An auxiliary analysis system based on neural activity synchronicity

Abstract

The application relates to an auxiliary analysis system based on neural activity synchrony, comprising: a cooperative task output module outputting a multi-person cooperative task; a task execution module adjusting each target object to synchronously participate in the cooperative task in the same time window based on the multi-person cooperative task; a data acquisition module acquiring brain function signals and behavior data of each target object in the execution process of the multi-person cooperative task; an interference control module acquiring interference information; a group neural activity synchrony calculation module used for quantitatively calculating neural activity synchrony indexes of any two target objects in the cooperative task process based on the acquired brain function signals, behavior data and interference information of each target object; obtaining a group synchrony index and a group synchrony network structure based on the neural activity synchrony indexes; and a risk monitoring module used for obtaining an auxiliary analysis result, so that the system can realize objective auxiliary identification, dynamic evaluation and closed-loop optimization of an ASD risk state.

Description

Technical Field

[0001] This application relates to the field of human-computer interaction technology, and in particular to an auxiliary analysis system based on the synchronization of neural activity. Background Technology

[0002] Autism Spectrum Disorder (ASD) is a neurodevelopmental disorder characterized by impaired social interaction, limited communication, and repetitive, stereotyped behaviors. Its pathogenesis is complex, and its manifestations exhibit significant heterogeneity and individual variability. Numerous studies have shown that the core issues of ASD are not only reflected in functional abnormalities in single brain regions, but are also closely related to impaired functional connectivity between multiple brain regions, dyssynergistic effects across brain networks, and abnormal synchronicity of neural activity in social interaction scenarios. Therefore, how to objectively, quantitatively, and dynamically assess the brain activity and interactions of multiple individuals in natural scenarios that closely resemble real-world social interactions has become an important topic in current brain science and clinical research. Existing methods for ASD risk identification and assessment mainly rely on behavioral scales, clinical interviews, and functional magnetic resonance imaging (fMRI). Among these, behavioral scales have limitations such as high subjectivity, reliance on assessor experience, and difficulty in capturing hidden neural abnormalities; while fMRI offers high spatial resolution, its large equipment size, high cost, and high requirements for subject cooperation make it difficult to apply to children and natural interaction scenarios, and also hinder simultaneous measurement and real-time control of multiple individuals. The aforementioned limitations, to some extent, restrict the early screening, risk stratification, and dynamic evaluation of intervention effects for ASD.

[0003] In recent years, functional near-infrared spectroscopy (fNIRS) has gradually become an important tool in cognitive neuroscience and clinical research due to its advantages such as non-invasiveness, strong resistance to motion artifacts, portable equipment, and suitability for children and real-world experiments. Hyperscanning technology, developed based on fNIRS, can simultaneously acquire brain blood oxygenation signals from multiple individuals, providing a new technical approach for studying social interaction, cooperative behavior, and group cognition. Existing research shows that in social interaction tasks such as cooperation, communication, or imitation, different individuals exhibit specific patterns of neural activity synchronicity, and individuals with ASD often show significant abnormalities in these synchronicity indicators.

[0004] Although numerous studies in recent years have begun to employ dual-person synchronous data acquisition and analysis techniques, related work has primarily focused on the social cognitive mechanisms behind interpersonal neural synchronization in scenarios such as emotion regulation and parent-child interaction. A comprehensive autism risk identification and assessment technology system based on abnormal neural synchronicity, geared towards multi-person natural interaction scenarios, has yet to be established. Particularly in multi-person scenarios, due to the increased number of participants and more complex group relationship structures, brain coupling patterns are no longer limited to simple pairwise synchronization but exhibit higher-dimensional, more dynamic, and more significantly individual-differentiated group collaboration characteristics. Existing methods generally lack specific adaptations for such complex interaction structures, making it difficult to achieve stable identification and refined stratification of ASD risk. Current research typically considers less critical issues such as experimental task adaptation for children's application scenarios, signal quality control during multi-person synchronous data acquisition, and multi-source behavioral information fusion, thus limiting the translational application of related technologies towards clinical auxiliary screening, risk warning, and intervention assessment. Therefore, there is an urgent need to propose an auxiliary analysis system based on the synchronization of neural activities suitable for multi-person natural interaction scenarios. By synchronously collecting and jointly analyzing the brain function signals and behavioral signals of multiple individuals, a quantitative index system that can characterize the synchronization of group neural activities, subgroup collaboration patterns and abnormal interaction characteristics can be constructed to achieve objective auxiliary identification, dynamic assessment and closed-loop optimization of ASD risk status. Summary of the Invention

[0005] Therefore, it is necessary to provide an auxiliary analysis system based on neural activity synchronization that can achieve objective auxiliary identification, dynamic assessment and closed-loop optimization of ASD risk status in order to address the above-mentioned technical problems.

[0006] This application provides an auxiliary analysis system based on the synchronicity of neural activity, the system comprising:

[0007] The collaborative task output module is used to output multi-person collaborative tasks;

[0008] The task execution module is used to adjust the participation of each target object in the cooperative task synchronously within the same time window based on the output of the multi-person cooperative task.

[0009] The data acquisition module is used to collect brain function signals and behavioral data of each target object during the execution of the multi-person collaborative task.

[0010] The interference control module is used to collect interference information during the execution of multi-person collaborative tasks;

[0011] The group neural activity synchronization calculation module is used to quantify the neural activity synchronization of any two target objects in the cooperative task process based on the collected brain function signals, behavioral data and interference information of each target object, to obtain a neural activity synchronization index; and to obtain a group synchronization index and a group synchronization network structure based on the neural activity synchronization index.

[0012] The risk monitoring module is used to obtain auxiliary analysis results based on the neural activity synchronization index and the group synchronization network structure.

[0013] In some alternative embodiments, the data acquisition module includes:

[0014] The brain function signal acquisition primary unit is used to synchronously initialize each acquisition device through a unified trigger signal or a unified sampling clock, and to acquire the brain function signals of each target object based on each acquisition device after synchronous initialization.

[0015] The behavior data acquisition unit is used to synchronously collect the task response time, operation sequence, and interaction event markers of each target object.

[0016] The timestamp alignment primary unit is used to uniformly timestamp the collected brain function signals and behavioral data to obtain a data stream, and to write the switching of each task stage of the multi-person collaborative task, the interaction event markers, and the system feedback nodes as event markers into the data stream.

[0017] In some alternative embodiments, the group neural activity synchronization calculation module includes:

[0018] The data correction primary unit is used to correct the collected brain function signals of each of the target objects based on the interference information to obtain the corrected brain function signals.

[0019] The first-level unit for dual-person synchronization calculation is used to quantify the synchronization of neural activities of any two target objects in a cooperative task based on the corrected brain function signals and behavioral data, and obtain a dual-person neural activity synchronization index.

[0020] A group synchronization fusion primary unit is used to fuse the synchronization indices of the two-person neural activities to obtain group synchronization indices;

[0021] The subgroup aggregation first-level unit is used to identify synchronized subgroups based on the dual-person neural activity synchronization index;

[0022] The first-level directional computing unit is used to determine the temporal sequence of neural activity changes among target objects in the synchronized subgroup based on the corrected brain function signals.

[0023] The group synchronization network represents the first-level unit, which is used to represent each target object with nodes, determine the edges between nodes based on the group synchronization index, determine the arrow direction of the edges based on the temporal relationship of the neural activity changes among the target objects, and generate the group synchronization network structure based on the nodes, edges and the arrow direction.

[0024] In some alternative embodiments, the data correction primary unit includes:

[0025] The secondary unit for determining data credibility weights is used to obtain data credibility weights based on the head motion index, signal quality index, and environmental interference index of the target object.

[0026] The secondary correction unit is used to obtain the corrected brain function signal based on the data confidence weight and the brain function signal of the target object.

[0027] In some optional embodiments, the subgroup aggregation first-level unit is specifically used to confirm that the two target objects constitute a synchronized subgroup when the synchronization index of the two-person neural activities corresponding to any two target objects is greater than or equal to the significant synchronization level, and to determine that the target object is an independent individual when the synchronization index of the two-person neural activities of any one target object and other target objects is less than the significant synchronization level.

[0028] In some alternative embodiments, the first-level unit for directionality calculation includes:

[0029] The delayed secondary unit is used to perform temporal alignment of the modified brain functional signals of different target objects in the same brain region, and to calculate the first follow-up change index value of the first target object following the second target object and the second follow-up change index value of the second target object following the first target object under different delay times.

[0030] The direction determination secondary unit is used to determine the first maximum neural activity synchronicity based on the first follow-change index value, determine the second maximum neural activity synchronicity based on the second follow-change index value, and determine the temporal sequence of neural activity changes of the first target object and the second target object based on the difference between the first maximum neural activity synchronicity and the second maximum neural activity synchronicity.

[0031] In some optional embodiments, the risk monitoring module includes:

[0032] The first risk assessment unit is used to obtain the first offset of each brain region of interest based on the group synchronicity index, the mean of synchronicity of the age-matched norm group, and the standard deviation of synchronicity of the age-matched norm group, and to determine the group synchronicity risk component based on the first offset and the weight corresponding to each brain region of interest.

[0033] The second risk assessment unit is used to determine the participation rate of the brain region of interest for each target object based on the synchronized subgroup, and to obtain the subgroup participation risk component based on the subgroup size and the average synchronization degree of the subgroup in which the target object participates.

[0034] The third risk assessment unit is used to obtain the directional consistency risk component based on the time sequence and the role of each target object.

[0035] The integrated first-level unit is used to obtain a comprehensive risk based on the group synchronization risk component, the subgroup participation risk component, and the direction consistency risk component, and to obtain auxiliary analysis results based on the comprehensive risk.

[0036] In some alternative embodiments, the system further includes:

[0037] An anomaly detection module is used to determine anomaly states based on the group synchronization network structure, the synchronization index of two-person neural activities, and the group synchronization index.

[0038] The annotation module is used to annotate the task stage, brain region, and target object corresponding to the abnormal state.

[0039] In some alternative embodiments, the system further includes:

[0040] The interaction target selection module is used to determine any one of the key individual nodes, key edges, or key subgroups as the interaction target based on the annotation results of the abnormal state.

[0041] An interaction parameter generation module is used to generate interaction parameters based on the interaction objective and the comprehensive risk.

[0042] An interaction module is used to stimulate the interaction target based on the interaction parameters;

[0043] The interaction effect determination module is used to compare the group synchronization network structure before and after the interaction to determine the interaction effect, and update the parameter generation strategy corresponding to the interaction parameter generation module based on the interaction effect.

[0044] In some alternative embodiments, the multi-person collaborative task is configured according to the type of cognitive processing and the degree of collaborative dependence, and the multi-person collaborative task includes one or more of the following: collaborative task based on action synchronization, collaborative task based on information sharing, collaborative task based on joint decision-making, and collaborative task based on creative expression.

[0045] Each of the target objects is divided into at least two teams, and each team includes a guide role and a collaborator role.

[0046] The aforementioned auxiliary analysis system based on neural activity synchronization includes: a cooperative task output module for outputting multi-person cooperative tasks; a task execution module for adjusting the synchronous participation of each target object in the cooperative task within the same time window based on the output multi-person cooperative task; a data acquisition module for acquiring brain function signals and behavioral data of each target object during the execution of the multi-person cooperative task; an interference control module for acquiring interference information during the execution of the multi-person cooperative task; a group neural activity synchronization calculation module for quantifying the neural activity synchronization of any two target objects during the cooperative task based on the acquired brain function signals, behavioral data, and interference information of each target object, obtaining a neural activity synchronization index; and obtaining a group synchronization index and a group synchronization network structure based on the neural activity synchronization index; and a risk monitoring module for obtaining auxiliary analysis results based on the neural activity synchronization index and the group synchronization network structure. This system can simultaneously acquire brain imaging data and multimodal behavioral signals of multiple individuals, and obtain a group synchronization index and a group synchronization network structure based on the neural activity synchronization assessment of two individuals, thereby achieving a quantitative assessment of the neural activity synchronization of multiple individuals. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 This is a block diagram of an auxiliary analysis system based on neural activity synchronization in one embodiment;

[0049] Figure 2 This is a schematic diagram of a group resource allocation and negotiation task in one embodiment;

[0050] Figure 3 This is a schematic diagram of near-infrared ultrasound scanning of multiple people in one embodiment;

[0051] Figure 4 This is a schematic diagram of the brain region of interest in one embodiment;

[0052] Figure 5 This is a schematic diagram of two-person synchronization calculation in one embodiment;

[0053] Figure 6 This is a schematic diagram of directional calculation in one embodiment. Detailed Implementation

[0054] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0055] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0056] Combination Figure 1 As shown, this application provides an auxiliary analysis system based on neural activity synchronization. The system includes: a cooperative task output module, a task execution module, a data acquisition module, a group neural activity synchronization calculation module, and a risk monitoring module.

[0057] The collaborative task output module is used to output multi-person collaborative tasks.

[0058] The collaborative task output module provides multiple children (at least three) with collaborative tasks appropriate to their cognitive development levels to induce neural activity patterns related to social interaction, joint attention, group collaborative processing, role division, and group decision-making. These collaborative tasks include, but are not limited to: collaborative building / puzzle work, collaborative drawing, co-creating stories, rhythmic synchronization, collaborative problem-solving, role-playing, group resource allocation and negotiation, collaborative mathematical cognition, group singing, and ensemble performance. Through task rule constraints, role assignments (such as organizer / executor / observer / coordinator), and interactive process design, multiple children are guided to engage in sustained social interaction under a common goal, forming an analyzable group interaction structure. Figure 1 The children are categorized as A, B, C, D, E, F, and G. A, B, C, and D are grouped into one group, and F and E are grouped into another.

[0059] The task execution module is used to adjust the synchronous participation of each target object in the collaborative task within the same time window based on the output of the multi-person collaborative task.

[0060] The task execution module is used to organize and coordinate multiple children to participate in collaborative tasks simultaneously within the same time window, and to manage the global timing and role / stage orchestration of the task execution process. This module includes at least three units: a global time anchoring unit to unify the start, stage switching, key event triggering, and end times of the task; a stage orchestration unit to divide the multi-person task into baseline, interaction, collaboration, conflict / negotiation, and review stages; and a role management unit to record / adjust the role status of each child in the task stage, providing a basis for subsequent analysis of synchronization, directionality, and leader-follower relationships.

[0061] Among them, multi-person collaborative tasks are cognitive tasks completed by multiple children together. These tasks are configured according to the type of cognitive processing and the degree of cooperation dependence to induce changes in children's neural activity during social interaction and collaborative goal completion. Multi-person collaborative tasks include one or more of the following: collaborative tasks based on action synchronization, collaborative tasks based on information sharing, collaborative tasks based on joint decision-making, and collaborative tasks based on creative expression. Each target group is divided into at least two teams, and each team includes a facilitator role and a collaborator role.

[0062] Collaborative cognitive tasks encompass various forms of cooperation, including but not limited to those based on action synchronization, information sharing, joint decision-making, and creative expression, thus covering a wide range of social collaboration scenarios. For example... Figure 2 As shown, Figure 2This is a schematic diagram of a group resource allocation and negotiation task in one embodiment, taking the task as an example. In this task, multiple children are organized into an interactive group, with each group consisting of six children, and randomly divided into two teams, each containing three children. Before the task begins, each team conducts internal negotiation through a pre-set brief communication session, and selects one child as the facilitator for that subgroup through voting or system-designated methods, with the remaining children acting as collaborators, thus forming a facilitator-collaborator role structure within the group. This role structure remains stable during subsequent task execution, used to induce directional social interaction behaviors. The task is presented in a story-like context, including various task scenarios, such as guardianship and exploration scenarios. In these task scenarios, each team accumulates task resources through collaboration and discussion, and receives group rewards based on the scenario objectives. By switching between the different task scenarios, children are guided to generate diverse decision-making behaviors and interaction strategies during cooperation, thereby enhancing the distinguishability of the synchronous characteristics of neural activity. The task operates in a multi-round structure, with each round including a decision-making phase, a waiting phase, and a feedback phase. During the decision-making phase, each child independently makes resource allocations or behavioral choices based on task prompts provided by the system within a unified time window. During the waiting phase, the system calculates and integrates the results of this round of tasks. In the feedback phase, the system presents all children with the involvement of each role within the group, the overall group performance, and the results of this round of tasks. By clearly presenting the behavioral differences between the guiding and collaborating roles during the feedback phase, the system guides collaborating roles to follow or adjust the guiding role's behavior in subsequent rounds, thereby gradually forming a quantifiable alignment relationship between guiding and collaborating behaviors at the group level.

[0063] in Figure 2 The text provides a diagram of a task, which is "Protect the castle from being destroyed by the storm. When the game ends, you will receive a reward based on the number of 'five-pointed stars' you have." The first round (1) is "You have 20 'five-pointed stars'. To resist the storm's attack, your team needs 15 'five-pointed stars'. You can contribute your stars ( )". The first round (2) is "Wait for others". The first round (3) is "You have 8 'five-pointed stars', the captain has 5 'five-pointed stars', and the team members have 6 'five-pointed stars'". The first round (4) is "Success". The first round (5) is "Everyone discusses". The second round (1) is "You have 20 'five-pointed stars'. To resist the storm's attack, your team needs 10 'five-pointed stars'. You can contribute your stars ( )". The task continues until the end of the task.

[0064] The data acquisition module is used to collect brain function signals and behavioral data of each target object during the execution of multi-person collaborative tasks.

[0065] The data acquisition module is used to simultaneously collect brain function signals and related behavioral data from multiple children during collaborative tasks. The brain function signals include functional near-infrared brain imaging (fIR) superscan signals; the behavioral data includes, but is not limited to: task reaction time, operation sequences, interactive event markers, voice communication characteristics, eye movement / fixation events, posture / motion characteristics, touch / object manipulation records, etc. This module supports simultaneous triggering by multiple individuals and uses a unified sampling clock to ensure temporal consistency of data across individuals.

[0066] The interference control module is used to collect interference information during the execution of multi-person collaborative tasks.

[0067] The interference control module is used to detect and control motion artifacts, environmental noise, and inter-individual interference factors in multi-person natural interaction scenarios. This includes, but is not limited to, head movement monitoring, contact quality monitoring, ambient light interference assessment, and behavioral event conflict detection. Interference information is used as the basis for subsequent synchronization calculations to eliminate or adjust weights, thereby improving the reliability of multi-person synchronization indicators.

[0068] In some optional embodiments, the system further includes a data processing and alignment module: this module is used to preprocess, standardize, and perform cross-individual alignment processing on brain functional signals and behavioral data from multiple children, outputting normalized input suitable for synchronous computation. The processing includes, but is not limited to: signal denoising, artifact correction, motion regression, time synchronization, task segmentation, baseline correction, multimodal event alignment, feature extraction, and cross-individual feature normalization.

[0069] The group neural activity synchronization calculation module is used to quantify the neural activity synchronization of any two target objects in the cooperative task process based on the collected brain function signals, behavioral data and interference information of each target object, and obtain the neural activity synchronization index; and obtain the group synchronization index and the group synchronization network structure based on the neural activity synchronization index.

[0070] The group neural activity synchronization calculation module is used to quantify the neural activity synchronization of multiple children during a cooperative task based on preprocessed multi-person brain functional signals, and to construct a synchronization structure index reflecting the degree of group collaboration. This module includes at least the following calculation units: a two-person synchronization calculation unit, which calculates the synchronization of any two children at the corresponding brain region / network level; a group synchronization fusion unit, which merges two-person synchronization into a group synchronization index; a subgroup aggregation unit, which identifies synchronized subgroups formed during interaction; a directionality calculation unit, which, combined with role arrangement, calculates synchronization features with directional properties such as leader-follower and active-passive; and a group synchronization network representation unit, which uses nodes to represent individuals and related brain regions, edges to represent synchronization strength, and arrows to represent directionality combined with role attributes, used to characterize the neural collaboration structure of multi-person interaction.

[0071] The risk monitoring module is used to obtain auxiliary analysis results based on neural activity synchronization indicators and group synchronization network structure.

[0072] This risk monitoring module is used to assess and monitor children's ASD-related risk status based on group synchronization network characteristics and multimodal behavioral characteristics. The module's output includes at least: an individual risk score, reflecting the degree of abnormal social processing in group interactions; a group structure risk index, reflecting the existence of abnormal structures in group interactions (such as isolated nodes, weakly connected nodes, abnormal subgroups, etc.); and risk stratification results, classifying individuals into different risk levels and providing explanations of their interaction roles and abnormal patterns within the group.

[0073] The aforementioned auxiliary analysis system based on neural activity synchronization includes: a cooperative task output module for outputting multi-person cooperative tasks; a task execution module for adjusting the synchronous participation of each target object in the cooperative task within the same time window based on the output multi-person cooperative task; a data acquisition module for acquiring brain function signals and behavioral data of each target object during the execution of the multi-person cooperative task; an interference control module for acquiring interference information during the execution of the multi-person cooperative task; a group neural activity synchronization calculation module for quantifying the neural activity synchronization of any two target objects during the cooperative task based on the acquired brain function signals, behavioral data, and interference information of each target object, obtaining a neural activity synchronization index; and obtaining a group synchronization index and a group synchronization network structure based on the neural activity synchronization index; and a risk monitoring module for obtaining auxiliary analysis results based on the neural activity synchronization index and the group synchronization network structure. This system can simultaneously acquire brain imaging data and multimodal behavioral signals of multiple individuals, and obtain a group synchronization index and a group synchronization network structure based on the neural activity synchronization assessment of two individuals, thereby achieving a quantitative assessment of the neural activity synchronization of multiple individuals.

[0074] In some optional embodiments, the data acquisition module includes: a brain function signal acquisition primary unit, a behavioral data acquisition primary unit, and a timestamp alignment primary unit.

[0075] The primary unit for brain function signal acquisition is used to synchronously initialize each acquisition device through a unified trigger signal or a unified sampling clock, and to acquire brain function signals of each target object based on the synchronously initialized acquisition devices.

[0076] Before the start of a multi-person collaborative task, uniform sampling parameters are configured for all brain function and behavior acquisition devices used by participating children. Each acquisition device is then synchronously initialized using a unified trigger signal or a unified sampling clock to establish a consistent time reference across individuals. Based on this unified time reference, brain function signals from multiple children are simultaneously acquired, including functional near-infrared brain imaging signals, and the acquired brain function signals are continuously recorded. (See Formula 1 and...) Figure 3 As shown, where Indicates the first Children at a unified time benchmark The functional near-infrared brain imaging signals contained in this data include One channel, Indicates the first Children at a unified time benchmark The next Functional near-infrared brain imaging signals from multiple channels.

[0077]

[0078] The primary unit for behavioral data acquisition is used to synchronously collect the task response time, operation sequence, and interaction event markers of each target object.

[0079] While collecting brain function signals, behavioral data related to task execution is collected simultaneously. The behavioral data includes at least task reaction time, operation sequence and interactive event markers, and may further include voice communication features, eye movement or gaze events, posture or movement features and touch or object manipulation records.

[0080] The timestamp alignment primary unit is used to uniformly timestamp the collected brain function signals and behavioral data to obtain a data stream, and writes the switching of each task stage of the multi-person collaborative task, the marking of interactive events, and the system feedback nodes as event markers into the data stream.

[0081] The collected brain function signals and behavioral data are uniformly timestamped, and task phase transitions, key interaction events, and system feedback nodes are written into the data stream as event markers to achieve time alignment between brain function signals and multimodal behavioral data. As shown in Formula 2, where... Indicates the first Children at a unified time benchmark The following dataset contains: the corrected first... Children at a unified time benchmark Functional near-infrared brain imaging signals ;No. Children at a unified time benchmark behavioral feature vectors ;No. Children at a unified time benchmark Event tag vectors for each stage below .

[0082]

[0083] In some optional embodiments, the swarm neural activity synchronization computation module includes: a data correction primary unit, a two-person synchronization computation primary unit, a swarm synchronization fusion primary unit, a subgroup aggregation primary unit, a directionality computation primary unit, and a swarm synchronization network representation primary unit, wherein:

[0084] The data correction primary unit is used to correct the brain function signals of each target object based on interference information, so as to obtain the corrected brain function signals.

[0085] The optional data correction primary unit includes: a data credibility weight determination secondary unit and a correction secondary unit.

[0086] The secondary unit for determining data credibility weights is used to obtain data credibility weights based on the head motion index, signal quality index, and environmental interference index of the target object.

[0087] The modified second-level unit is used to obtain the modified brain function signal based on the data confidence weight and the brain function signal of the target object.

[0088] This process involves detecting head movements or large-amplitude body movements in children based on motion sensor data or signal change characteristics, and marking time periods that may introduce motion artifacts. The status of the near-infrared probe is monitored during brain function acquisition to identify poor contact, signal attenuation, or abnormal drift, generating corresponding quality labeling information. External factors such as changes in illumination or background noise in the task environment are evaluated to determine their impact on brain function signal quality. In multi-person natural interaction scenarios, simultaneous high-intensity interaction events, abnormal behavioral states, or task interruption events are detected, and the corresponding time, location, and involved individuals are recorded. Information on detected motion artifacts, abnormal contact quality, environmental interference, and behavioral event conflicts is summarized and associated with the original data in the form of event annotations or quality indicators. Based on the interference annotation results, corresponding data correction strategies are generated, including removing interfered data segments, reducing their weight, or marking them as low-confidence data, and these correction strategies are passed to subsequent neural activity synchronization calculation steps. As shown in Formulas 3 and 4, Indicates the first Children at a unified time benchmark The data credibility weight is as follows: Represents the normalized i-th Children at a unified time benchmark The following head motion indicators, Represents the normalized i-th Children at a unified time benchmark The probe signal quality index is as follows: Indicates the normalized result on a unified time base. The environmental interference indicators below These are used to adjust the weight of a specific quality indicator. Indicates the corrected number Children at a unified time benchmark Functional near-infrared brain imaging signals below, Indicates the first Children at a unified time benchmark Functional near-infrared brain imaging signals.

[0089]

[0090]

[0091]

[0092] The dual-person synchronization calculation first-level unit is used to quantify the synchronization of neural activities of any two target objects during a cooperative task based on the corrected brain function signals and behavioral data, and obtain the dual-person neural activity synchronization index.

[0093] After completing the synchronous acquisition of brain functional signals and multimodal behavioral data, as well as interference detection and labeling, preprocessing steps were performed on the obtained raw data to improve data quality and provide stable and comparable input data for subsequent neural activity synchronization calculations. Specifically, the preprocessing process included denoising and artifact correction of brain functional signals. Data segments affected by motion artifacts, abnormal contact quality, or environmental interference were removed or downweighted to reduce the impact of non-task-related factors on brain functional signals. Simultaneously, the filtered light intensity signals were denoised, including suppression of high-frequency noise and slow drift. High-frequency components unrelated to hemodynamic response were filtered out to reduce the influence of physiological noise such as heart rate and respiration; low-frequency drift was corrected to eliminate baseline shifts caused by probe contact changes or environmental factors during long-term acquisition. After denoising, the light intensity signals were converted into a signal representation reflecting changes in blood oxygen dynamics. By converting light intensity changes at different wavelengths, the changes in oxyhemoglobin and deoxyhemoglobin related to neural activity were obtained, and these were used as the main input for subsequent analysis. In addition, to reduce the impact of individual differences on synchronicity analysis, the blood oxygen dynamics signals of different children were normalized or standardized so that they could be compared on a uniform scale, thereby ensuring that subsequent synchronicity indicators mainly reflect the relative consistency of neural activity changes during multi-person interaction.

[0094] The standard brain space was divided into several brain regions using a pre-defined brain region segmentation atlas (Destrieux), and several regions of interest were identified from these regions. (See Equation 5 and...) Figure 4 As shown, Figure 4 The region of interest (ROI) is shown as region of interest 1, region of interest 2, region of interest 3, region of interest 4, and region of interest 5. In practical applications, This represents a collection of brain regions of interest. This represents a collection of voxels belonging to the dorsolateral prefrontal cortex. This represents the set of voxels belonging to the superior frontal gyrus. This represents the set of voxels belonging to the inferior frontal gyrus. This represents the set of voxels belonging to the temporal pole. This represents the set of voxels belonging to the inferior temporal gyrus. This represents the set of voxels belonging to the middle temporal gyrus. This represents the set of voxels belonging to the temporoparietal junction.

[0095]

[0096] Based on the spatial location information of the functional near-infrared probe on the scalp, and combined with the head model or standard spatial registration relationship, each near-infrared measurement channel is mapped to its corresponding location in the standard brain space. After establishing the correspondence between channels and brain regions, the blood oxygen dynamics signals from multiple near-infrared measurement channels belonging to the same brain region of interest are converged. This convergence processing includes weighted averaging or statistical summarization of changes in oxyhemoglobin or deoxyhemoglobin within the brain region to obtain a brain region-level time-series signal characterizing the overall neural activity changes in that brain region. Through this process, the conversion from channel-level signals to brain region-level signals is achieved.

[0097] Based on preprocessed multi-person brain functional signals, this study quantifies the neural activity synchronicity of multiple children during collaborative tasks and constructs a synchronicity structural index reflecting the degree of group collaboration to characterize neural collaboration patterns in multi-person natural interaction scenarios. First, at the voxel level of brain regions, the neural activity synchronicity between any two children is calculated. Specifically, this unit analyzes the temporal changes of two children in corresponding brain regions or network nodes based on brain region-level blood oxygenation time series mapped to a set of brain regions of interest, obtaining a two-person synchronicity index reflecting the degree of neural activity collaboration during the collaborative task. The calculation of synchronicity characteristics between different individual pairs provides a foundation for subsequent group-level fusion and structural analysis. Figure 5 And as shown in formulas 6 and 7, Representing the children and the The number of children in the The degree of synchronization on individual elements for and covariance, for variance Represents the corrected number The number of children in the Brain function signals on individual pixels Represents the corrected number The number of children in the Brain function signals on individual pixels To The results after performing Fisher's Z-transform are used to facilitate subsequent summarization and comparison.

[0098]

[0099]

[0100] The group synchronization fusion first-level unit is used to fuse the synchronization indices of two individuals' neural activities to obtain the group synchronization index.

[0101] This process involves fusing multiple sets of pair synchronization results to construct a group synchronization index reflecting the overall level of group collaboration. By statistically summarizing or weighting the synchronization characteristics of all pairs of individuals within the group, a comprehensive index describing the overall neural collaboration level of multiple individuals is formed, characterizing the overall coordination state of the group during cooperative tasks. As shown in Formula 8, For the first Group synchronicity on individual factors For the specified group size, For the standardized first children and the The number of children in the Synchronization level at the voxel level. Based on the voxel calculation method, the group synchronicity of voxels contained in all brain regions of interest is summarized to obtain a group synchronicity index. , used to represent the first Synchronization of brain regions of interest.

[0102]

[0103] The subgroup aggregation first-level unit is used to identify synchronized subgroups based on the synchronization index of two-person neural activity.

[0104] Based on the results of two-person synchronization calculations, synchronized subgroups formed during multi-person interactions are identified. Specifically, by analyzing the distribution of neural activity synchronization between any pair of individuals within the group, a subset of individuals with relatively high synchronization levels is automatically identified. Combined with the corresponding brain regions of interest, their collaborative activity patterns at the brain region level are extracted, thereby characterizing the core collaborative subgroups, peripheral subgroups, or isolated individual structures that may exist within the group. These synchronized subgroups reflect the degree of participation and neural synergistic relationships of different individuals in group collaboration tasks, and serve as an important basis for subsequent anomaly identification and individualized interaction.

[0105] Optionally, the subgroup aggregation first-level unit is specifically used to confirm that two target objects constitute a synchronized subgroup when the synchronization index of the two-person neural activities corresponding to any two target objects is greater than or equal to the significant synchronization level, and to determine that the target object is an independent individual when the synchronization index of the two-person neural activities of any one target object and other target objects is less than the significant synchronization level.

[0106] The determination of synchronized subgroups is based on a two-person synchronization threshold condition. For a given brain region of interest... When the synchronicity index of any two individuals within a subset of individuals in a given brain region both reach a preset significance level, the subset of individuals is determined to constitute a synchronized subgroup in that brain region. For example, for a group of three children, if the synchronicity index of any two children in the brain region reaches a preset significance level, the subset of individuals is considered to constitute a synchronized subgroup in that brain region. If the neural activity synchronicity in all brain regions of interest reaches a significant level, then the three-person group is considered to have achieved significant synchronicity in the first stage. A synchronized subgroup is formed on each brain region of interest, representing the coordinated processing state of the corresponding functions in that brain region. If the first... The child and the first The number of children reached a significant synchronization level, the number of children... The child and the first The number of children reached a significant level of synchronization, but compared with the number of children... If a child does not reach a significant synchronization level, then the child in the first stage... The child and the first The children form a synchronous subgroup, the first... The child and the first Children constitute a synchronous subgroup, but three children do not form a synchronous subgroup. If a child is not significantly synchronous with any other child, then that child is an isolated individual. As shown in Equations 9-12, This indicates the Fisher's Z-statistic threshold corresponding to a significance level of 0.05 (default). For the standardized first children and the The number of children in the The degree of synchronization in individual brain regions For the standardized first children and the The number of children in the The degree of synchronization in individual brain regions For the standardized first children and the The number of children in the The degree of synchronization in individual brain regions Representative at the Synchronization subgroups in brain regions, Representative at the Synchronization subgroup 1 on a brain region, Representative at the Synchronization subgroups 2 on brain regions.

[0107]

[0108]

[0109]

[0110]

[0111] The first-level unit for directional computation is used to determine the temporal sequence of neural activity changes among target objects in a synchronized subgroup based on the corrected brain functional signals.

[0112] In some alternative embodiments, the first-level unit for directionality calculation includes a second-level unit for delay and a second-level unit for direction determination.

[0113] The delayed secondary unit is used to perform temporal alignment of the corrected brain functional signals of different target objects in the same brain region, and to calculate the first follow-up change index value of the first target object following the second target object and the second follow-up change index value of the second target object following the first target object under different delay times.

[0114] The direction determination secondary unit is used to determine the first maximum neural activity synchronicity based on the first follow-change index value, determine the second maximum neural activity synchronicity based on the second follow-change index value, and determine the temporal sequence of neural activity changes of the first target object and the second target object based on the difference between the first maximum neural activity synchronicity and the second maximum neural activity synchronicity.

[0115] Based on the synchronicity of neural activity in two individuals, this study further characterizes the temporal sequence of neural activity changes among individuals in multi-person collaborative tasks to quantify the directional features in synchronicity. Directional analysis is performed on pairs of individuals that have reached a significant level of synchronization. By introducing a delay parameter on the time axis, the neural activity changes of different individuals in the same brain region are temporally aligned, and the correlation strength under different delay conditions is calculated to determine the sequential relationship of neural activity changes among individuals. Furthermore, by comparing the bidirectional delay correlation results, the directional features in the synchronization relationship are determined to distinguish different types of group interaction patterns such as leader-follower and active-passive. In one implementation, the directional calculation results can be jointly constrained with the role arrangement information in the collaborative task. When the directional determination result is consistent with the task role, its confidence level is increased; otherwise, it is marked as a weak or abnormal directional relationship. The directional calculation unit ultimately outputs synchronization features containing directional information, which are used to construct a group neural activity synchronization network with directional attributes and provide a basis for subsequent anomaly identification, key individual location, and personalized interaction strategy generation. (See formulas 13-16). Figure 6 As shown, Indicates a time delay of When, calculate the first Did the children follow the number? A function that changes in the number of children. Represents the corrected number The number of children in the Brain function signals in each brain region Represents the corrected number The number of children in the On a unified time reference, individual brain regions The upload was delayed. The logic of brain function signals in seconds lies in the fact that if... The sequence is shifted to the right, and... If the synchronization is stronger, it means It happened earlier, Passed It took a few seconds to catch up, which, in terms of direction, is the [number]th [time]. The children followed the first Children, Indicates within a preset time interval The first The children followed the number The greatest change in neural activity synchronicity among children Indicates within a preset time interval The first The children followed the number The greatest change in neural activity synchronicity among children This indicates the difference between the two values ​​mentioned above. If the value is greater than 0, it indicates that it is the [number]th [unit]. The children followed the number The magnitude of change in the first child was greater than that in the second child. The children followed the number The magnitude of the changes in the children, then overall the direction is the first The children followed the first 100 children.

[0116]

[0117]

[0118]

[0119]

[0120] The first-level unit of the group synchronization network is used to represent each target object with nodes. The edges between nodes are determined based on the group synchronization index, and the arrow direction of the edges is determined based on the temporal relationship of the neural activity changes among the target objects. The group synchronization network structure is generated based on the nodes, edges, and arrow directions.

[0121] In some optional embodiments, the risk monitoring module includes: a first risk assessment primary unit, a second risk assessment primary unit, a third risk assessment primary unit, and a comprehensive primary unit.

[0122] The first risk assessment unit is used to obtain the first offset of each brain region of interest based on the population synchronicity index, the mean of synchronicity of the age-matched norm population, and the standard deviation of synchronicity of the age-matched norm population, and to determine the population synchronicity risk component based on the first offset and the weight corresponding to each brain region of interest.

[0123] The risk component is calculated based on group synchronization deviation characteristics. By fusing the synchronization results among all individuals within the group, it characterizes whether the group can form a stable overall neural coordination state during cooperative tasks. When the group synchronization level is significantly lower than the age-appropriate norm, it indicates that the group has overall deficiencies in collaborative processing, joint attention, or information integration. Since ASD-related disorders often manifest as difficulty integrating into group collaboration in social interactions, this feature is used to assess the overall coordination state of children in their social environment from a macro perspective, providing background reference for subsequent individual risk assessment. As shown in Formulas 17-18... Representing the The deviation of group synchronicity of brain regions of interest relative to age-matched norms Indicates the first Synchronization of brain regions of interest in a group For the first Mean synchronicity of age-matched norm groups for brain regions of interest For the first The standard deviation of synchronicity among age-matched norm groups of brain regions of interest As a component of group synchronicity risk, For the first The weights of each brain region of interest (assuming all brain regions of interest have equal weights, but can also be differentiated based on prior theories or literature results). This is the set of all brain regions of interest.

[0124]

[0125]

[0126] The second risk assessment unit is used to determine the participation rate of the brain region of interest for each target object based on the synchronous subgroup, and to obtain the subgroup participation risk component based on the subgroup size and the average synchronization degree of the subgroup in which the target object participates.

[0127] The risk component is calculated based on subgroup participation characteristics. This characteristic not only considers whether a child participates in any synchronous subgroup, but also comprehensively considers the size of the subgroup and the average synchronization intensity within the subgroup, thereby distinguishing different collaboration patterns such as "stable participation in core collaborations," "occasional participation in peripheral collaborations," or "long-term isolation." For children at high risk of ASD, it is often difficult to form stable collaborative relationships with multiple peers, making them easily excluded from highly synchronous subgroups, or only able to maintain limited interaction in very small-scale collaborations. Therefore, subgroup participation characteristics reveal a child's position and degree of participation in multi-person collaborative networks from the perspective of the group's internal structure. As shown in formulas 19-24, Indicates the first The participation rate of brain regions of interest in each child. Indicates the first The child in Whether the brain region has entered any synchronization subgroup (if so, the value is 1; otherwise, the value is 0). For the set of all brain regions of interest, if the first The child in If a brain region participates in at least one synchronization subgroup, then the size of the subgroup it participates in is further calculated. and the average synchronization level within the subgroup , For the standardized first children and the The number of children in the The degree of synchronization in individual brain regions Representing the The number of children in the Synchronization subgroups involved in each brain region This represents the average participation rate of brain regions of interest among age-matched norms. The standard deviation of the participation rate in brain regions of interest for age-matched norms. For the risk component of the brain region of interest, For the first Mean synchronization of subgroups of age-matched norms for brain regions of interest For the first Standard deviation of synchronization of subgroups of brain regions of interest with age-matched norms For the risk component of subgroup synchronization, For the first The weights of each brain region of interest (assuming all brain regions of interest have equal weights, but can also be differentiated based on prior theories or literature results). For subgroups to participate in risk weighting and These are the weighting coefficients that control the participation rate of brain regions of interest and the degree of synchronization of subgroups, respectively.

[0128]

[0129]

[0130]

[0131]

[0132]

[0133]

[0134] The third risk assessment unit is used to obtain the directional consistency risk component based on the chronological order and the roles of each target object.

[0135] Risk components are calculated based on directional consistency characteristics. This feature analyzes the temporal relationship of neural activity changes between children and others to determine whether a child can form a neural synchronization pattern of being followed by others under a guiding role, or an effective following pattern of the guiding role under a collaborative role. Simultaneously, the obtained directional results are compared with the pre-set role information in the task to assess whether the directional synchronization is consistent with role expectations. For individuals at risk of ASD, this often manifests as difficulty in understanding or internalizing social role rules; their neural synchronization may lack a stable guiding-following direction, or exhibit interaction patterns inconsistent with the task role. Therefore, directional consistency characteristics are used to assess children's role alignment ability in social collaboration from the perspective of interaction mechanisms. As shown in formulas 25 and 26, the risk component is calculated based on different stages of the task and the arrangement of different roles. The child in Has any change been made in the brain region that aligns with the task role? (Taking the group resource allocation and negotiation task described in step 1 as an example, for a child in the role of team leader, there should be more team members following along, but the value is 1 when the situation matches the role identity, otherwise the value is 0). The percentage of brain regions of interest that match the role's identity. For the set of all brain regions of interest, The percentage of brain regions of interest that match the role identity is the average among age-matched norms. The standard deviation of the percentage of brain regions of interest that match the role identity among age-appropriate norms. The risk component is determined by the ratio of brain regions of interest to roles and identities.

[0136]

[0137]

[0138] The integrated first-level unit is used to obtain the comprehensive risk based on the group synchronization risk component, the subgroup participation risk component, and the direction consistency risk component, and to obtain auxiliary analysis results based on the comprehensive risk.

[0139] After calculating the group synchronization characteristics, subgroup participation characteristics, and directional consistency characteristics, the first-level unit integrates the above multidimensional risk components to output the autism spectrum disorder risk stratification results for children. This is used to adjust the relative contribution of different risk dimensions in the comprehensive assessment and can be preset or adaptively adjusted according to task type, age group, or clinical needs. As shown in Formula 27, For ASD comprehensive risk score, As a component of group synchronicity risk, For subgroups to participate in risk weighting For brain regions of interest, the risk component is determined by the ratio of role identity. These are the weighting coefficients for controlling the three risk components.

[0140]

[0141] In some optional embodiments, the system further includes an anomaly detection module and an annotation module.

[0142] The anomaly detection module is used to determine abnormal states based on the group synchronization network structure, the synchronization index of two-person neural activities, and the group synchronization index.

[0143] The annotation module is used to annotate the task stage, brain region, and target object corresponding to abnormal states.

[0144] When the synchronization characteristics of the group network or individuals deviate from the preset normal range, this module labels and records the anomaly. The labeling includes at least: the task stage in which the anomaly occurred, the target brain region / network associated with the anomaly, the individual involved in the anomaly, and the anomalous relationship pairing or anomalous subgroup structure. This result serves as the basis for subsequent interactive target selection and parameter generation.

[0145] First, based on the group synchronization network structure and individual synchronization characteristics, a comparison is made with a preset normal reference range. When a significant deviation is detected in the group synchronization index, subgroup structure, or directional consistency characteristics, the abnormal state is automatically labeled and recorded. The abnormal labeling includes at least the task stage in which the abnormality occurred, the target brain region or functional network associated with the abnormality, the individual involved in the abnormality, and the relationship pairing or subgroup structure form manifested by the abnormality, thereby providing a structured basis for subsequent interactive decision-making.

[0146] In some alternative embodiments, the system further includes:

[0147] The interaction target selection module is used to determine any one of the following as the interaction target based on the annotation results of abnormal states: a key individual node, a key edge, or a key subgroup. This module automatically determines the interaction target based on the abnormal annotation results, including at least: key individual node selection, such as the individual with the lowest connectivity or the largest deviation in a group network; key relationship edge selection, such as the continuous abnormality in synchronicity between two specific individuals at a critical stage; and key subgroup selection, such as the break in coupling between a core collaborative subgroup and a peripheral subgroup.

[0148] The interaction parameter generation module is used to generate interaction parameters based on the interaction goal and comprehensive risk. This module generates optical stimulation interaction parameters that match the individual's neural state, based on the interaction goal and risk stratification results. These interaction parameters include, but are not limited to: stimulation intensity, stimulation frequency, stimulation rhythm, stimulation sequence, stimulation duration, target brain region, and the order / coordination strategy of stimulating multiple subjects.

[0149] The interaction module is used to stimulate interactive targets based on interaction parameters. This module is used to perform non-invasive light stimulation interaction on multiple children based on these parameters, in order to modulate the group synchronization network structure and improve the level of neural activity synchronization in multi-person interactions. This module supports: single-point interaction, stimulating only key individuals to drive group synchronization; multi-point interaction, simultaneously stimulating multiple key individuals or key brain regions; and collaborative interaction, providing phased stimulation to key relational edges or key subgroups. Furthermore, it dynamically adjusts the interaction parameters through real-time synchronization feedback, forming a closed-loop optimization.

[0150] The interaction effect determination module compares the group synchronization network structure before and after the interaction to determine the interaction effect, and updates the parameter generation strategy of the interaction parameter generation module based on the interaction effect. This module is used to compare and evaluate the group synchronization network, individual risk scores, and behavioral coordination indicators before and after the interaction, and update the reference model or individualized parameter library to support long-term follow-up and individualized interaction iteration.

[0151] The interaction target selection module, interaction parameter generation module, interaction module, and interaction effect determination module are used to automatically determine the interaction target based on the anomaly annotation results. The target includes at least: key individual node selection, such as the individual with the lowest connectivity or the largest deviation in the group network; key relationship edge selection, such as the continuous anomaly in synchronicity between two specific individuals at a critical stage; and key subgroup selection, such as the coupling break between the core collaborative subgroup and the peripheral subgroup.

[0152] After completing the calculation of the synchronization of neural activities of multiple people and the risk stratification analysis, this application further introduces anomaly detection and closed-loop interaction mechanism to achieve accurate identification and control of abnormal neural coordination patterns in the process of multi-person interaction.

[0153] The interaction target selection module automatically filters and locates interaction targets based on anomaly labeling results. Starting from the group network topology and synchronization characteristic distribution, this module comprehensively determines the priority targets for interaction. These targets include at least: key individual nodes with significantly reduced connectivity or the greatest deviation in the group synchronization network; individual pairs or relationship edges that consistently exhibit synchronization anomalies during critical task phases; and core or peripheral cooperative subgroups showing significant coupling breaks during group collaboration. This hierarchical selection of interaction targets avoids homogenized interaction for all individuals, improving the targeting and efficiency of interaction. Subsequently, the interaction parameter generation module, combining the interaction targets and individual risk stratification results, automatically generates optical stimulation interaction parameters that match the child's current neural state. These parameters include at least stimulation intensity, stimulation frequency, stimulation rhythm, stimulation sequence, stimulation duration, and the target brain region. When multiple interaction targets are involved, the module further determines the order of stimulation or a collaborative stimulation strategy for multiple targets to ensure coordination between the interaction process and the rhythm of group interaction. During the interactive implementation phase, the interaction module performs non-invasive photostimulation on multiple children based on generated interaction parameters to regulate their neural activity synchronization and optimize the group synchronization network structure. This module supports multiple interaction modes, including: single-point interaction targeting key individuals to improve overall group synchronization by regulating their neural activity; multi-point parallel interaction targeting multiple key individuals or key brain regions; and phased collaborative interaction targeting abnormal relationships or key subgroups. During the interaction, the module dynamically adjusts the stimulation parameters based on real-time synchronization feedback, forming a closed-loop optimization mechanism that allows the interaction process to adaptively evolve with changes in the group's neural coordination state. Finally, the interaction effect determination module compares and analyzes the group synchronization network structure, individual risk scores, and behavioral coordination indicators before and after the interaction to evaluate the effectiveness and stability of the interaction. Based on the evaluation results, the reference model parameters or individualized interaction parameter library are updated to support risk assessment and interaction decisions in subsequent tasks, thereby achieving long-term tracking, individualized regulation, and continuous optimization of children's neural coordination abilities.

[0154] It should be fully understood that the user information involved in this application (including but not limited to user physiological information, user personal information, etc.) is information and data authorized by the user or fully authorized by all parties. The use of user information shall comply with privacy policies and practices that are often considered to meet or exceed industry requirements for maintaining user privacy. The collection, use and processing of related data shall comply with relevant laws, regulations and standards, and provide corresponding operation access points for users to choose to authorize or refuse.

[0155] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0156] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0157] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. An auxiliary analysis system based on the synchronization of neural activity, characterized in that, The system includes: The collaborative task output module is used to output multi-person collaborative tasks; The task execution module is used to adjust the participation of each target object in the cooperative task synchronously within the same time window based on the output of the multi-person cooperative task. The data acquisition module is used to collect brain function signals and behavioral data of each target object during the execution of the multi-person collaborative task. The interference control module is used to collect interference information during the execution of multi-person collaborative tasks; The group neural activity synchronization calculation module is used to quantify the neural activity synchronization of any two target objects in the cooperative task process based on the collected brain function signals, behavioral data and interference information of each target object, to obtain a neural activity synchronization index; and to obtain a group synchronization index and a group synchronization network structure based on the neural activity synchronization index. The risk monitoring module is used to obtain auxiliary analysis results based on the neural activity synchronicity index and the group synchronization network structure. The group neural activity synchronization calculation module includes: The data correction primary unit is used to correct the collected brain function signals of each of the target objects based on the interference information to obtain the corrected brain function signals. The first-level unit for dual-person synchronization calculation is used to quantify the synchronization of neural activities of any two target objects in a cooperative task based on the corrected brain function signals and behavioral data, and obtain a dual-person neural activity synchronization index. A group synchronization fusion primary unit is used to fuse the synchronization indices of the two-person neural activities to obtain group synchronization indices; The subgroup aggregation first-level unit is used to identify synchronized subgroups based on the dual-person neural activity synchronization index; The first-level directional computing unit is used to determine the temporal sequence of neural activity changes among target objects in the synchronized subgroup based on the corrected brain function signals. The group synchronization network represents the first-level unit, which is used to represent each target object with nodes, determine the edges between nodes based on the group synchronization index, determine the arrow direction of the edges based on the temporal relationship of the neural activity changes between each target object, and generate the group synchronization network structure based on the nodes, edges and the arrow direction. The risk monitoring module includes: The first risk assessment unit is used to obtain the first offset of each brain region of interest based on the group synchronicity index, the mean of synchronicity of the age-matched norm group, and the standard deviation of synchronicity of the age-matched norm group, and to determine the group synchronicity risk component based on the first offset and the weight corresponding to each brain region of interest. The second risk assessment unit is used to determine the participation rate of the brain region of interest for each target object based on the synchronized subgroup, and to obtain the subgroup participation risk component based on the subgroup size and the average synchronization degree of the subgroup in which the target object participates. The third risk assessment unit is used to obtain the directional consistency risk component based on the time sequence and the role of each target object. The integrated first-level unit is used to obtain a comprehensive risk based on the group synchronization risk component, the subgroup participation risk component, and the direction consistency risk component, and to obtain auxiliary analysis results based on the comprehensive risk.

2. The system according to claim 1, characterized in that, The data acquisition module includes: The brain function signal acquisition primary unit is used to synchronously initialize each acquisition device through a unified trigger signal or a unified sampling clock, and to acquire the brain function signals of each target object based on each acquisition device after synchronous initialization. The behavior data acquisition unit is used to synchronously collect the task response time, operation sequence, and interaction event markers of each target object. The timestamp alignment primary unit is used to uniformly timestamp the collected brain function signals and behavioral data to obtain a data stream, and to write the switching of each task stage of the multi-person collaborative task, the interaction event markers, and the system feedback nodes as event markers into the data stream.

3. The system according to claim 1, characterized in that, The data correction primary unit includes: The secondary unit for determining data credibility weights is used to obtain data credibility weights based on the head motion index, signal quality index, and environmental interference index of the target object. The secondary correction unit is used to obtain the corrected brain function signal based on the data confidence weight and the brain function signal of the target object.

4. The system according to claim 1, characterized in that, The subgroup aggregation first-level unit is specifically used to confirm that the two target objects constitute a synchronized subgroup when the synchronization index of the two-person neural activities corresponding to any two target objects is greater than or equal to the significant synchronization level, and to determine that the target object is an independent individual when the synchronization index of the two-person neural activities of any one target object and other target objects is less than the significant synchronization level.

5. The system according to claim 1, characterized in that, The first-level unit for directional calculation includes: The delayed secondary unit is used to perform temporal alignment of the modified brain functional signals of different target objects in the same brain region, and to calculate the first follow-up change index value of the first target object following the second target object and the second follow-up change index value of the second target object following the first target object under different delay times. The direction determination secondary unit is used to determine the first maximum neural activity synchronicity based on the first follow-change index value, determine the second maximum neural activity synchronicity based on the second follow-change index value, and determine the temporal sequence of neural activity changes of the first target object and the second target object based on the difference between the first maximum neural activity synchronicity and the second maximum neural activity synchronicity.

6. The system according to claim 1, characterized in that, The system also includes: An anomaly detection module is used to determine anomaly states based on the group synchronization network structure, the synchronization index of two-person neural activities, and the group synchronization index. The annotation module is used to annotate the task stage, brain region, and target object corresponding to the abnormal state.

7. The system according to claim 6, characterized in that, The system also includes: The interaction target selection module is used to determine any one of the key individual nodes, key edges, or key subgroups as the interaction target based on the annotation results of the abnormal state. An interaction parameter generation module is used to generate interaction parameters based on the interaction objective and the comprehensive risk. An interaction module is used to stimulate the interaction target based on the interaction parameters; The interaction effect determination module is used to compare the group synchronization network structure before and after the interaction to determine the interaction effect, and update the parameter generation strategy corresponding to the interaction parameter generation module based on the interaction effect.

8. The system according to any one of claims 1 to 7, characterized in that, The multi-person collaborative task is configured according to the type of cognitive processing and the degree of collaborative dependence. The multi-person collaborative task includes one or more of the following: collaborative task based on action synchronization, collaborative task based on information sharing, collaborative task based on joint decision-making, and collaborative task based on creative expression. Each of the target objects is divided into at least two teams, and each team includes a guide role and a collaborator role.

Images