Cognitive assessment and intervention methods for rescue personnel under brain signal and physiological load regulation

Abstract

The application provides a method for cognitive evaluation and intervention of rescue personnel under brain signal and physiological load regulation, and belongs to the field of cognitive evaluation. The method comprises the following steps: collecting brain function signals and peripheral physiological signals of rescue personnel during a task and extracting corresponding features; constructing multiple cognitive state indexes based on the brain function features and constructing a physiological load index based on the peripheral physiological features; taking the physiological load index as a regulation factor, mapping it into a regulation factor, combining a preset regulation sensitivity coefficient of each cognitive dimension, correcting the cognitive state indexes, and obtaining corrected cognitive state indexes; comparing the corrected cognitive state indexes with a preset threshold to determine whether each cognitive dimension is in an abnormal state; and generating an intervention strategy according to the determination result and executing the intervention strategy by dynamically adjusting the output mode of task information. In this way, the non-specific interference of physiological signals on cognitive evaluation is effectively isolated, and appropriate intervention is performed according to the evaluation result.

Description

Technical Field

[0001] This invention belongs to the field of cognitive assessment, specifically relating to a method for cognitive assessment and intervention of rescue personnel under the regulation of brain signals and physiological load. Background Technology

[0002] In complex rescue environments, rescuers' cognitive state is influenced by a combination of factors, making them prone to problems such as slowed information perception, inattention, memory disturbance, and slower logical reasoning, thereby increasing the risk of operational errors and personal injury. Therefore, it is essential to conduct real-time cognitive state assessments of rescuers and take effective intervention measures when necessary.

[0003] Early studies often assessed cognitive states based on single signals, such as using only brain function signals or single peripheral physiological indicators to characterize attention levels or workload. While these methods are relatively easy to implement, they struggle to consistently reflect the dynamic changes in cognitive states within complex task environments due to the limited information source. Furthermore, the assessment results are easily affected by individual differences and transient fluctuations, resulting in insufficient accuracy.

[0004] Building upon this foundation, to improve the accuracy of cognitive state assessment, multimodal signal fusion methods are increasingly being adopted. These methods use brain functional signals and peripheral physiological signals as inputs, directly mapping them through neural network models to obtain cognitive state indicators. However, direct fusion of multi-source signals typically uses both brain functional and physiological signals as input variables, mapping them through a unified model. While this improves model accuracy, it makes it difficult to distinguish the specific impact of different signals on the results, reducing the interpretability of the assessment and hindering the implementation of subsequent intervention measures. Furthermore, some peripheral physiological signals exhibit strong nonspecificity and are easily affected by environmental conditions, emotional states, and other factors. During direct fusion modeling, these interferences may be mistakenly identified as changes in cognitive state, thus affecting the stability of the assessment.

[0005] Therefore, in complex rescue environments, existing cognitive state assessment methods that rely solely on a single signal are unable to stably reflect dynamic changes in cognition. When brain function signals are directly fused with peripheral physiological signals for modeling, non-specific interference in the physiological signals can easily be mistaken for changes in cognitive state, leading to unstable assessment results with poor interpretability. It is impossible to clearly distinguish whether the decline in cognitive ability stems from changes in brain function itself or fluctuations in physiological load, and it is also difficult to implement targeted interventions accordingly. Summary of the Invention

[0006] To address the issues of low interpretability and poor stability in existing technologies for assessing the cognitive state of rescue personnel, this invention provides a method for assessing and intervening in the cognitive state of rescue personnel under the regulation of brain signals and physiological load.

[0007] To achieve the above objectives, the present invention provides the following technical solution: A method for cognitive assessment and intervention of rescue personnel under brain signal and physiological load regulation, the method comprising: Collect brain function signals and peripheral physiological signals of rescue personnel during the mission, and extract brain function characteristics and peripheral physiological characteristics; Based on the aforementioned brain functional characteristics, multiple cognitive state indicators are constructed, corresponding to different cognitive dimensions in attention, perception, working memory, and decision processing. A physiological load index is constructed based on the aforementioned peripheral physiological characteristics to characterize the overall physiological load level of an individual; Using the physiological load index as a moderating factor, it is mapped to a moderating factor, and combined with the preset moderating sensitivity coefficients of each cognitive dimension, the cognitive state index is modified to obtain the modified cognitive state index. The corrected cognitive state index is compared with a preset threshold to determine whether each cognitive dimension is in a declining or abnormal state. An intervention strategy is generated based on the judgment result, and the output method of the display device and voice output device for task information is dynamically adjusted based on the intervention strategy.

[0008] Optionally, multiple cognitive state indicators are constructed based on the aforementioned brain functional characteristics, including: At least one of the following is extracted from brain functional signals: frequency band power features, event-related potential features, and brain region functional connectivity features. The corresponding cognitive dimension index values ​​are obtained by normalization and weighted combination mapping. Among them, the attention state index is constructed based on the relative relationship between theta band power and alpha band power, the perceptual response index is constructed based on the amplitude and / or latency of event-related potentials, the working memory capacity index is constructed based on the theta band power change in the prefrontal region and brain region functional connectivity features, and the decision processing index is constructed based on the functional connectivity strength or synchronicity between multiple brain regions.

[0009] Optionally, constructing a physiological load index based on the peripheral physiological characteristics includes: At least one of the peripheral physiological signals, including heart rate, heart rate variability, body temperature, and dehydration-related parameters, is normalized based on individual baselines, and the physiological load index is obtained by linear weighting or nonlinear combination.

[0010] Optionally, the formula for obtaining the physiological load index through normalization based on individual baselines and by linear weighting or nonlinear combination is as follows: ; in, Let be the normalized deviation of the j-th peripheral physiological characteristic relative to its individual baseline. For the corresponding weighting coefficients; or, for the normalized indicators of heart rate, heart rate variability, and body temperature. , and Nonlinear combinations are used to reflect the coupling effect of multiple physiological indicators: .

[0011] Optionally, the physiological load index is used as a moderating factor, mapped to a moderating factor, and modified by combining preset moderating sensitivity coefficients for each cognitive dimension, including: The physiological load index was statistically standardized to obtain the standardized load value. Using a nonlinear function with optimal interval characteristics to... Mapped to regulation factor Based on this regulating factor, a modified cognitive state index was obtained. : ; in, This is an indicator of the original cognitive state. Let be the moderating sensitivity coefficient of the i-th cognitive dimension, and be the moderating factor. The formula for determining it is: ; in, The location parameters are used to determine the optimal cognitive state corresponding to the standardized load.

[0012] Optionally, the preset threshold is set based on individual baseline data or historical statistical data, and is set separately for different cognitive dimensions; when the corrected cognitive state index is lower than the corresponding threshold, or shows a continuous downward trend within the preset time window, the cognitive dimension is determined to be in a declining or abnormal state.

[0013] Optionally, the intervention strategy is generated according to a preset mapping rule, which maps the judgment results of different cognitive dimensions to the corresponding information output control parameters respectively; wherein, when attention declines, the display intensity of key information is enhanced; when working memory declines, information is output in stages to reduce the burden of immediate information; when perception declines, the output of non-key information is suppressed to reduce interference; when multiple cognitive dimensions are abnormal at the same time, the control parameters of each strategy are coordinated according to the preset priority of key task capabilities.

[0014] A device for assessing and intervening in the cognitive state of rescue workers, the device comprising: The data acquisition module is used to collect brain function signals and peripheral physiological signals of rescue personnel during the mission, and to extract brain function features and peripheral physiological features. The module is used to construct multiple cognitive state indicators based on the brain functional characteristics, corresponding to different cognitive dimensions in attention, perception, working memory, and decision processing; and to construct a physiological load index based on the peripheral physiological characteristics to characterize the individual's overall physiological load level. The correction module is used to map the physiological load index as a regulation factor and combine it with the preset regulation sensitivity coefficients of each cognitive dimension to correct the cognitive state index and obtain the corrected cognitive state index. The judgment module is used to compare the corrected cognitive state index with a preset threshold to determine whether each cognitive dimension is in a declining or abnormal state. The intervention module is used to generate intervention strategies based on the judgment results, and dynamically adjust the output mode of the display device and voice output device to the task information based on the intervention strategies.

[0015] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned method for cognitive assessment and intervention of rescue personnel under the regulation of brain signals and physiological load.

[0016] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the aforementioned method for cognitive assessment and intervention of rescue personnel under the regulation of brain signals and physiological load.

[0017] The cognitive assessment and intervention method for rescue personnel under the regulation of brain signals and physiological load provided by this invention has the following beneficial effects: This invention introduces physiological load as an independent modulatory factor into the cognitive assessment process, rather than directly fusing it with brain function signals. This hierarchical processing architecture ensures that peripheral physiological signals are used solely to generate a physiological load index reflecting the overall load level, which is then used to correct the assessment results for different cognitive dimensions. This approach effectively isolates the influence of non-specific interference in physiological signals on cognitive assessment, making the corrected cognitive state indicators closer to the actual cognitive abilities of rescue personnel, thus improving the stability and anti-interference capability of the assessment. Furthermore, because brain function assessment and physiological load regulation are independent in their processes, the pathways of each signal's impact on the final result are clear, resulting in highly interpretable assessment results. Based on this, abnormal dimensions can be accurately identified according to changes in cognitive state, and the display intensity, priority, and output rhythm of task information can be dynamically adjusted. This achieves closed-loop intervention in cognitive state without altering the content of the task information, thereby reducing the risk of human error by rescue personnel in complex environments. Attached Figure Description

[0018] To more clearly illustrate the embodiments and design schemes of the present invention, the accompanying drawings required for this embodiment will be briefly described below. The drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a flowchart illustrating a method for cognitive assessment and intervention of rescue personnel under the regulation of brain signals and physiological load, provided by the present invention according to an exemplary embodiment.

[0020] Figure 2 This is a flowchart illustrating another method for cognitive assessment and intervention of rescue personnel under the regulation of brain signals and physiological load, provided by the present invention according to an exemplary embodiment.

[0021] Figure 3 This is a schematic diagram illustrating a cognitive state correction principle according to an exemplary embodiment of the present invention.

[0022] Figure 4 This is a schematic diagram of an intervention strategy execution process provided by the present invention according to an exemplary embodiment.

[0023] Figure 5 This is a block diagram of a cognitive assessment and intervention device for rescue personnel under the regulation of brain signals and physiological load, provided by the present invention according to an exemplary embodiment. Detailed Implementation

[0024] To enable those skilled in the art to better understand and implement the technical solutions of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. The following embodiments are only used to more clearly illustrate the technical solutions of the present invention and should not be construed as limiting the scope of protection of the present invention.

[0025] The technical solutions provided by the various embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0026] First, this invention provides a method for cognitive assessment and intervention of rescue personnel under the regulation of brain signals and physiological load, specifically as follows: Figure 1 As shown, it includes the following steps: S101. Collect brain function signals and peripheral physiological signals of rescue personnel during the mission, and extract brain function characteristics and peripheral physiological characteristics.

[0027] In this step, rescuers wear EEG collection caps and chest-strap physiological sensors. While performing pathfinding tasks within buildings, they simultaneously collect EEG signals, heart rate, and heart rate variability signals in fixed 10-second time windows. The EEG signals are bandpass filtered from 0.5 to 40 Hz, and independent component analysis is used to remove artifacts from electrooculography (EOG) and electromyography (EMG). The absolute power of the theta and alpha bands is extracted, normalized, and used as brain function features. For peripheral physiological signals, outliers caused by movement interference are removed, and sliding smoothing is applied. The root mean square of the average heart rate and the difference between adjacent RR intervals within the current window is calculated as heart rate variability indicators. Using the individual's resting baseline as a reference, heart rate deviation and heart rate variability deviation are calculated, normalized, and used as peripheral physiological features. These brain function features and peripheral physiological features form a feature vector, which is input into subsequent cognitive assessment and load regulation processes.

[0028] S102. Based on this brain functional characteristic, construct multiple cognitive state indicators, which correspond to different cognitive dimensions in attention, perception, working memory and decision processing, respectively; based on this peripheral physiological characteristic, construct a physiological load index to characterize the overall physiological load level of an individual.

[0029] In this step, at least one of the following is extracted from brain functional signals: frequency band power features, event-related potential features, and brain region functional connectivity features. These are then normalized and weighted to obtain the corresponding cognitive dimension index values. Specifically, the attention state index is constructed based on the relative relationship between theta band power and alpha band power; the perceptual response index is constructed based on the amplitude and / or latency of event-related potentials; the working memory capacity index is constructed based on the theta band power changes in the prefrontal region and brain region functional connectivity features; and the decision processing index is constructed based on the functional connectivity strength or synchronicity between multiple brain regions. Next, at least one of the peripheral physiological signals—heart rate, heart rate variability, body temperature, and dehydration-related parameters—is normalized based on an individual baseline. The physiological load index formula is then obtained through linear weighting or nonlinear combination:

[0030] ; in, Let be the normalized deviation of the j-th peripheral physiological characteristic relative to its individual baseline. For the corresponding weighting coefficients; or, for the normalized indicators of heart rate, heart rate variability, and body temperature. , and Nonlinear combinations are used to reflect the coupling effect of multiple physiological indicators: .

[0031] S103. Using the physiological load index as a moderating factor, map it into a moderating factor, and combine it with the preset moderating sensitivity coefficients of each cognitive dimension to modify the cognitive state index, thereby obtaining the modified cognitive state index.

[0032] In this step, the physiological load index is statistically standardized to obtain the standardized load value. Using a nonlinear function with optimal interval characteristics to... Mapped to regulation factor Based on this regulating factor, a modified cognitive state index was obtained. : ; in, This is an indicator of the original cognitive state. Let be the moderating sensitivity coefficient for the i-th cognitive dimension, and this moderating factor The formula for determining it is: ; in, The location parameters are used to determine the optimal cognitive state corresponding to the standardized load.

[0033] S104. Compare the revised cognitive state index with the preset threshold to determine whether each cognitive dimension is in a declining or abnormal state; generate an intervention strategy based on the determination result, and execute the intervention strategy by dynamically adjusting the output method of task information.

[0034] The adjustment of this output method includes modifying one or more of the following: display intensity, information priority, information density, and output rhythm. The preset threshold is set based on individual baseline data or historical statistical data, and is set separately for different cognitive dimensions. When the corrected cognitive state index is lower than the corresponding threshold, or shows a continuous downward trend within a preset time window, the cognitive dimension is determined to be in a declining or abnormal state. The intervention strategy is generated according to a preset mapping rule, which maps the judgment results of different cognitive dimensions to the corresponding information output control parameters. Specifically, when attentional state declines, the display intensity of key information is enhanced; when working memory declines, information is output in stages to reduce the immediate information burden; and when perceptual ability declines, non-key information is suppressed to reduce interference. When multiple cognitive dimensions are abnormal simultaneously, the control parameters of each strategy are coordinated according to the preset priority of key task abilities.

[0035] By employing the aforementioned method, a hierarchical processing architecture is constructed by introducing physiological load as an independent modulatory factor into the cognitive assessment process, rather than directly fusing it with brain function signals. This architecture ensures that peripheral physiological signals are used solely to generate a physiological load index reflecting the overall load level, which is then used to uniformly correct the assessment results for different cognitive dimensions. This approach effectively isolates the influence of non-specific interference in physiological signals on cognitive assessment, making the corrected cognitive state indicators closer to the actual cognitive abilities of rescue personnel, thus improving the stability and anti-interference capability of the assessment. Furthermore, because brain function assessment and physiological load regulation are independent in their processes, the pathways of each signal's impact on the final result are clear, resulting in highly interpretable assessment results. Based on this, abnormal dimensions can be accurately identified according to changes in cognitive state, and the display intensity, priority, and output rhythm of task information can be dynamically adjusted. This achieves closed-loop intervention in cognitive state without altering the content of the task information, thereby reducing the risk of human error by rescue personnel in complex environments.

[0036] Based on the above method steps, the present invention also provides an embodiment of the execution steps.

[0037] For rescue personnel, this invention constructs a complete "assessment-correction-judgment-intervention" process. First, it acquires brain function signals and peripheral physiological signals, optionally including some environmental information. Cognitive state is analyzed based on brain function signals, and then regulated through physiological load. The corrected assessment results are compared with preset thresholds, and different information output methods are adjusted according to the judgment results to achieve intervention in the cognitive state. Figure 2 As shown, the key steps include the following:

[0038] a) Collect brain function signals, peripheral physiological signals, and some environmental information of rescue personnel during the mission, and preprocess and extract features from various signals.

[0039] The brain functional signals include at least one or more of electroencephalogram (EEG) signals and near-infrared spectral signals. The peripheral physiological signals include at least one or more of heart rate, heart rate variability, skin conductance, respiratory rate, skin temperature, and core body temperature. Optional environmental information is included, primarily to provide task context and information source.

[0040] The acquired brain function signals undergo preprocessing operations such as denoising, filtering, and artifact removal. In one embodiment, bandpass filtering is performed on the EEG signals to remove low-frequency drift and high-frequency noise, and artifacts such as electrooculogram (EOG) and electromyogram (EMG) signals are removed using independent component analysis and threshold determination. Near-infrared brain function signals undergo baseline correction and low-pass filtering to eliminate motion interference and high-frequency noise.

[0041] Temporal and standardization processes are performed on peripheral physiological signals to improve data stability, including signal alignment, outlier removal, and smoothing filtering.

[0042] After processing, feature extraction was performed on brain functional signals and peripheral physiological signals. Among these, brain functional signal features ( This includes at least frequency domain features, time domain features, and spatial distribution features. Based on near-infrared signals, blood oxygen concentration variation features and corresponding temporal variation features are extracted. Furthermore, frequency domain features include the power of each frequency band and the ratio of partial power spectra; time domain features include the amplitude and latency of event-related potentials; and spatial distribution features include the strength or synchronicity of functional connectivity between different brain regions.

[0043] Characteristics of peripheral physiological signals ( At least statistical characteristics and trend characteristics should be included. For example, the average heart rate and heart rate variability coefficient can be extracted from the heart rate signal, the skin conductance level and transient response amplitude can be extracted from the skin conductance signal, and the respiratory rate and its fluctuation characteristics can be extracted from the respiratory signal.

[0044] The above processing yields a set of brain functional features and a set of physiological features, providing data for subsequent cognitive state assessment and physiological load modeling.

[0045] b) Furthermore, multiple cognitive state indicators are constructed based on brain functional characteristics ( (and input the data). Different indicators correspond to different cognitive dimensions such as attention state, perceptual response, working memory capacity, and decision-making ability.

[0046] In the mapping process, the data of various brain functional features are first normalized to eliminate the dimensional differences between different features. Then, based on the established empirical function and weights, the relevant features are processed and combined to obtain multiple cognitive state indicators.

[0047] In one implementation, the cognitive state indicators can be calculated as follows: ; in, For the j-th brain functional feature, Features Corresponding to cognitive state indicators The mapping function is implemented using a nonlinear transformation function in this embodiment. These are the corresponding weighting coefficients.

[0048] Specifically: (1) An attention state index is constructed based on the relative relationship between the power of the θ band and the power of the α band. The power ratio of the θ band and the θ / α ratio are calculated, and the attention state index is obtained after normalization. This type of feature reflects the changes in the allocation of attention resources and the suppression process.

[0049] (2) Construct a perception response index based on event-related potential characteristics. Extract the potential amplitude or peak value within a preset time window, standardize it, and then obtain the perception response index by weighted combination, which is used to characterize the response intensity during the processing of external stimuli.

[0050] (3) A working memory capacity index was constructed based on the power variation of the theta band in the prefrontal region and the functional connectivity characteristics of brain regions. The entropy of this spectrum was calculated and used as the main input of the working memory index, which reflects the body's ability to maintain and update information.

[0051] (4) Construct decision processing indicators based on the functional connectivity characteristics between multiple brain regions. Calculate the synchronicity or coupling strength between brain regions, and after normalization, use it to characterize the level of neural activity in the information integration and decision-making process.

[0052] It should be noted that there is actually a certain coupling relationship between the various cognitive dimensions, but considering the complexity of the model, they are regarded as approximately independent in this model. In practical applications, they can be extended through multi-dimensional combination.

[0053] After completing the above calculations, a unified scale mapping is performed on multiple cognitive state indicators to ensure that their values ​​are within the same range for subsequent comparison and judgment.

[0054] c) Further, a physiological load modeling process is introduced. By processing peripheral physiological signals, a physiological load index (PLI) reflecting the overall physiological load level is constructed.

[0055] The peripheral physiological signals include at least heart rate, heart rate variability, body temperature, and dehydration-related parameters. Different physiological indicators reflect the body's workload from different perspectives. For example, heart rate reflects overall metabolic and stress levels, heart rate variability reflects the autonomic nervous system's regulatory capacity, body temperature reflects heat stress levels, and dehydration-related parameters reflect fluid balance. Processing these physiological indicators allows for a comprehensive reflection of an individual's physiological workload in complex task environments.

[0056] Specifically, multiple physiological signals are first normalized to calculate their deviation from the individual baseline. In one implementation, each physiological characteristic can be represented as:

[0057] ; in, Let j be the current measured value of the j-th physiological indicator. For the corresponding baseline value, This is a preset range of variation.

[0058] After normalization, multiple physiological characteristics are weighted and combined to obtain a unified physiological load index. In one implementation, it can be expressed as:

[0059] ; in, This represents the weighting coefficient for each physiological indicator. This weighting coefficient reflects the degree of influence of different physiological conditions on the overall physiological load indicators. Its value can be set based on pre-experimental data, historical statistical data, or empirical parameters, while also taking into account individual conditions.

[0060] It should be noted that different physiological indicators have varying degrees of influence on cognitive state, but this difference is mainly reflected in their impact on overall physiological load indicators, rather than directly affecting changes in specific cognitive state indicators. Based on this physiological load indicator, cognitive state indicators can be adjusted in subsequent processes, and their differentiated responses can be reflected through adjustment sensitivity parameters of different cognitive dimensions.

[0061] Through the above modeling process, multiple peripheral physiological signals were converted into unified quantitative indicators, thereby providing a basis for correcting the results of cognitive state assessment.

[0062] d) Furthermore, this physiological workload index can be used to regulate cognitive state indicators. For example... Figure 3 As shown, this adjustment refers to mapping the physiological load index to a moderating factor and combining it with sensitivity coefficients of multiple cognitive dimensions to perform nonlinear correction on the cognitive state indicators. The nonlinear correction takes into account that the impact of physiological load on individual cognitive performance in existing studies is often nonlinear.

[0063] First, the physiological load index is standardized and mapped to reflect its position relative to the individual's historical distribution, thereby eliminating the influence of individual differences.

[0064] In one implementation, the physiological load index can be statistically standardized based on individual historical data or pre-experimental data to obtain a standardized load value: ; in, and These represent the mean and standard deviation from the individual's baseline or historical data, respectively. First, the physiological load index is standardized to indicate its position relative to the individual's historical distribution, thereby eliminating the influence of individual differences.

[0065] Based on this, a regulation model with optimal interval characteristics is constructed. Considering the often nonlinear relationship between physiological load and cognitive performance, in one implementation, the standardized physiological load is mapped to a regulation factor. Its expression is:

[0066] ; Where θ represents the position of the optimal cognitive state corresponding to the standardized load, which is usually taken as 0 or determined through pre-experimentation.

[0067] To improve engineering feasibility, in practical applications, this function can be discretized or implemented using a lookup table, that is, pre-defined based on different... Calculate the corresponding value And it is called directly during runtime.

[0068] Subsequently, the moderating factor was applied to each cognitive state indicator to obtain the corrected results: ; in, This is an indicator of the original cognitive state. This is the revised cognitive state indicator. This is the moderating sensitivity coefficient corresponding to the cognitive dimension. The source remains consistent with the aforementioned physiological load modeling process. In one implementation, The determination process includes the following steps:

[0069] (1) Collect physiological load index and cognitive state index sequences simultaneously in pre-experiment or historical data.

[0070] (2) Standardized physiological load values With various cognitive indicators Perform correlation analysis to calculate the relationship of change within a time window. For example, obtain the slope or magnitude of change of each cognitive index relative to changes in physiological load through linear regression or local fitting methods.

[0071] (3) Normalize the data obtained from the above calculations to obtain the relative sensitivity of each cognitive dimension, and map it to a preset range (e.g., 0 to 1) as the corresponding moderating sensitivity coefficient. .

[0072] During system operation, the sensitivity coefficient can be dynamically adjusted based on individual historical data to adapt to differences in cognitive response among different individuals under varying physiological loads.

[0073] Through the above process, the cognitive state results are corrected based on the physiological load index, so that the cognitive state index is adjusted accordingly when the physiological load deviates from the optimal range, while ensuring the differentiated response characteristics between different cognitive dimensions.

[0074] e) After obtaining the revised cognitive state indicators, determine their cognitive state.

[0075] Specifically, each cognitive state indicator is compared with a preset threshold or range. When an indicator falls below the corresponding threshold or shows a continuous downward trend within a preset time window, the cognitive dimension is determined to be in an abnormal or declining state.

[0076] Regarding threshold setting, the preset threshold or range can be determined in the following ways: (1) Based on individual baseline data. Specifically, collect cognitive state index data under low load or normal conditions, calculate their mean and fluctuation range, and determine the threshold by considering the acceptable degree of deviation (e.g., mean minus a certain multiple of the standard deviation) to characterize the lower bound of the individual's normal cognitive level.

[0077] (2) Based on pre-experiment or historical data. By analyzing the distribution of cognitive indicators under different task states, we can distinguish between good cognitive performance and declining performance, and select the distinction boundary as the threshold.

[0078] (3) Set the threshold by combining statistical quantile method. When a high sensitivity to abnormal states is required, select the low quantile in the historical data as the threshold.

[0079] It should be noted that the above thresholds or ranges can be dynamically adjusted based on individual circumstances and task requirements. When the task intensity or environmental complexity is high, the thresholds can be appropriately relaxed or tightened to adapt to different application scenarios. Furthermore, thresholds corresponding to different cognitive dimensions can be set separately to reflect the varying degrees of influence of different cognitive dimensions on task completion.

[0080] Then, the cognitive state indicators are determined. In one implementation, the state determination steps are as follows: The corrected cognitive state indicators are stored as time series. Data within a time window is extracted and subjected to sliding statistical processing to obtain the mean or rate of change for each time window. The statistical results are then compared with corresponding thresholds. When the window mean is below the threshold, or shows a downward trend over multiple consecutive time windows, the cognitive ability status for that dimension is determined.

[0081] f) After completing the state determination, the system generates corresponding intervention strategies based on the identified abnormal cognitive dimensions.

[0082] Specifically, such as Figure 4As shown, the system first summarizes the judgment results of each cognitive dimension to determine the set of cognitive dimensions that are currently in an abnormal or declining state. Then, according to the preset strategy mapping rules, it selects the corresponding intervention method from the strategy library and generates control instructions.

[0083] In one implementation, the policy mapping rule can be implemented through a rule table or parameter mapping. For example, different judgment results such as "decreased attention" or "decreased perception" can be mapped to corresponding sets of output adjustment parameters.

[0084] At the execution level, this intervention strategy is implemented by adjusting the way information is output, rather than changing the content of the task information itself. Specifically, it includes the following types of control parameters:

[0085] (1) Display intensity control: Adjust the brightness, contrast or color prominence of key information according to the cognitive state; (2) Display priority control: Adjust the presentation order or display level of different information elements to make key targets appear first; (3) Information density control: Screen or compress non-critical content to reduce information interference; (4) Output rhythm control: process information presentation time and stage, and break down complex information into multiple stages for gradual output.

[0086] It should be noted that the task information or environmental information originates from external task systems or other information-providing devices or methods, including but not limited to work instruction systems, environmental sensing systems, or manual input. This invention does not limit the generation process of the information content itself. This invention does not generate or semantically modify the information content itself, but rather adjusts its output method based on the already determined information.

[0087] g) Finally, the intervention strategy is implemented through a display device and / or a voice output device to dynamically adjust the way task information is output, thereby improving cognitive state and increasing task performance efficiency.

[0088] Specifically, corresponding control parameters are generated based on the selected intervention strategy, such as adjusting the rendering parameters of the display interface or the rhythm parameters of the voice output, and the control command is executed through the display device and / or the voice output device.

[0089] In addition, when multiple cognitive dimension indicators show abnormalities at the same time, different intervention strategies can be combined and coordinated according to preset priorities or weights to avoid conflicts between different regulatory measures.

[0090] Through the above methods, the process from cognitive intervention strategy to specific execution is realized, enabling the system to dynamically adjust the way information is presented without changing the content of the task information, thereby improving cognitive state and enhancing task execution efficiency.

[0091] By cyclically executing the above steps within each time window, real-time monitoring and dynamic intervention of cognitive states can be achieved.

[0092] By adopting the methods and steps of the above embodiments, a physiological load regulation mechanism is introduced, which is closer to the real state than the cognitive assessment results of a single signal; while fusing multimodal signals, the interpretability and stability of the cognitive state assessment results are guaranteed; a complete control process from assessment to intervention is constructed, realizing the dynamic regulation of cognitive state; the intervention strategy is based on information output control, which has good engineering feasibility and application adaptability.

[0093] In addition to the above implementation steps, the present invention also provides embodiments for rescue personnel to conduct status assessment and intervention in two different situations.

[0094] Example 1: Assess and intervene in the attentional state of rescue personnel during the path search task.

[0095] In this embodiment, the S1 signal acquisition and feature extraction first collects the electroencephalogram (EEG) signals and peripheral physiological signals of the rescue personnel during the mission. The power features of the theta and alpha bands, heart rate (HR) and heart rate variability (HRV) in the EEG signals are extracted and normalized as input data for subsequent processing.

[0096] In this embodiment, a fixed time window method is used for data processing. The time window length is 10 seconds, and the system completes a complete evaluation and intervention process within each time window.

[0097] S2 Cognitive Index Construction. In this embodiment, an attention state index is constructed using a linear weighting method, specifically:

[0098] Here, and These are the normalized powers of the θ and α bands, respectively. , These are preset weighting coefficients. These weighting coefficients are set based on pre-experimental data or experience, and are taken as follows: =0.6, =0.4, thus obtaining the attention status index value within the current time window.

[0099] S3 Physiological Load Index Construction. In this stage, heart rate and heart rate variability were selected as input indicators.

[0100] Physiological indicators were normalized using differences based on individual baselines. and .

[0101] In this embodiment, a linear weight combination method is used to obtain... in, , This represents the weighting coefficient. In this embodiment, the weights can be set according to the contribution of different indicators to physiological load, taking... =0.7、 =0.3, thus obtaining the current physiological load level.

[0102] S4 adjusts the cognitive state index based on physiological load. To reduce the impact of individual differences on the assessment results, a statistical standardization method based on historical data is used to process the physiological load index, resulting in a standardized physiological load value. .

[0103] In the cognitive state correction phase, a nonlinear adjustment function is used to map physiological load. In this embodiment, a Gaussian function is selected to construct the adjustment function, and the sensitivity coefficient is obtained through regression analysis using pre-experimental data. The original cognitive indicators are then corrected to obtain the attention state indicators. .

[0104] S5 Status Determination. This stage uses a threshold method based on individual baseline statistics for determination.

[0105] In this embodiment, the threshold can be determined as follows: in, and These are the mean and standard deviation of the cognitive index at the baseline state, respectively.

[0106] When the revised cognitive index meets When this happens, it is determined that the current state of attention is in a decreased or abnormal state.

[0107] S6 Intervention Strategy Generation and Execution. In this stage, a rule-table-based strategy mapping method is used to generate the intervention strategy. In this embodiment, when a decrease in attention is detected, the system selects the "enhanced display intensity" strategy and generates corresponding control parameters, including: increasing the brightness and contrast of key information; highlighting key targets; and reducing the visual weight of background information.

[0108] S7 executes in a loop. The above control parameters act on the information output process and are executed through the display device, thereby adjusting the way information is presented.

[0109] The above steps are executed cyclically within each time window to achieve real-time monitoring and dynamic intervention of attentional state.

[0110] Example 2: Assess and intervene in the multidimensional cognitive state of rescue personnel in decision-making tasks in complex environments.

[0111] In this embodiment, the system uses a sliding time window for continuous processing, with a window length of 15s and a step size of 5s, updating the cognitive state assessment results within each window.

[0112] S1 signal acquisition and feature extraction. Electroencephalogram (EEG) signals and peripheral physiological signals, including heart rate, heart rate variability, and body temperature, were collected from rescue personnel during the mission. Corresponding power spectra, event potentials, and prefrontal cortex connectivity were extracted for subsequent analysis and processing.

[0113] S2 Multidimensional Cognitive Index Construction. In this embodiment, different features and mapping methods are selected to construct cognitive indicators for different cognitive dimensions. Details are as follows:

[0114] (1) Pay attention to the indicators: An entropy-stability function mapping is established based on an empirical model: ; : No. Characteristic subvectors of frontal midline / frontal lobe theta wave power in a frequency band or time window. : No. The proportion of the theta wave power at a given time point or frequency band to the entire eigenvector. : A measure of the "dispersion" of a distribution

[0115] (2) Perception indicators: A nonlinear combination mapping of ERP amplitude and latency is employed: ; : Amplitude normalization index. : Normalized index of incubation period.

[0116] (3) Working memory index: a coupling index was constructed based on the theta power in the prefrontal cortex and the connectivity strength of brain regions.

[0117] ; This is the brain region coherence / functional connectivity matrix. Prefrontal-parietal lobe brain region pairing. The number of prefrontal-parietal lobe regions

[0118] All the above indicators were normalized and mapped to 0-1 to obtain the initial state indicators for each cognitive dimension.

[0119] S3 physiological load modeling. Heart rate, heart rate variability, and body temperature were selected as input indicators and normalized based on individual baselines.

[0120] In this embodiment, a nonlinear combination method is used to construct the physiological load index to reflect the coupling relationship between different physiological indicators, thereby obtaining the current physiological load level. Specifically:

[0121] ; This expression is used to illustrate the combined effect of high heart rate and high body temperature, as well as the reverse regulatory effect of decreased HRV on load.

[0122] S4 Cognitive State Adjustment. A statistical standardization method based on historical data was used to obtain standardized physiological load values. Based on these physiological load indicators, adjustments were made to various cognitive state indicators.

[0123] In this embodiment, physiological load is first mapped to a moderating factor, and then the cognitive indicators are modified by combining the sensitivity coefficients corresponding to each cognitive dimension, so that they can more accurately reflect the cognitive state under varying load conditions. The sensitivity coefficients are determined through regression analysis of historical data.

[0124] After correction, a multidimensional cognitive state index was obtained. , , .

[0125] S5 Status Determination. In this stage, a threshold method based on historical data distribution is used to determine the determination criteria. For each cognitive indicator, the 50th percentile of its historical data is set as the trigger threshold. The 25th percentile is used as the bottom line threshold. .

[0126] When a certain cognitive indicator is lower than When this occurs, the cognitive dimension is judged to be in a declining or abnormal state. When any indicator is below... Strengthen the strategy in time.

[0127] S6 Intervention Strategy Generation. Based on the judgment results, determine the cognitive dimension of the current abnormality and generate the corresponding intervention strategy.

[0128] In this embodiment: when attention is diminished, key information is enhanced. When working memory declines, information is output in stages. When perceptual ability declines, non-key information is suppressed.

[0129] When multiple cognitive dimensions are abnormal simultaneously, a priority ranking method based on the key capability requirements of the task is adopted to coordinate the control parameters of each strategy, and to prioritize and coordinate them according to the criticality of the task. This priority is obtained based on experimental data and expert analysis.

[0130] S7 Intervention Execution. The generated control parameters are applied to the information output process, and corresponding intervention strategies are executed through display and voice output devices to dynamically adjust the information presentation method. The system continuously updates the above process according to a sliding time window, enabling dynamic assessment and intervention of multidimensional cognitive states.

[0131] Secondly, this invention also provides a device for cognitive assessment and intervention of rescue personnel under the regulation of brain signals and physiological load, such as... Figure 5 As shown, it includes: The acquisition module 201 is used to acquire brain function signals and peripheral physiological signals of rescue personnel during the mission, and to extract brain function features and peripheral physiological features.

[0132] The construction module 202 is used to construct multiple cognitive state indicators based on the brain functional characteristics, which correspond to different cognitive dimensions in attention, perception, working memory and decision processing; and to construct a physiological load index based on the peripheral physiological characteristics to characterize the overall physiological load level of an individual.

[0133] The correction module 203 is used to map the physiological load index as a regulating factor and, in conjunction with the preset regulation sensitivity coefficients of each cognitive dimension, correct the cognitive state index to obtain the corrected cognitive state index.

[0134] The judgment module 204 is used to compare the corrected cognitive state index with a preset threshold to determine whether each cognitive dimension is in a declining or abnormal state.

[0135] The intervention module 205 is used to generate an intervention strategy based on the judgment result, and dynamically adjust the output mode of the display device and the voice output device to the task information based on the intervention strategy.

[0136] The present invention also provides a computer-readable storage medium storing a computer program that can be used to execute the above-described... Figure 1 The steps of the provided brain signal and physiological load regulation method for cognitive assessment and intervention of rescue personnel.

[0137] This invention also provides a computer device. At the hardware level, the computer device includes a processor, an internal bus, a network interface, memory, and non-volatile memory, and may also include other hardware required for various operations. The processor reads the corresponding computer program from the non-volatile memory into memory and then executes it to achieve the above-mentioned functions. Figure 1 The steps of the provided brain signal and physiological load regulation method for cognitive assessment and intervention of rescue personnel.

[0138] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0139] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0140] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0141] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0142] It should be noted that the specific embodiments described above enable those skilled in the art to more fully understand the present invention, but do not limit the present invention in any way. Therefore, although the present invention has been described in detail in this specification, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the present invention; and all technical solutions and improvements that do not depart from the spirit and scope of the present invention are covered within the protection scope of the patent of the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A method for cognitive assessment and intervention of rescue personnel under the regulation of brain signals and physiological load, characterized in that, The method includes: Collect brain function signals and peripheral physiological signals of rescue personnel during the mission, and extract brain function characteristics and peripheral physiological characteristics; Based on the aforementioned brain functional characteristics, multiple cognitive state indicators are constructed, corresponding to different cognitive dimensions in attention, perception, working memory, and decision processing. A physiological load index is constructed based on the aforementioned peripheral physiological characteristics to characterize the overall physiological load level of an individual; Using the physiological load index as a moderating factor, it is mapped to a moderating factor, and combined with the preset moderating sensitivity coefficients of each cognitive dimension, the cognitive state index is modified to obtain the modified cognitive state index. The corrected cognitive state index is compared with a preset threshold to determine whether each cognitive dimension is in a declining or abnormal state. An intervention strategy is generated based on the judgment result, and the output method of the display device and voice output device for task information is dynamically adjusted based on the intervention strategy.

2. The method according to claim 1, characterized in that, Based on the aforementioned brain functional characteristics, several cognitive state indicators are constructed, including: At least one of the following is extracted from brain functional signals: frequency band power features, event-related potential features, and brain region functional connectivity features. The corresponding cognitive dimension index values ​​are obtained by normalization and weighted combination mapping. Among them, the attention state index is constructed based on the relative relationship between theta band power and alpha band power, the perceptual response index is constructed based on the amplitude and / or latency of event-related potentials, the working memory capacity index is constructed based on the theta band power change in the prefrontal region and brain region functional connectivity features, and the decision processing index is constructed based on the functional connectivity strength or synchronicity between multiple brain regions.

3. The method according to claim 1, characterized in that, The physiological load index is constructed based on the aforementioned peripheral physiological characteristics, including: At least one of the peripheral physiological signals, including heart rate, heart rate variability, body temperature, and dehydration-related parameters, is normalized based on individual baselines, and the physiological load index is obtained by linear weighting or nonlinear combination.

4. The method according to claim 3, characterized in that, The formula for obtaining the physiological load index through normalization based on individual baselines and by linear weighting or nonlinear combination is as follows: ； in, Let be the normalized deviation of the j-th peripheral physiological characteristic relative to its individual baseline. For the corresponding weighting coefficients; or, for the normalized indicators of heart rate, heart rate variability, and body temperature. , and Nonlinear combinations are used to reflect the coupling effect of multiple physiological indicators: ； in, This is the physiological load index.

5. The method according to claim 1, characterized in that, Using the physiological load index as a moderating factor, mapping it to a moderating factor, and combining it with preset moderating sensitivity coefficients for each cognitive dimension, the cognitive state index is modified as follows: The physiological load index was statistically standardized to obtain the standardized load value. Using a nonlinear function with optimal interval characteristics to... Mapped to regulation factor Based on this regulating factor, a modified cognitive state index was obtained. : ； in, This is an indicator of the original cognitive state. Let be the moderating sensitivity coefficient of the i-th cognitive dimension, and be the moderating factor. The formula for determining it is: ； in, The location parameters are used to determine the optimal cognitive state corresponding to the standardized load.

6. The method according to claim 1, characterized in that, The preset threshold is set based on individual baseline data or historical statistical data, and is set separately for different cognitive dimensions; when the corrected cognitive state index is lower than the corresponding threshold, or shows a continuous downward trend within the preset time window, the cognitive dimension is determined to be in a declining or abnormal state.

7. The method according to claim 1, characterized in that, The intervention strategy is generated according to a preset mapping rule, which maps the judgment results of different cognitive dimensions to the corresponding information output control parameters. Specifically, when attention is reduced, the display intensity of key information is enhanced; when working memory is reduced, information is output in stages to reduce the burden of immediate information; and when perception is reduced, the output of non-key information is suppressed to reduce interference. When multiple cognitive dimensions are abnormal at the same time, the control parameters of each strategy are coordinated according to the preset priority of key task capabilities.

8. A device for cognitive assessment and intervention of rescue personnel under the regulation of brain signals and physiological load, characterized in that, The device includes: The data acquisition module is used to collect brain function signals and peripheral physiological signals of rescue personnel during the mission, and to extract brain function features and peripheral physiological features. The module is used to construct multiple cognitive state indicators based on the brain functional characteristics, corresponding to different cognitive dimensions in attention, perception, working memory, and decision processing; and to construct a physiological load index based on the peripheral physiological characteristics to characterize the individual's overall physiological load level. The correction module is used to map the physiological load index as a regulation factor and combine it with the preset regulation sensitivity coefficients of each cognitive dimension to correct the cognitive state index and obtain the corrected cognitive state index. The judgment module is used to compare the corrected cognitive state index with a preset threshold to determine whether each cognitive dimension is in a declining or abnormal state. The intervention module is used to generate intervention strategies based on the judgment results, and dynamically adjust the output mode of the display device and voice output device to the task information based on the intervention strategies.

9. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the method described in any one of claims 1 to 7.

10. A computer device, characterized in that, The method includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method described in any one of claims 1 to 7.

Images