ASD emotional breakdown risk prediction and collaborative intervention method and system, medium

Abstract

The application provides an ASD emotional breakdown risk prediction and collaborative intervention method and system, medium, and relates to the cross technical field of biomedical engineering, emotional computing and intelligent Internet of Things, through the modeling of the nonlinear trend of physiological signals by the TFT model, the application can give an early warning 3-5 minutes earlier than the traditional threshold detection method, significantly reduce the frequency and intensity of breakdown; using the adaptive denoising technology driven by IMU, the application can still maintain a high signal-to-noise ratio when ASD children appear stereotyped shaking or running, overcoming the defect that the traditional wearable device is disorderly reported when moving; the application constructs a central-peripheral physiological coupling model, improves the prediction accuracy: the application does not look at the brain or hand alone, but captures the physiological conduction mechanism of emotional loss of control by calculating the time series cross-correlation of EEG and EDA; the application realizes automatic closed loop to reduce the care burden, and provides objective basis for the follow-up analysis of the inducing factors of children by the rehabilitation therapist.

Description

Technical Field

[0001] This invention relates to the interdisciplinary fields of biomedical engineering, affective computing, and the Internet of Things (IoT), and particularly to methods, systems, and media for predicting and coordinating interventions for ASD emotional breakdown risk. Background Technology

[0002] Currently, the management of emotional breakdowns in children with ASD mainly relies on the following methods: Human observation: Relies on the subjective observation of parents or therapists to intervene after abnormal behavior (such as screaming, self-harm) is discovered.

[0003] Single-modal monitoring devices: These are primarily consumer-grade wristbands that monitor only changes in heart rate or electrical conductance of the skin (EDA) to determine stress levels.

[0004] Clinical EEG monitoring: Electroencephalogram (EEG) signals are collected using a traditional wet electrode cap to analyze neural states.

[0005] Existing technologies suffer from the following main technical problems: delayed early warning and lack of foresight: most existing products are post-event or critical detection. By the time drastic fluctuations in physiological indicators are detected, children are often already in a state of collapse, missing the optimal window for comforting and intervention (usually 3-5 minutes before the collapse).

[0006] Weak resistance to motion interference (low signal-to-noise ratio): Children with ASD often exhibit stereotyped movements (such as shaking their bodies or clapping) or ADHD. Traditional physiological signal filtering methods (such as simple bandpass filtering) cannot effectively distinguish between artifacts generated by muscle movements and genuine neurophysiological signals, resulting in an extremely high false alarm rate in daily activities.

[0007] Multimodal feature fragmentation: Existing studies usually analyze EEG (reflecting central nervous system cognitive load) or EDA (reflecting peripheral autonomic nervous system activation) separately, lacking technical means to explore the central-peripheral temporal coupling relationship between the two, and thus failing to fully reflect the physiological evolution of emotional breakdown.

[0008] Lack of closed-loop intervention (testing without intervention): Existing wearable devices only provide alerts (such as pop-up windows on mobile phones) and cannot automatically adjust in conjunction with the physical environment (such as lights and sounds). In situations where caregivers cannot immediately leave or react in time, effective closed-loop intervention cannot be formed.

[0009] Therefore, it is necessary to provide a new approach to solve the aforementioned technical problems. Summary of the Invention

[0010] To achieve the above-mentioned objectives and other advantages of the present invention, the first objective of the present invention is to provide a method for predicting and coordinating intervention for the risk of emotional breakdown in ASD, comprising the following steps: Acquired frontal electroencephalogram (EEG) signals, wrist electrodermal activity signals, and inertial motion signals from individuals with ASD using wearable devices for synchronous acquisition; The motion intensity is evaluated in real time based on the inertial motion signal. When the motion intensity exceeds a preset threshold, an adaptive filtering algorithm is activated. The inertial motion signal is used as reference noise to filter out motion artifacts from the contaminated EEG and skin conductance signals, and the cleaned physiological signal is obtained. Features are extracted from the cleaned physiological signals and coupling features are calculated. Based on the coupling features, the probability of an ASD individual experiencing an emotional breakdown event within a preset time period in the future is predicted. The predicted probability is compared with a preset dynamic risk threshold, and multi-level intervention measures are triggered based on the comparison results; wherein, the intervention measures include automatic adjustment of environmental equipment and early warning notification to guardians.

[0011] Furthermore, the adaptive filtering algorithm is a variable step size minimum mean square error algorithm, whose step size factor is dynamically adjusted according to the real-time motion intensity. The higher the motion intensity, the larger the step size factor, and the faster the filtering convergence speed.

[0012] Furthermore, the step of extracting features from the cleaned physiological signals and calculating coupling features, and predicting the probability of an ASD individual experiencing an emotional breakdown within a predetermined time period based on the coupling features, includes: Frequency band energy features characterizing cognitive load were extracted from cleaned EEG signals, and phase component features characterizing sympathetic nerve activation were extracted from cleaned electrodermal activity signals. The maximum cross-correlation coefficient between the frequency band energy characteristics and the phase component characteristics within a sliding time window is calculated as a coupling characteristic for quantifying the synchronicity of the central nervous system and the peripheral nervous system. The coupling features, temporal physiological features, and individual static features are input into the prediction model, which outputs the probability that the individual with ASD will experience an emotional breakdown within a preset time period in the future.

[0013] Furthermore, the prediction model is a temporal fusion attention network, which contains a variable selection network; the variable selection network dynamically adjusts the weights of EEG signal features, skin conductance signal features, and coupling features in the model prediction based on real-time calculated signal quality indicators.

[0014] Furthermore, the multi-level intervention measures include at least: Level 1 warning: When the predicted probability is in the first risk range, the risk trend information will only be displayed on the guardian's user terminal; Second-level environmental regulation: When the predicted probability is within the second risk range, environmental devices are automatically controlled via IoT protocol to adjust environmental sensory stimuli in a smooth and gradual manner; wherein, the second risk range is higher than the first risk range; Level 3 Emergency Alarm: When the predicted probability is in the third risk range, a strong reminder device is triggered for the guardian, and an emergency intervention suggestion is pushed; wherein, the third risk range is higher than the second risk range.

[0015] Furthermore, the second-level environmental conditioning includes: controlling the intelligent lighting device to linearly reduce the color temperature and brightness within a specified time period, and / or controlling the audio playback device to play pink noise with a masking and soothing effect.

[0016] Furthermore, prior to the step of acquiring the frontal EEG signals, wrist skin conductance signals, and inertial motion signals of the ASD individual simultaneously collected using a wearable device, a baseline learning step is also included: During the initial use of the system, the baseline of physiological signals of the ASD individual in a calm state is recorded, and the dynamic risk threshold is set based on the statistical distribution of the baseline.

[0017] The second objective of this invention is to provide a system for predicting and coordinating intervention for the risk of emotional breakdown in individuals with ASD, which utilizes the aforementioned method and includes: The signal acquisition module includes a flexible forehead patch integrating EEG electrodes and a smart wristband integrating electrodermal activity electrodes and an inertial measurement unit. Signal processing and prediction module: used to perform real-time assessment of motion intensity based on the inertial motion signal; when the motion intensity exceeds a preset threshold, to activate an adaptive filtering algorithm, using the inertial motion signal as reference noise, to filter out motion artifacts from contaminated EEG and EEG activity signals to obtain cleaned physiological signals; and to extract features from the cleaned physiological signals and calculate coupling features, and to predict the probability of an ASD individual experiencing an emotional breakdown within a preset time period based on the coupling features, including a processor and stored algorithm program; Collaborative intervention execution module: This module compares the predicted probability with a preset dynamic risk threshold and triggers multi-level intervention measures based on the comparison results, including IoT gateways and controlled environmental control equipment, and early warning notification equipment.

[0018] Furthermore, the EEG electrodes on the flexible forehead patch are ultrathin dry tattoo electrodes based on nano-silver or conductive polymer materials; The flexible forehead patch in the signal acquisition module and the smart wristband are networked via Bluetooth Low Energy, and a unified timestamp ensures millisecond-level synchronization of multimodal signals.

[0019] A third object of the present invention is to provide a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the above-described method.

[0020] Compared with the prior art, the beneficial effects of the present invention are: This invention provides a method, system, and medium for predicting and coordinating interventions for ASD emotional breakdown risk, enabling early warning and securing the golden intervention time: By modeling the nonlinear trend of physiological signals using a TFT model, this invention can issue an early warning 3-5 minutes earlier than traditional threshold detection methods. This allows caregivers or the system sufficient time to intervene before the breakdown fully erupts, thereby significantly reducing the frequency and intensity of breakdowns.

[0021] This invention solves the signal failure problem in high-dynamic scenarios: by utilizing IMU-driven adaptive denoising technology, this invention can maintain a high signal-to-noise ratio even when children with ASD exhibit stereotyped shaking or running. This overcomes the shortcomings of traditional wearable devices that issue false alarms with slight movement, greatly improving the system's usability in real-life scenarios.

[0022] This invention constructs a central-peripheral physiological coupling model, improving prediction accuracy: instead of focusing solely on the brain or hand, this invention captures the physiological transmission mechanism of emotional dysregulation by calculating the temporal cross-correlation between EEG and EDA. Experiments show that the prediction accuracy and recall of the feature fusion approach are significantly better than the single-modal approach.

[0023] This invention achieves an automated closed-loop system to reduce the burden of care: it directly translates prediction results into adjustments to the physical environment (light, sound), bridging the gap between monitoring and intervention. This non-invasive environmental intervention is more suitable for children with ASD who are sensitive to their senses, while also reducing the psychological burden on parents who need to monitor data around the clock.

[0024] The model provided by this invention is interpretable: the attention mechanism of the TFT model can output which feature dominated the alarm, whether it was due to excessive cognitive load or physiological irritation caused by external fright, providing objective evidence for rehabilitation therapists to analyze the inducing factors in children.

[0025] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it according to the contents of the specification, the preferred embodiments of the present invention are described in detail below with reference to the accompanying drawings. Specific embodiments of the present invention are given in detail below with reference to the accompanying drawings. Attached Figure Description

[0026] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings: Figure 1 This is a diagram illustrating the overall architecture of a system for predicting and collaboratively intervening in the risk of emotional breakdown in individuals with ASD (Acute Disorder). Figure 2 Diagram of the flexible EEG acquisition patch on the forehead; Figure 3 Layout diagram of sensors for a smart wristband; Figure 4 A flowchart for predicting and collaboratively intervening in the risk of emotional breakdown in ASD; Figure 5 Schematic diagram of IMU-driven adaptive noise reduction; Figure 6 Flowchart for spatiotemporal coupling feature extraction and prediction; Figure 7 This is a structural diagram of the Temporal Fusion Attention Network (TFT) model. Figure 8 This is a schematic diagram of central-peripheral coupling feature extraction; Figure 9 A flowchart of a closed-loop, tiered intervention strategy; Figure 10 A flowchart of multi-level intervention measures; Figure 11 This is a schematic diagram of a computer device. Figure 12 This is a schematic diagram of a computer-readable storage medium. Detailed Implementation

[0027] The present invention will now be further described with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. It should be noted that, without conflict, the various embodiments or technical features described below can be arbitrarily combined to form new embodiments.

[0028] Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without inventive effort are within the scope of protection of this invention.

[0029] The drawing numbers in this application are only used to distinguish the steps in the scheme and are not used to limit the execution order of the steps. The specific execution order is as described in the specification.

[0030] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention.

[0031] This invention provides a system and method for children with autism spectrum disorder (ASD), which utilizes wearable sensors (EEG, EDA, IMU) for multimodal physiological signal acquisition, deep learning-based temporal model-based risk prediction, and coordinated with environmental physical devices for automated closed-loop intervention. The specific solution is as follows: Example 1 A system for predicting and coordinating the risk of emotional breakdown in people with ASD, such as Figure 1 As shown, it includes: The signal acquisition module includes a flexible forehead patch integrating EEG electrodes and a smart wristband integrating electrodermal activity electrodes and an inertial measurement unit. Signal processing and prediction module: used to perform real-time assessment of motion intensity based on the inertial motion signal; when the motion intensity exceeds a preset threshold, to activate an adaptive filtering algorithm, using the inertial motion signal as reference noise, to filter out motion artifacts from contaminated EEG and EEG activity signals to obtain cleaned physiological signals; and to extract features from the cleaned physiological signals and calculate coupling features, and to predict the probability of an ASD individual experiencing an emotional breakdown within a preset time period based on the coupling features, including a processor and stored algorithm program; Collaborative intervention execution module: This module compares the predicted probability with a preset dynamic risk threshold and triggers multi-level intervention measures based on the comparison results, including IoT gateways and controlled environmental control equipment, and early warning notification equipment.

[0032] To address the issues of tactile sensitivity and poor compliance in children with ASD, the EEG electrodes on the flexible forehead patch are ultra-thin dry tattoo electrodes based on nano-silver or conductive polymer materials. Specifically, the flexible forehead patch integrates Fp1 / Fp2 channel EEG electrodes, and the smart wristband integrates EDA electrodes and a 6-axis IMU inertial sensor. This embodiment uses nano-silver / conductive polymer tattoo electrodes instead of traditional wet electrodes, improving wearing comfort and utilizing the IMU to capture motion posture data in real time.

[0033] Forehead EEG acquisition module such as Figure 2 As shown, the electrode is an ultra-thin tattoo electrode based on PEDOT:PSS (poly(3,4-ethylenedioxythiophene):polystyrene sulfonate) conductive polymer. This electrode is less than 100 μm thick, breathable, and is attached to the Fp1 and Fp2 positions on the forehead using water transfer printing technology. Signals are extracted via a disposable flexible FPC (flexible printed circuit board). The analog front-end (AFE) uses a low-noise biopotential acquisition chip (such as ADS1299), with a sampling rate set to 250 Hz and a gain set to 24. Specifically, the signal characteristics are focused on acquiring signals in the 0.5-45 Hz frequency band, covering delta, theta, alpha, beta, and low-frequency gamma waves.

[0034] Wrist physiological and kinematic data acquisition module, such as Figure 3 As shown, the electrical conductance-dermal (EDA) sensor is located on the ventral side of the inner wristband, employing two 8mm diameter dry stainless steel electrodes spaced 15mm apart, with a low DC voltage (0.5V) applied to measure skin conductance (GSR). The inertial measurement unit (IMU) integrates a 6-axis sensor (such as the MPU6050), including a 3-axis accelerometer and a 3-axis gyroscope, with a sampling rate set to 50Hz, to capture stereotyped hand movements such as clapping and shaking.

[0035] To achieve data transmission and synchronization, the frontal EEG acquisition module and the wrist physiological and motor acquisition module are networked via Bluetooth Low Energy (BLE 5.0). The system uses the wrist module as the master node, and adds a unified Unix timestamp to the data packets to ensure millisecond-level alignment of EEG, EDA, and IMU data before transmitting them to edge computing terminals, such as smartphones or edge gateways.

[0036] For a detailed description of the ASD emotional breakdown risk prediction and collaborative intervention method applied by this system, please refer to the corresponding description in the following method embodiments, which will not be repeated here.

[0037] Example 2 A method for predicting and collaboratively intervening in the risk of ASD emotional breakdown is applied to the aforementioned system. For a detailed description of the system, please refer to the corresponding description in the system embodiments described above; it will not be repeated here. Figure 4 As shown, the method includes the following steps: S100: Acquire frontal electroencephalogram (EEG) signals, wrist skin electrical activity signals, and inertial motion signals of individuals with ASD, collected synchronously using wearable devices. S200. Based on the inertial motion signal, the motion intensity is evaluated in real time. When the motion intensity exceeds a preset threshold, an adaptive filtering algorithm is activated. The inertial motion signal is used as reference noise to filter out motion artifacts from the contaminated EEG signal and skin conductance signal, and the cleaned physiological signal is obtained. Preferably, the adaptive filtering algorithm is a variable step size minimum mean square error algorithm, whose step size factor is dynamically adjusted according to the real-time motion intensity. The higher the motion intensity, the larger the step size factor, and the faster the filtering convergence speed.

[0038] To address the technical challenge of signal distortion caused by hyperactivity in children with ASD, this embodiment employs a signal cleaning strategy, unlike traditional methods that directly remove bad segments. For example... Figure 5 As shown, the specific steps include: Motion intensity determination: Real-time calculation of the composite vector magnitude of IMU triaxial acceleration Set the motion threshold. Value (e.g., 1.2g). If The signal is determined to be in a static / micro-motion state, and only a 0.5-45Hz bandpass filter is applied to the EEG signal. If... If the state is determined to be a high motion artifact state, the adaptive LMS (Least Mean Square) filtering program is started.

[0039] Adaptive LMS filtering process: This process transforms the original noisy EEG signal... As the main input, the IMU's acceleration signal is mapped to a reference noise signal. .

[0040] Let the filter order be... The weight vector is Calculate the filter output (estimated motion artifacts): Calculate the error signal (i.e., the cleaned EEG signal): Update weights: ; Step size factor It is not a constant, but varies with the intensity of motion. Dynamic adjustment. The more intense the exercise, the better. Increase the value to accelerate the convergence speed.

[0041] This embodiment introduces the IMU signal as a reference noise source. The system calculates the acceleration amplitude of the IMU in real time. When high-intensity motion is detected (exceeding a preset threshold), the variable step size minimum mean square error (LMS) adaptive filtering algorithm is activated to subtract the motion component captured by the IMU from the contaminated EEG / EDA signal, thereby filtering out motion artifacts while preserving the characteristics of the neural signal.

[0042] S300. Extract features from the cleaned physiological signals and calculate coupling features, and predict the probability of an ASD individual experiencing an emotional breakdown within a preset time period based on the coupling features. In order to uncover the physiological evolutionary patterns of emotional breakdown from the data, such as Figures 6-8 As shown, the steps of extracting features from the cleaned physiological signals and calculating coupling features, and predicting the probability of an ASD individual experiencing an emotional breakdown within a preset time period based on the coupling features, include: S310. Extract the frequency band energy features representing cognitive load from the cleaned EEG signals, and extract the phase component features representing sympathetic nerve activation from the cleaned skin conductance activity signals. For example, after cleaning Perform short-time Fourier transform (STFT) to extract Relative power spectral energy in the band (13-30Hz) . A sustained increase in the wave indicates cognitive overload or anxiety buildup.

[0043] The EDA signal is decomposed into Tonic (slow baseline) and Phasic (fast pulse) components using the convex optimization algorithm (CVXEDA), and the peak frequency and amplitude of the Phasic component are extracted. .

[0044] S320. Calculate the frequency band energy characteristics. With the phase component characteristics The maximum cross-correlation coefficient within a sliding time window (e.g., 60 seconds) serves as a coupling characteristic for quantifying the synchronicity of responses between the central and peripheral nervous systems; the cross-correlation function formula is: in, This refers to the physiological transmission delay. This feature quantifies the degree of synchronization between brain anxiety and physical stress, and is a precursor indicator for predicting breakdown.

[0045] S330. Input the coupling features, temporal physiological features and individual static features into the prediction model, and output the probability that the ASD individual will experience an emotional breakdown event within a preset time period in the future. Preferably, the prediction model is a temporal fusion attention network, which includes a variable selection network; the variable selection network dynamically adjusts the weights of EEG signal features, skin conductance signal features, and coupling features in the model prediction based on real-time calculated signal quality indicators.

[0046] Specifically, a Temporal Fusion Transformer is used for multi-step prediction. Its input layer includes static variables: child's age, gender, and sensory sensitivity type (auditory / visual sensitivity); dynamic known variables: current time and whether it is a medication window; and dynamic observed variables. Trend, EDA Phasic peak, SCI coupling index, IMU motion intensity.

[0047] The TFT's built-in Variable Selection Network (VSN) layer automatically weights features based on the current Signal Quality Index (SQI). For example, when the IMU detects violent motion that reduces the reliability of the EEG, the VSN automatically reduces the weight of the EEG features and increases the weight of the EDA features to ensure prediction robustness.

[0048] Finally, the output layer outputs the probability distribution (P10, P50, P90 quantiles) of an emotional breakdown occurring in the next N minutes (e.g., 5 minutes).

[0049] This embodiment extracts the beta band energy trend of EEG and the peak value of the phase component of EDA, and calculates the maximum cross-correlation coefficient between the two within a sliding window to quantify the degree of synchronization between cognitive load and sympathetic nerve activation. A temporal fusion attention network (TFT) is used, with input coupling features, raw temporal data, and static variables. The model utilizes a variable selection network (VSN) to dynamically adjust the weights of different modalities based on the current signal quality (SQI), outputting the probability value of emotional breakdown occurring within the next N minutes.

[0050] S400. Compare the predicted probability with a preset dynamic risk threshold, and trigger multi-level intervention measures based on the comparison result; wherein, the intervention measures include automatic adjustment of environmental equipment and early warning notification to guardians.

[0051] To translate the predicted results into physical-world interventions and form a closed loop, a baseline learning step is also included: during the initial use of the system, the baseline physiological signals of the individual with ASD in a resting state are recorded, and the dynamic risk threshold is set based on the statistical distribution of this baseline. For example, during the first week of system use, the system is in learning mode, recording the child's physiological baseline in a resting state and setting the dynamic risk threshold. .

[0052] In some embodiments, such as Figure 9 , Figure 10 As shown, the multi-level intervention measures include at least: S410, Level 1 Warning: When the predicted probability is in the first risk range, the risk trend information is only displayed on the guardian's user terminal; S420, Second-level environmental adjustment: When the predicted probability is in the second risk range, the environmental equipment is automatically controlled through the Internet of Things protocol to adjust the environmental sensory stimulation in a smooth and gradual manner; wherein, the second risk range is higher than the first risk range; Preferably, the second-level environmental adjustment includes: controlling the intelligent lighting device to linearly reduce the color temperature and brightness within a specified time, and / or controlling the audio playback device to play pink noise with a masking and soothing effect.

[0053] S430, Level 3 Emergency Alarm: When the predicted probability is in the third risk range, a strong reminder device is triggered for the guardian, and an emergency intervention suggestion is pushed; wherein, the third risk range is higher than the second risk range.

[0054] Specifically, the system updates the probability of a crash over the next 5 minutes every 10 seconds. Level 1 intervention ( (Warning period): A yellow warning icon will only be displayed on the guardian's APP to remind the guardian to pay attention, but the guardian will not intervene in the child's activities.

[0055] Secondary intervention ( (Automatic Adjustment Period): Action A (Light Environment): Sends a command to the smart bulb to linearly adjust the color temperature within 45 seconds: gradually changing from the current value (e.g., 6000K cool white) to 2700K (warm yellow), and reducing the brightness to 40%. This avoids sudden changes in light that could cause fright. Action B (Sound Environment): Triggers the smart speaker to play pink noise or a pre-recorded soothing voice from parents, with the volume fading from 20dB to 50dB to remove sensory stimuli from the environment and mask sudden noises with white noise.

[0056] Level 3 intervention ( Emergency response period: The wristband worn by the guardian emits a continuous strong vibration, the APP pops up a full screen window and emits a high-frequency prompt sound to call for immediate human intervention to prevent self-harm.

[0057] when If the reading remains below 0.4 for three consecutive minutes, the system determines that the crisis has passed. The IoT device will not immediately return to its original state, but will instead slowly restore the lights and sounds to normal levels in 5-minute cycles to prevent secondary irritation.

[0058] This embodiment triggers tiered interventions based on predicted probability and individual baselines, including Level 1 warning, which is the trend displayed on the APP; Level 2 intervention, which is to control the smart lights to smoothly transition color temperature and brightness through the IoT protocol and control the speaker to play pink noise; and Level 3 intervention, which is to trigger strong vibration of the guardian's wristband and push suggestions.

[0059] Example 3 A computer device 500, such as Figure 11 As shown, the system includes a memory 510, a processor 520, and a computer program 530 stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of a method for predicting and coordinating the risk of ASD emotional breakdown. For a detailed description of the method, please refer to the corresponding description in the above method embodiments; it will not be repeated here.

[0060] Example 4 A computer-readable storage medium, such as Figure 12 As shown, a computer program is stored thereon. When executed by a processor, the computer program implements the steps of a method for predicting and coordinating the risk of ASD emotional breakdown. For a detailed description of the method, please refer to the corresponding description in the above method embodiments, which will not be repeated here.

[0061] The number of devices and processing scale described herein are for the purpose of simplifying the description of the invention. Applications, modifications, and variations of the invention will be readily apparent to those skilled in the art.

[0062] Although embodiments of the present invention have been disclosed above, they are not limited to the applications listed in the specification and embodiments. They can be applied to various fields suitable for the present invention. For those skilled in the art, other modifications can be easily made. Therefore, without departing from the general concept defined by the claims and their equivalents, the present invention is not limited to the specific details and illustrations shown and described herein.

[0063] The apparatus, computer device, and non-volatile computer storage medium and method provided in the embodiments of this specification are corresponding. Therefore, the apparatus, computer device, and non-volatile computer storage medium also have similar beneficial technical effects as the corresponding method. Since the beneficial technical effects of the method have been described in detail above, the beneficial technical effects of the corresponding apparatus, computer device, and non-volatile computer storage medium will not be repeated here.

[0064] Those skilled in the art will also know that, besides implementing the controller in the form of purely computer-readable program code, the same functions can be achieved by logically programming the method steps, making the controller take the form of logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers (PLCs), and embedded microcontrollers. Therefore, such a controller can be considered a hardware component, and the devices included within it for implementing various functions can also be considered structures within that hardware component. Alternatively, the devices for implementing various functions can be considered as both software units implementing the method and structures within a hardware component.

[0065] The systems, apparatuses, or units described in the above embodiments can be implemented by computer chips or physical entities, or by products with certain functions. For ease of description, the above apparatuses are described separately as various units based on their functions. Of course, when implementing one or more embodiments of this specification, the functions of each unit can be implemented in one or more software and / or hardware.

[0066] Those skilled in the art will understand that the embodiments of this specification can be provided as methods, systems, or computer program products. Therefore, the embodiments of this specification can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the embodiments of this specification can take the form of a computer program product implemented 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.

[0067] This specification is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this specification. It will be understood that each block of the flowchart illustrations and / or block diagrams, and 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 illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0068] 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.

[0069] 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.

[0070] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0071] This specification may be described in the general context of computer-executable instructions, such as program units, that are executed by a computer. Generally, program units include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This specification may also be practiced in distributed computing environments, where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program units may reside in local and remote computer storage media, including storage devices.

[0072] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0073] The above description is merely an embodiment of this specification and is not intended to limit the scope of one or more embodiments of this specification. Various modifications and variations can be made to one or more embodiments of this specification by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of one or more embodiments of this specification should be included within the scope of the claims of one or more embodiments of this specification.

Claims

1. A method for predicting and coordinating intervention for the risk of emotional breakdown in ASD, characterized in that, Includes the following steps: Acquired frontal electroencephalogram (EEG) signals, wrist electrodermal activity signals, and inertial motion signals from individuals with ASD using wearable devices for synchronous acquisition; The motion intensity is evaluated in real time based on the inertial motion signal. When the motion intensity exceeds a preset threshold, an adaptive filtering algorithm is activated. The inertial motion signal is used as reference noise to filter out motion artifacts from the contaminated EEG and skin conductance signals, and the cleaned physiological signal is obtained. Features are extracted from the cleaned physiological signals and coupling features are calculated. Based on the coupling features, the probability of an ASD individual experiencing an emotional breakdown event within a preset time period in the future is predicted. The predicted probability is compared with a preset dynamic risk threshold, and multi-level intervention measures are triggered based on the comparison results; wherein, the intervention measures include automatic adjustment of environmental equipment and early warning notification to guardians.

2. The method for predicting and coordinating intervention for ASD emotional breakdown risk as described in claim 1, characterized in that, The adaptive filtering algorithm is a variable step size minimum mean square error algorithm. Its step size factor is dynamically adjusted according to the real-time motion intensity. The higher the motion intensity, the larger the step size factor, and the faster the filtering convergence speed.

3. The method for predicting and coordinating intervention for ASD emotional breakdown risk as described in claim 1, characterized in that, The steps of extracting features from the cleaned physiological signals and calculating coupling features, and predicting the probability of an ASD individual experiencing an emotional breakdown within a preset time period based on the coupling features, include: Frequency band energy features characterizing cognitive load were extracted from cleaned EEG signals, and phase component features characterizing sympathetic nerve activation were extracted from cleaned electrodermal activity signals. The maximum cross-correlation coefficient between the frequency band energy characteristics and the phase component characteristics within a sliding time window is calculated as a coupling characteristic for quantifying the synchronicity of the central nervous system and the peripheral nervous system. The coupling features, temporal physiological features, and individual static features are input into the prediction model, which outputs the probability that the individual with ASD will experience an emotional breakdown within a preset time period in the future.

4. The method for predicting and coordinating intervention for ASD emotional breakdown risk as described in claim 3, characterized in that, The prediction model is a temporal fusion attention network, which contains a variable selection network. The variable selection network dynamically adjusts the weights of EEG signal features, skin conductance signal features, and coupling features in the model prediction based on real-time calculated signal quality indicators.

5. The method for predicting and coordinating intervention for ASD emotional breakdown risk as described in claim 1, characterized in that, The multi-level intervention measures include at least: Level 1 warning: When the predicted probability is in the first risk range, the risk trend information will only be displayed on the guardian's user terminal; Second-level environmental regulation: When the predicted probability is within the second risk range, environmental devices are automatically controlled via IoT protocol to adjust environmental sensory stimuli in a smooth and gradual manner; wherein, the second risk range is higher than the first risk range; Level 3 Emergency Alarm: When the predicted probability is in the third risk range, a strong reminder device is triggered for the guardian, and an emergency intervention suggestion is pushed; wherein, the third risk range is higher than the second risk range.

6. The method for predicting and coordinating intervention for ASD emotional breakdown risk as described in claim 5, characterized in that, The second level of environmental conditioning includes: controlling intelligent lighting devices to linearly reduce color temperature and brightness within a specified time period, and / or controlling audio playback devices to play pink noise with masking and soothing effects.

7. The method for predicting and coordinating intervention for ASD emotional breakdown risk as described in claim 1, characterized in that, Before the step of acquiring the frontal EEG signals, wrist skin conductance signals, and inertial motion signals of the ASD individual synchronously collected using a wearable device, a baseline learning step is also included: During the initial use of the system, the baseline of physiological signals of the ASD individual in a calm state is recorded, and the dynamic risk threshold is set based on the statistical distribution of the baseline.

8. A system for predicting and coordinating intervention for the risk of emotional breakdown in patients with acute mental illness (ASD), using the method described in any one of claims 1 to 7, characterized in that, include: The signal acquisition module includes a flexible forehead patch integrating EEG electrodes and a smart wristband integrating electrodermal activity electrodes and an inertial measurement unit. Signal processing and prediction module: used to perform real-time evaluation of motion intensity based on the inertial motion signal. When the motion intensity exceeds a preset threshold, an adaptive filtering algorithm is activated to use the inertial motion signal as reference noise to filter out motion artifacts from the contaminated EEG and skin conductance signals, and obtain the cleaned physiological signal. The algorithm program includes a processor and storage. It also includes the steps of extracting features from the cleaned physiological signals and calculating coupling features, and predicting the probability of an ASD individual experiencing an emotional breakdown within a preset time period based on the coupling features. Collaborative intervention execution module: This module compares the predicted probability with a preset dynamic risk threshold and triggers multi-level intervention measures based on the comparison results, including IoT gateways and controlled environmental control equipment, and early warning notification equipment.

9. The ASD emotional breakdown risk prediction and collaborative intervention system as described in claim 8, characterized in that, The EEG electrodes on the flexible forehead patch are ultra-thin dry tattoo electrodes based on nano-silver or conductive polymer materials. The flexible forehead patch in the signal acquisition module and the smart wristband are networked via Bluetooth Low Energy, and a unified timestamp ensures millisecond-level synchronization of multimodal signals.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 7.

Images