the section range of the mutation in the multi-channel data, obtaining the amplitude mutation section, wherein, represents the total number of data in the amplitude change data, represents the data item in the th amplitude change data, represents the data item of the th channel instantaneous amplitude in the current section, represents the data item of the th channel instantaneous amplitude in the previous section, represents the normalization denominator item, the number of segments representing the mutated segments, a data item representing the start of the th mutated segment, a data item representing the end of the th mutated segment; The change fluctuation sequence refers to an ordered sequence formed by the amplitude changes of multi-channel electroencephalogram signals between adjacent segments after statistics, aggregation and calculation, used to describe the fluctuation intensity and change amplitude of electroencephalogram signals in each acquisition segment. It is a comprehensive measurement result of the fluctuation of multi-channel electroencephalogram signals between different acquisition windows, and is one of the key data bases in the analysis process.

[0041] The number of participating items constituting the difference is counted, and the total number of samples is obtained in combination with the channel data in the previous stage The amplitude change items participating in the calculation are extracted as follows: 、 、 、 After normalization, the following are obtained: 、 、 、 ; The current segment channel instantaneous amplitude data are as follows: 、 、 、 ; After normalization, the following are obtained: 、 、 、 ; The previous period channel instantaneous amplitude data are as follows: 、 、 、 ; After normalization, the following are obtained: 、 、 、 ; The sampling denominator normalization item is set as , which is a constant item constructed based on the maximum amplitude range and standard deviation of the sample; The mutated segments collected contain two channel time period ranges, the start data items are 、 , and after normalization, the following are obtained: 、 ; the end data items are , , normalized as , , therefore .

[0042] Substitute the above values into the original formula: ; The first part, in turn, expands the square term and the product term: ; ; ; ; The sum of the above four terms: ; Divide by and take the square root: ; The second part of the calculation: ; Combination operation: ; The result indicates that, in the current acquisition window, the aggregated fluctuation value of the multi-channel in the continuous sampling segment is , which can be further used in subsequent steps to extract and distinguish the amplitude mutation segment based on this value, and the segment is included in the data source term of subsequent stability and extreme value segmentation analysis. The larger the value, the stronger the channel aggregation fluctuation. The formula considers both the change intensity and the consistency of the direction by connecting the square and cross product, avoiding the situation of misjudging the aggregated fluctuation caused by single channel offset.

[0043] S313: According to the amplitude mutation segment, filter the multi-channel data, optimize the amplitude extreme features in the segment, and count the segment types and amplitude stability in each segment to obtain a set of electroencephalogram segmentation features; First, the multi-channel data segment with obvious amplitude change marked in the previous step is screened out, for example, the forehead and the top of the head of node 1 are recorded as mutation segments in the second acquisition window, and then the refined original sampling value sequence is re-read in the mutation segment, and the maximum value and the minimum value are counted, for example, the maximum amplitude of the forehead channel is 2.1 millivolts and the minimum amplitude is 0.4 millivolts, and the span is 1.7 millivolts, then according to the amplitude stability interval [0, 0.5] is stable, [0.6, 1.2] is medium fluctuation, [1.3, 2.0] is high fluctuation, 1.7 millivolts is judged as high fluctuation segment, and the length of continuous increasing or decreasing of the sampling data in the segment is counted, if the continuous increasing lasts for 150 milliseconds which is greater than the stable reference 100 milliseconds, it is recorded as a long increasing segment, and the type is marked as a strong increasing segment type, in this way, the maximum and minimum value difference operation is performed on the original data in each mutation segment, the length of continuous trend is counted, whether it exceeds the stable threshold is compared, and the segment label is assigned accordingly, finally, the segment type and amplitude stability statistical result of all segments of the node are summarized to form the brain electrical segmentation feature set, node 1 in the second window is recorded as a strong increasing high fluctuation segment, and node 1 in the third window is recorded as a medium fluctuation short increasing segment if the amplitude span is 0.9 millivolts, and the final result is arranged to generate the electroencephalogram segmentation feature set.

[0044] Please refer to Figure 5 The extreme value response feature interval acquisition step is specifically as follows: S411: Based on the electroencephalogram segmentation feature set, the amplitude characteristics and fluctuation performance of each segment are analyzed, the extreme value amplitude and the minimum amplitude of each cognitive branch task stage are compared, the amplitude difference is calculated, and the associated segment is screened, and the extreme value amplitude difference segment is obtained; The maximum amplitude value and the minimum amplitude value stored in each segment in the set are called in turn, and the corresponding segment start and end time is called, then the maximum value and the minimum value of the same segment are directly subtracted to obtain the amplitude difference value, then the difference value is divided according to the preset amplitude difference interval [0, 0.5], [0.6, 1.2], [1.3, 2.0], for example, when the maximum amplitude of segment 1 is 2.1 millivolt and the minimum amplitude is 0.7 millivolt, the subtraction is 1.4 millivolt, which falls into the [1.3, 2.0] interval and is judged as a significant difference segment, then the corresponding cognitive branch task phase identifier of the current segment is read, if segment 1 belongs to the 3rd phase of "attention test", the segment is recorded separately under the branch task, then the above calculation is repeated for all EEG segments, the maximum value and the minimum value of each segment are directly compared to obtain the amplitude difference size, and then compared with the reference value of each interval to determine the interval grade, if the maximum value of segment 2 is 1.0 millivolt and the minimum value is 0.8 millivolt, the subtraction is only 0.2 millivolt, which falls into the [0, 0.5] interval and is judged as a slight difference, by reading the amplitude value, performing direct subtraction and comparing with the preset difference interval, the amplitude difference level of all segments is determined, all segments in the medium and significant difference grades are screened in the segment registration table to form a preliminary list, then the task phase corresponding to the segment is checked, the segment numbers appearing multiple times with high amplitude difference in the same task phase are merged into the summary table of the phase, for example, if there are three segments with amplitude difference above 1.3 millivolt under "stress task 2nd phase", they are recorded together, by reading, calculating, grading and merging the segments in the phase in this way, the cognitive task phase and the segment information corresponding to all segments with large amplitude difference are registered in the summary table to form the extreme value amplitude difference segment.

[0045] S412: Based on the extreme value amplitude difference segment, compare with the time distribution of user behavior response time, through the corresponding relationship of the task phase, judge the time sequence of the extreme value fluctuation and the response behavior, identify the combination with close corresponding relationship, obtain the extreme value response coupling combination; First, the start time and end time of each extreme value difference section are read in sequence, and the operation record time points of the user under the same task in the behavior data log file are called to perform direct comparison to determine whether the user behavior response time falls within the time range of the electroencephalogram extreme value difference section, for example, if the extreme value section lasts from the 15th second to the 20th second and the user has a click operation at the 17th second, it is recorded as interval overlap, and the other behavior time distribution under the same task is further checked, if there are two or more user click or drag operations accumulated in the extreme value difference section, it is determined that there is a close correspondence relationship, then the identification information of the task stage is read, for example, “psychological release task phase 1”, and the combination is added to the coupling check result table, then each extreme value amplitude difference section and all user behavior times recorded in the task stage are repeatedly compared, for example, if the extreme value section is from the 22nd second to the 26th second and the user only has an operation at the 28th second, it is not considered as close correspondence, by repeatedly reading the extreme value section time value and the user behavior time value, comparing the time sequence and whether it is within the time range after each reading, and counting the number of user operations in each section, more than 2 times of the set threshold value is considered as a close correspondence section, the threshold value is 2, which is rounded down from the average behavior trigger times of 2.4 times in the analysis history of 20 similar psychological courses, used to determine that at least 2 times of overlap is required to consider that there is obvious coupling, after repeating all records in sequence, a coupling combination table is generated, and all extreme value response coupling combinations that are determined to have close correspondence in time sequence between extreme value fluctuation and user behavior response are summarized.

[0046] S413: Based on the extreme value response coupling combination, analyze the coupling performance of amplitude change and behavior interval between groups, using the formula: ; Screen the key sections related to association to obtain the extreme value response feature interval, wherein, represents the coupling interval association coefficient, represents the extreme value duration of the first group, represents the amplitude transformation rate of the first group, represents the behavior response interval of the first group, represents the amplitude difference of the first group, represents the total number of coupling data pairs.

[0047] The coupling interval correlation coefficient refers to a quantitative coefficient for reflecting the close degree of coupling between the extreme value change and the behavior response in the same task stage after calculating a plurality of data such as the extreme value duration, the amplitude transformation rate and the behavior response interval of each group of extreme value coupling behaviors, representing the overall correlation level between the extreme value fluctuation characteristics and the user behavior, which can be used to analyze and judge the synchronization, correlation or linkage strength between the EEG extreme value change and the user behavior response in different sections, and further provide a basis for subsequent discrimination such as screening and risk grading of the extreme value response characteristic interval.

[0048] The corresponding data is extracted from the selected task stage 3, task stage 4 and task stage 5, and the extreme value duration, amplitude transformation rate, behavior response interval and task stage amplitude difference of each combination are obtained. The original extreme value duration in task stage 3 is , and after normalization ; the amplitude transformation rate is synthesized from local fluctuation data, and the original calculation is , and after normalization ; the behavior response interval is originally , and after normalization ; the amplitude difference of this stage is originally , and after normalization , and the formula is calculated as: In task stage 3, it is expanded as: ; Similarly, in task stage 4, , , , , calculate: ; In task stage 5, , , , , calculate: ; Compare the above results with the reference coefficient , because the three groups of data are all less than the reference coefficient, no high coupling stage is screened out, and the judgment of forming the extreme value response characteristic interval needs to be further calculated in other task stages or more combinations. The formula introduces the difference between the extreme value duration and the amplitude transformation rate, and combines the behavior response interval and the amplitude difference to form a complex ratio, which helps to more objectively reflect the overall coupling characteristics of the EEG fluctuation and the behavior response, so as to realize the effective quantification of grouping screening in multi-stage tasks. ​​​

[0049] Referring to Figure 6 , the obtaining step of the psychological risk classification index is specifically: S511: Based on the extreme value response feature interval, analyze the extreme value sequence duration interval and the user's behavior response time interval, compare the change trend of each corresponding node, and sort out the difference performance under all nodes, identify the data node with consistent difference characteristics, and obtain the extreme value behavior difference node; The extreme value sequence duration recorded in the feature interval is read in sequence, the maximum amplitude maintenance time and the minimum amplitude maintenance time corresponding to each node are extracted, and then the user's behavior response time list in the virtual scene task under the same node is called. Through item-by-item comparison of the two groups of time periods, the difference value of each corresponding node is obtained by directly performing difference operation on the duration interval length of the extreme value sequence and the behavior response time interval. According to the preset difference judgment interval [0, 0.5 seconds] for slight difference, [0.6, 1.5 seconds] for moderate difference, and [1.6, 3 seconds] for obvious difference, the interval is judged. For example, node 1 has a maximum amplitude maintenance time of 2.1 seconds, and the user's click behavior interval is 1.0 seconds, with a difference of 1.1 seconds falling into the moderate difference zone, and node 2 has a maximum amplitude maintenance time of 3.5 seconds, and the user's response delay is 0.5 seconds, with a difference of 3.0 seconds falling into the obvious difference zone. The node difference level is recorded one by one, and the data trend of the same node in different stages is compared. If node 1 gets difference values of 0.9 seconds, 1.2 seconds and 1.3 seconds in three stages respectively, it is recorded as an increasing trend, and if node 2 gets difference values of 2.9 seconds, 3.1 seconds and 3.0 seconds in three stages, it is recorded as a stable high difference value trend. By reading the time value of each node, performing direct difference operation, and then judging the level according to the interval and comparing the trend in multiple stages, all nodes with the same increasing or stable high value trend are singled out to form a consistent difference node list, and the node is singled out and identified as an extreme value behavior difference node.

[0050] S512: Based on the extreme value behavior difference node, judge the matching of each node and the high risk duration interval reference and the delay time reference on the feature distribution, identify the nodes that meet the high risk judgment standard, and obtain the high risk distribution interval; The extreme value duration length and the user response time delay recorded by the node are read in sequence, and then compared with the static configuration values of the high-risk duration reference and the delay time reference, the high-risk duration reference is 2.5 seconds, and the high-risk delay reference is 1.2 seconds, if the extreme value duration length recorded by the node is greater than 2.5 seconds and the behavior response delay time is greater than 1.2 seconds, it is determined that the high-risk standard is met at the same time, for example, the extreme value length of node 2 is 3.0 seconds, and the delay time is 1.4 seconds, both of which exceed the corresponding reference, node 2 is marked as a high-risk node, and the extreme value length of node 1 is 1.3 seconds, which is less than the reference, so it is directly marked as a low-risk node, and the delay is not compared, by reading the two time values of the node and the reference value in sequence, and using the double condition to meet as the basis for determining whether it belongs to high risk, the risk classification state of each node is recorded by sequentially judging all nodes, and the node numbers of all nodes meeting the high-risk condition are included in the high-risk distribution interval list, forming the high-risk distribution interval corresponding to the node.

[0051] S513: Based on the high-risk distribution interval, compare with the risk classification basis, analyze its distribution range and grouping condition, filter the data meeting the high-risk grouping, and the rest is classified as low-risk grouping, to obtain the psychological risk classification index; The distribution start and end time of each node on the time axis is read, and then the grouping condition threshold recorded in the risk classification basis file is called, the threshold file sets the length of the accumulated high-risk interval greater than 5 seconds as the high-risk grouping critical value, and then the length of the accumulated high-risk section time of each node is read, for example, node 2 accumulates 6.3 seconds, which exceeds the 5-second reference and is directly classified into the high-risk grouping, while node 3 accumulates 4.2 seconds, which does not exceed the 5-second reference and is classified into the low-risk grouping, the accumulated time length of each node is compared with the fixed grouping reference value to determine, and the results are recorded in sequence, all nodes exceeding the reference value are finally included in the high-risk grouping and generate node identification, and the remaining nodes not exceeding the reference value are uniformly included in the low-risk grouping, to form the psychological risk classification index, and the high-risk nodes and low-risk nodes are recorded in the output structure according to the grouping label for subsequent data use.

[0052] The above is only a preferred embodiment of the present application, and does not limit the present application in other forms, any skilled person in the art can modify or change the above disclosed technical content to equivalent embodiments applied to other fields, but any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present application without departing from the technical solution content of the present application still belongs to the protection scope of the present application.

Claims

1. A VR-based mental health assessment method, characterized in that, The method comprises the following steps: S1: based on the context event trigger node in the virtual scene, analyze whether the event type is consistent with the preset type, judge the user interaction state and the specification requirement, combine the event priority ranking, compare each node item by item, screen the nodes that meet the adaptation standard, and obtain the event adaptation judgment sequence; S2: based on the event adaptation judgment sequence, analyze the event visual complexity, compare the correlation between the interaction behavior click frequency and the event priority, calculate the adjustment start and end interval of each node collection window, and obtain the collection window adjustment section; S3: based on the collection window adjustment section, judge the instantaneous amplitude change of the brain waves of each channel, compare the average amplitudes of the current interval and the previous interval, analyze the brain wave trend of the difference interval, identify the segmentation time, and obtain the brain wave segmentation feature set; S4: based on the brain wave segmentation feature set, judge each brain wave signal interval, analyze the cognitive task node, compare the maximum and minimum brain wave amplitudes to form an extreme value sequence, compare the corresponding interval correlation by comparing the user behavior response time, and obtain the extreme value response feature interval.

2. The VR-based mental health assessment method of claim 1, wherein, The event adaptation judgment sequence includes node effectiveness, node influence degree, and node hierarchical information. The collection window adjustment section includes window division method, window identifier, and window related event. The brain wave segmentation feature set includes segmentation type, segmentation stability, and segmentation range index. The extreme value response feature interval includes response coupling coefficient, extreme value interval distribution, and behavior synchronization characteristics.

3. The VR-based mental health assessment method of claim 1, wherein, The acquisition step of the event adaptation judgment sequence is specifically: S111: based on the context event trigger node in the virtual scene, analyze the type and preset standard of the context event in the virtual scene, compare the matching relationship between each event type and the standard type, judge whether the behavior parameters of the user in the interaction process meet the standard, screen the nodes that meet the standard, and obtain the node adaptation comparison sequence; S112: according to the node adaptation comparison sequence, combine the event priority, calculate the relative position difference of each node in the sorting process, screen the node content that needs to be focused on, and obtain the node screening and sorting sequence; S113: according to the node screening and sorting sequence, compare the effectiveness, influence degree and hierarchical attribute of each node, obtain the node adaptation difference, and judge the adaptation of each node to the reference, and obtain the event adaptation judgment sequence.

4. The VR-based mental health assessment method of claim 1, wherein, The acquisition step of the collection window adjustment section is specifically: S211: based on the event adaptation judgment sequence, for the visual performance of each node, count the color, shape, dynamic element and visual hierarchical features in the scene, and combine the node influence degree and hierarchical relationship to judge the visual load of the node in the scene, and obtain the event visual load index; S212: based on the event visual load index, analyze the relationship between the interaction behavior click and the node criticality in the event node, calculate the response distribution of the interaction operation in the difference node, and screen the data segments with prominent correlation, and obtain the interaction coupling distribution sequence; S213: According to the interaction coupling distribution sequence, the visual load of each node and the interaction distribution change are combined to determine the interval of the acquisition window of each event node that needs to be extended or reduced, optimize the start and end limits of the acquisition time period, and obtain the acquisition window adjustment section.

5. The VR-based mental health assessment method of claim 1, wherein, The acquisition step of the EEG segmentation feature set is specifically: S311: Based on the acquisition window adjustment section, analyze the continuous change of the instantaneous amplitude of multi-channel EEG, compare the amplitude change of each channel in the difference section, judge the fluctuation between each section by constructing the amplitude change sequence between multi-channel data, and obtain the amplitude change data; S312: Based on the amplitude change data, analyze the difference of the continuous change section, obtain the change fluctuation sequence, judge the section range where the mutation occurs in the multi-channel data, and obtain the amplitude mutation section; S313: According to the amplitude mutation section, filter the multi-channel data, optimize the extreme amplitude features in the section, count the segmentation type and amplitude stability in each section, and obtain the EEG segmentation feature set.

6. The VR-based mental health assessment method of claim 1, wherein, The acquisition step of the extreme value response feature interval is specifically: S411: Based on the EEG segmentation feature set, analyze the amplitude features and fluctuation performance of each segment, compare the extreme amplitude and minimum amplitude of each cognitive branch task stage, calculate the amplitude difference, and filter the associated section to obtain the extreme amplitude difference section; S412: Based on the extreme amplitude difference section, compare with the time distribution of user behavior response time, judge the time sequence of extreme fluctuation and response behavior through the corresponding relationship of task stage, identify the combination with close corresponding relationship, and obtain the extreme value response coupling combination; S413: Based on the extreme value response coupling combination, analyze the coupling performance of amplitude change and behavior interval in each group, filter the key section of association, and obtain the extreme value response feature interval.

7. The VR-based mental health assessment method of claim 1, wherein, The steps further include: S5: Based on the extreme value response feature interval, analyze the joint state of the length of the extreme value sequence interval and the behavior response interval, compare the matching with the high-risk section benchmark and the delay time benchmark, filter the high-risk group, and summarize the rest as a low-risk group, to obtain the psychological risk classification index; The psychological risk classification index includes risk level, grouping label, and risk interval boundary.

8. The VR-based mental health assessment method of claim 7, wherein, The acquisition step of the psychological risk classification index is specifically: S511: Based on the extreme value response feature interval, analyze the continuous interval of the extreme value sequence and the behavior response time interval of the user, compare the change trend of each corresponding node, and sort out the difference performance under all nodes, identify the data node with consistent difference characteristics, and obtain the extreme value behavior difference node; S512: Based on the extreme value behavior difference node, judge the matching of each node with the high-risk continuous section benchmark and the delay time benchmark in the feature distribution, identify the nodes that meet the high-risk judgment standard, and obtain the high-risk distribution interval; S513: Based on the high-risk distribution interval, compare with the risk classification basis, analyze the distribution range and grouping conditions, filter the data that meets the high-risk grouping, and the rest data is grouped as a low-risk group, to obtain the psychological risk classification index.

Images