399 results about "Eeg data" patented technology

The invention discloses a sparse dynamic graph convolution-based electroencephalogram signal decoding method and system. The method comprises the following steps of: acquiring a multi-channel electroencephalogram signal and preprocessing the multi-channel electroencephalogram signal; performing multi-band filtering on each channel signal, extracting statistical characteristics on each band signal, calculating a covariance matrix of a task electroencephalogram signal, constructing image electroencephalogram data by taking an electroencephalogram channel as an image node, the multi-band spliced statistical characteristics on the channel as a node feature vector, and the covariance matrix between the channels as an adjacent matrix; finally, a dynamic graph convolutional neural network model is constructed, the model constructs a graph convolutional neural network based on an autoregression moving average filter, graph electroencephalogram data is used as input, the category of electroencephalogram signals is used as output, an adjacency matrix is dynamically generated in combination with bilinear mapping, fuzzy label learning and sparse constraint are added to improve the decoding capacity of the model, and the dynamic graph convolutional neural network model is obtained. And the frequency domain response capability and robustness of the model to the graph structure are enhanced.

The invention discloses an attention deficit hyperactivity disorder recognition system and recognition method, and belongs to the technical field of attention deficit hyperactivity disorder. The data processing module is used for carrying out preprocessing and feature extraction on the acquired electroencephalogram data; a multi-source feature fusion mechanism is firstly used for the extracted original feature data, and then a data enhancement strategy is applied; a multi-source fusion feedback regulation network model is constructed, wherein the model is of a CNN-GRU parallel modeling structure; the training module is used for inputting the enhanced data into a model for training, key hyper-parameters are dynamically adjusted by a performance feedback adjusting mechanism in the training process, and the performance feedback adjusting mechanism is used for dynamically adjusting key training parameters according to the performance of the verification set; and the classification module is used for classifying to-be-detected samples through the trained multi-source fusion feedback regulation network model and outputting a final recognition result. The ADHD electroencephalogram recognition method effectively improves the accuracy, robustness and generalization performance of ADHD electroencephalogram recognition.

The invention relates to the technical field of artificial intelligence and biomedical sensing, and discloses a cognitive disorder risk identification device and method integrating multi-modal physiological data, and the device comprises an EEG module, an eye movement module, an fNIRS module, an edge calculation module and an acceleration sensor; the EEG module collects EEG data, the eye movement module collects eye movement eye-tracing data, and the fNIRS module collects near infrared spectrum fNIRS data and inputs the data to the edge calculation module; the edge calculation module runs the multi-modal space-time attention fusion model to output a cognitive impairment risk assessment result by using the multi-modal space-time attention fusion model; the acceleration sensor operates based on an operation dynamic filtering algorithm of a motion accelerometer to inhibit signal drift caused by head motion. According to the invention, through the modularized head-mounted multi-mode edge device, the physiological data is collected and edge end processing and cognitive disorder recognition and screening are carried out in combination with the edge calculation module, so that early recognition and screening of neurodegenerative diseases can be conveniently and effectively carried out at low cost.

The invention discloses a multi-path deep neural network-based ratchet wave superposition ripple detection method. The method comprises the following steps: 1, collecting electroencephalogram data of a real BECTS patient, preprocessing and slicing; 2, performing feature extraction and dynamic spectrum weighted fusion on the processed low-frequency and high-frequency electroencephalogram signals by using a deep neural network, and outputting a high-order feature map; 3, a multi-head attention mechanism and a coding layer are introduced, dynamic weight distribution and global dependency relationship modeling of electroencephalogram signal features are achieved, and a classification result is output; 4, dynamically processing a sample imbalance problem by adopting weighted cross entropy loss and differentiable AUC loss, and constructing and optimizing a classification model; 5, the optimized and adjusted RonS detection model is applied, and electroencephalogram RonS detection is achieved on the continuous EEG signals. According to the method, the multi-kernel causal convolution module is constructed to extract the time domain features of the ratchet waves, and the PSD weighting mechanism is combined to fuse the multi-band information, so that the detection precision and generalization ability of the RonS in the complex electroencephalogram signals are remarkably improved.

The invention discloses an intelligent asthenopia control method and system based on eye movement and electroencephalogram data. The method comprises the steps that eye movement time sequence data and electroencephalogram rhythm signals of a user are synchronously obtained through a multi-mode sensing unit; inputting the eye movement time sequence data and the electroencephalogram rhythm signal into a multi-modal fusion decision model, and generating a fatigue level through feature weighting based on an attention mechanism; at least one intervention mode is dynamically selected according to the asthenopia level, and the intervention modes comprise the first mode, the second mode and the third mode; in the first mode, a display interface adjusting instruction containing natural light simulation parameters is generated, and the display interface adjusting instruction is integrated with a vegetation color system compensation algorithm; and mode 2: starting a visual training program with ecological images, and generating a dynamic guide mark simulating natural motion on an interface. According to the invention, through a data-environment-physiology multi-dimensional collaborative innovative architecture, the spanning of asthenopia regulation from passive response to ecological active restoration is realized.

The invention discloses an electroencephalogram signal processing method and system based on motion artifact prediction. The method comprises the following steps: acquiring experimental electroencephalogram data when a testee executes a motion imagination task; discrete wavelet transform is carried out on experimental electroencephalogram data, and signals are decomposed into low-frequency components and high-frequency components through a low-pass filter and a high-pass filter; inputting the low-frequency component into a pre-constructed and trained ARIMA model, and predicting to obtain a linear artifact; inputting the high-frequency component into a pre-constructed and trained XGBoost regression model, and predicting to obtain a nonlinear artifact; combining the linear artifacts and the nonlinear artifacts to generate a complete artifact prediction signal; real electroencephalogram signals are separated through difference value calculation of the experimental electroencephalogram data and the artifact prediction signals. According to the method, the time sequence change of the motion artifacts is predicted by utilizing the ARIMA model, and the nonlinear artifact features are captured in combination with the XGBoost regression model, so that the motion artifacts can be effectively removed, and purer electroencephalogram signals can be recovered.

The invention relates to an instant denoising method for electroencephalogram signals of students in a classroom scene. The method comprises the following steps: based on electroencephalogram acquisition equipment and a Daisy expansion board, acquiring electroencephalogram signal data of a student through a dry electrode device; and carrying out sliding mean filtering on the electroencephalogram signal data to remove baseline drift, eliminating noise by adopting a band-pass filter to obtain preprocessed electroencephalogram signal data, and inputting the data into a lightweight neural network model to carry out channel-by-channel denoising processing to obtain a denoised electroencephalogram signal. The model comprises a multi-scale feature extraction layer, a time sequence modeling module and a residual connection structure. According to the method, through lightweight neural network design such as multi-scale feature extraction, time sequence modeling and a residual connection structure, the denoising effect is improved, the real-time performance and high efficiency of electroencephalogram denoising processing are guaranteed, and an electroencephalogram data basis with a high signal-to-noise ratio is provided for subsequent teaching quality evaluation.

The invention relates to an electroencephalogram signal classification method based on multi-scale causal convolution and KAN attention. The electroencephalogram signal classification method comprises the steps of 1, EEG data collection and preprocessing; 2, EEG data enhancement; and 3, motor imagery task classification based on EEG signals. According to the method, the spatial-temporal characteristics of the EEG signals under different frequencies can be effectively extracted, the multi-scale nonlinear characteristics are efficiently fused and weighted, and the problem that nonlinear fitting of the EEG signals in a motor imagery task is insufficient is solved, so that more accurate and reliable motor imagery classification is realized.

The invention relates to the technical field of determination of nerve regulation and control precise stimulation targets, and discloses an individualized multi-frequency regulation and control stimulation system for Rett regulation and control. The individualized multi-frequency regulation and control stimulation system comprises a data acquisition module, a structure and function brain image analysis module and a multi-frequency electroencephalogram self-adaptive regulation and control module, the data acquisition module is used for acquiring resting state functional magnetic resonance imaging data rs-fMRI, structural magnetic resonance imaging data sMRI and EEG data, and the structural and functional brain image analysis module is used for identifying an abnormal brain area of an RTT patient and determining an optimal stimulation target. The multi-frequency electroencephalogram self-adaptive regulation and control module comprises two sub-modules including an electroencephalogram state recognition module and an FUS self-adaptive stimulation module, and the electroencephalogram state recognition module is used for analyzing EEG signals, extracting frequency band power change and instantaneous phase information corresponding to a target brain region and providing real-time feedback signals for self-adaptive regulation and control. According to the system, the target spot is accurately determined, and the individualized accurate stimulation target spot is determined by combining the structure and function image and the disease exposure model.

The invention discloses a cross-subject electroencephalogram emotion recognition method and system based on dynamic domain invariant representation decoupling and recombination, and belongs to the technical field of artificial intelligence. The invention provides a non-personalized decoupling and recombination framework for cross-subject electroencephalogram emotion recognition, and aims to separate emotion-related individual invariant features from cross-subject individual invariant features through complex electroencephalogram signal characterization obtained through dynamic decoupling, so that individual differences are eliminated while emotion classification performance is guaranteed. Specifically, the method comprises the following steps: carrying out original EEG data analysis and preprocessing by using MATLAB and Python MNE libraries; frequency spectrum and space features of EEG signals are extracted through a multi-channel frequency spectrum space self-attention mechanism module, and capture of emotional features is enhanced in combination with a self-attention mechanism and a cross-attention mechanism; a joint distribution alignment method based on a category prototype is adopted, and decoupled feature distribution is optimized, so that invariant features in subjects and invariant features among subjects have higher distinction degree in emotion classification; the optimized decoupling features are recombined through a linear network, the two features are coordinated to perform more sufficient emotion representation extraction, and emotion classification is performed through a multi-layer perceptron. According to the method, excellent cross-subject emotion recognition performance is obtained on multiple data sets, and the emotion recognition rate of the electroencephalogram signals in a cross-subject scene can be effectively improved.

The invention provides a brain-computer interface-based TMS and VR collaborative rehabilitation training system, which is characterized in that a multi-mode collaborative positioning module is used for collecting VR interaction data and electroencephalogram data and determining the optimal stimulation state of a subject; the motion intention analysis module is used for matching a rehabilitation mode according to the collected data and in combination with the type of the subject; and the closed-loop rehabilitation interaction module is used for adjusting parameters of the TMS equipment according to the rehabilitation mode and stimulating the final stimulation area under the optimal stimulation state of the subject. For a mild stroke patient, the system dynamically monitors the plastic change of the autonomic nerve and adaptively adjusts the treatment parameters according to the plastic change of the autonomic nerve; for a severe stroke patient, the virtual reality situation is combined to effectively induce motor imagery and synchronously collect and analyze electroencephalogram signals, closed-loop nerve regulation is achieved, synchronous accurate intervention and curative effect optimization of the severe stroke patient are achieved, and an intelligent solution is provided for rehabilitation treatment of the nerve function of the severe stroke patient.

The invention discloses an electroencephalogram signal clustering method based on a Mama framework and comparative learning, and belongs to the crossing field of engineering application and information science, and the method comprises the following steps: collecting and preprocessing label-free electroencephalogram data; the method comprises the following steps of: constructing a Mama-based feature extractor, dividing an input electroencephalogram signal into a plurality of slices by adopting an electroencephalogram signal slice embedding strategy so as to obtain fine-grained local information, and modeling a potential context relationship by utilizing a slice perception scanning mechanism; the method comprises the following steps: constructing an original data view and an enhanced data view, and respectively inputting samples of the two data views into two Mamba feature extractors with shared network parameters for feature embedding; processing the feature pairs of the two views using an instance projector, constructing weighted instance level contrast learning wherein the distance in the original data space provides weight information; processing feature pairs of the two views by using a clustering projector, and constructing a clustering distribution discrimination contrast learning branch and a semantic perception contrast learning branch; and finally, performing joint training by using the comparative learning branches, and outputting a high-quality clustering result by ensuring feature, clustering distribution and semantic consistency.

The invention discloses a hair area hydrogel Ag3Sn tow brain electrode and application. The hair area hydrogel Ag3Sn tow brain electrode comprises an electrode base body and hydrogel. The electrode base body comprises an electrode upper shell, an electrode lower shell, a conducting strip and an electrode buckle; the lower electrode shell and the upper electrode shell are detachably connected; the hydrogel is fixed in an inner cavity of the electrode lower shell and extends out of the shell bottom surface of the electrode lower shell; an electrode buckle extends out of one end, far away from the electrode lower shell, of the electrode upper shell; the hydrogel is in contact with the conducting strip, and the conducting strip is connected with the electrode buckle; and the conducting strip comprises an Ag3Sn nano wire bundle modified with a conducting polymer. According to the hair area hydrogel Ag3Sn wire bundle brain electrode, the structure design of micro / nano wire bundles on the surface of the electrode is optimized, so that the electrode has good conductivity and more excellent charge storage capacity, and the interface impedance and the long-time electroencephalogram collection capacity are reduced.

The invention discloses an electroencephalogram cap for meditation training and a control method thereof, and relates to the technical field of intelligent wearable equipment, the electroencephalogram cap comprises a cap body, an information acquisition module, a signal processing module, a wireless transmission module, a power supply module and a control host; the information acquisition module, the signal processing module, the wireless transmission module and the power supply module are respectively arranged on the helmet body, and the control host is respectively and electrically connected with the information acquisition module, the signal processing module, the wireless transmission module and the power supply module; the information acquisition module is used for acquiring microvolt-level electroencephalogram signals of a corresponding brain region and providing a data source for state recognition, the signal processing module is used for real-time denoising, power frequency interference suppression and artifact preliminary recognition, and the wireless transmission module is used for outputting processed electroencephalogram data streams in real time and transmitting the processed electroencephalogram data streams to the corresponding brain region. The power supply module supplies power to the information acquisition module, the signal processing module and the wireless transmission module. The electroencephalogram cap is convenient to wear, multi-dimensional analysis is achieved, feedback is timely, and the comfort level of mind calming training experience and the training effect can be improved.

The invention relates to the field of disease prediction, in particular to a cognitive impairment prediction system based on synchronously acquired electroencephalogram-stomach electrical signals, comprising: a data acquisition module for acquiring data including electroencephalogram data and stomach and intestine electrical data; the first analysis module is used for obtaining an average electrode inconsistency index of the first channel and the second channel based on a first model according to the time sequence data; and the prediction module is used for comparing the average electrode inconsistency index of the subject with a preset evaluation threshold value so as to predict a cognitive impairment patient. The system provided by the invention can more accurately and efficiently distinguish patients with mild cognitive impairment from healthy people, and can be used as an auxiliary diagnosis tool to be applied to primary screening of mild cognitive impairment on relatively large people (such as communities and physical examination centers).

A method of predicting a seizure occurrence or a surgical outcome includes monitoring a patient using an electroencephalography (EEG) system, including recording EEG data that indicates a seizure, and extracting a peri-ictal segment of the EEG data that includes a pre-ictal period immediately preceding the seizure, or a post-ictal period immediately following the seizure. The method also includes processing the peri-ictal segment, including generating an electrode-wise power spectral density (PSD) feature across a frequency band defined by at least one of a delta frequency range, a theta frequency range, an alpha frequency range, a beta frequency range, and a gamma frequency range. The method also includes inputting the PSD feature into a model, wherein the model predicts a seizure occurrence or a surgical outcome of the patient.

The invention provides an electroencephalogram data transmission method combining lossless and lossy compression, belongs to the field of electroencephalogram data transmission and compression processing, and is used for solving the problems of low transmission efficiency of pure lossless compression and easy loss of key diagnosis information of pure lossy compression in related technologies. According to the method, multi-dimensional features of electroencephalogram data are extracted, a key area and a non-key area are divided, lossless compression and lossy compression processing are adopted respectively, the area division proportion is dynamically adjusted in combination with transmission bandwidth, signal features and application scenes, and corresponding data are transmitted through dual protocols. The transmission efficiency can be improved while the clinical diagnosis marker and the core characteristics are reserved, and the scene adaptability and the transmission reliability are achieved at the same time.

The invention relates to the technical field of neuroscience and artificial intelligence, in particular to an autism spectrum disorder prediction method based on EEG machine learning, which comprises the following steps: step 1, collecting EEG data for children with autism spectrum disorder and healthy control children; step 2, preprocessing the collected EEG data, then extracting sleep spindle wave signals in the EEG data, and analyzing the sleep spindle wave signals to obtain feature information data; step 3, performing standardization processing on the obtained feature information data to obtain a data set, and then dividing the data set into a training set and a test set; 4, training the training set through different machine learning algorithms to obtain corresponding training models, and selecting an optimal model from the training models through a five-fold cross validation strategy; and 5, inputting the test set into each optimal model for evaluation, and comparing evaluation results to obtain a target model. The method provides a scientific basis for early intervention of ASD.

The invention relates to the technical field of epilepsy detection, in particular to a self-supervised graph neural network epilepsy detection method based on Transform. The method comprises the following steps: S1, preprocessing original EEG data, and constructing an EEG graph; s2, building a graph neural network based on DCTran, and respectively capturing a space-time dependency relationship of the EEG signal through diffusion convolution and a Transform structure; and S3, training the model through spatio-temporal joint complementary double-branch pre-training based on the self-supervised prediction pre-training task and the mask reconstruction pre-training task. According to the epilepsy detection method based on the self-supervised graph neural network of the Transform, provided by the invention, EEG data is modeled into a graph structure, and a diffusion convolution space-time network DCTran based on the Transform is provided; the diffusion convolution accurately captures a complex spatial relationship between the electrodes through multi-order neighbor information propagation; and meanwhile, global context information of the EEG signal is fully utilized, and DCTran is introduced into a Transform structure to model a time sequence, so that the long-distance time dependency relationship in the signal is effectively captured.

The invention discloses an EEG intelligent agent automatic analysis method based on a large language model, which takes the large language model as a strategy engine, autonomously understands user analysis intentions and intelligently decomposes tasks, and further dynamically schedules and collaboratively integrates analysis resources including traditional feature engineering, diversified deep learning models and an external knowledge base. According to the method, end-to-end automatic coordination and execution of complex EEG analysis tasks such as signal preprocessing, feature extraction, event positioning, classification diagnosis, emotion recognition, sleep staging and the like can be realized, the limitation of a single detection or classification task is broken through, and through context perception and flexible space-time analysis capability, the accuracy of the EEG analysis is improved. And multitask and continuous deep reasoning and interpretation can be carried out on the complex EEG data. According to the method, the general planning and reasoning capability of a large language model is deeply fused with a special analysis model in the EEG field, so that the automation level, flexibility and clinical application potential of EEG analysis are remarkably improved.

The invention discloses a fusion enhancement online motor imagery intention recognition system and a training method thereof.The method comprises the steps that a first training data set and a second training data set of motor imagery of a user are collected, the first training data set is subjected to a fusion enhancement method containing at least one of noise enhancement, phase enhancement, frequency band enhancement and channel enhancement, and the fusion enhancement method can be implemented independently or in series; generating a first enhancement data set based on the initial fusion enhancement parameters, and training an identification model in a first training stage; in the evaluation stage, identifying a second training data set by using the model, and generating a second enhanced data set for correctly identified data by using the same fusion enhancement method and parameters; in the second training stage, the two types of enhanced data sets are combined to re-train the model for optimization; and meanwhile, traversing the fusion enhancement parameter space, and determining optimal parameters containing various enhancement parameters. Through multiple selectable data enhancement, online loop optimization and automatic parameter optimization, the problems that motor imagery electroencephalogram data is small in sample size, uneven in distribution and high in timeliness are effectively solved.

The embodiment of the invention belongs to the technical field of data processing, and relates to a user state detection method which comprises the following steps: acquiring respiration data, electroencephalogram data and basic sensing data of a user; performing feature extraction on the respiration data and the electroencephalogram data to obtain respiration feature data and electroencephalogram feature data; performing data fusion on the respiration feature data and the electroencephalogram feature data to obtain a joint feature vector; determining a target state detection model in each preset state detection model according to the basic sensing data; and performing state detection on the user according to the target state detection model and the joint feature vector, and determining the state of the user. According to the method and the device, the physiological data of the user is acquired from different dimensions, and the acquired data is subjected to multi-modal fusion and conjoint analysis, so that the user state detection in different scenes is realized, diversified scene requirements are met, and the accuracy of the intelligent wearable equipment in detecting the user state is improved.

The invention provides an electroencephalogram image reconstruction method based on frequency steering and bidirectional diffusion, and belongs to the technical field of brain-computer interfaces and computer vision. The method comprises the steps that EEG data and corresponding image data are acquired and preprocessed; constructing a reconstruction model comprising a frequency domain-space-time dynamic encoder and a bidirectional submerged space diffusion generator; wherein the frequency domain-space-time dynamics encoder adopts a frequency-oriented Mama architecture, explicitly models neural oscillation dynamics by constructing a block diagonal state matrix, and extracts features in combination with graph convolution and space-time convolution; the bidirectional submerged space diffusion generator comprises a symmetric EEG-to-image submerged space diffusion model and an image-to-EEG submerged space diffusion model, and training is carried out through generative cyclic consistency constraint; and finally, mapping the collected EEG data into image semantic features by using the trained model, and driving a pre-training generation model to reconstruct an image. According to the method, the problems that in the prior art, the electroencephalogram frequency domain specificity dynamic state is ignored, and cross-modal semantic alignment is weak are solved, and the semantic consistency of electroencephalogram decoding and the fidelity of image reconstruction are remarkably improved.

The invention discloses an emotion recognition method based on spatio-temporal multi-scale attention convolutional neural network, which comprises: collecting EEG data of subjects for preprocessing to obtain EEG data containing spatial dimension and temporal dimension; constructing a lightweight convolutional neural network including two-stream spatio-temporal feature construction layer, hybrid attention mechanism layer, high-order fusion layer and classification layer; wherein the two-stream spatio-temporal feature construction layer comprises a temporal feature extraction module and a parallel spatial feature extraction module; the high-order fusion layer is used to re-learn from the learned global convolution kernel to the representation of the local hemisphere convolution kernel; the trained lightweight convolutional neural network is used to identify EEG data, and the emotion recognition results of the subjects are obtained. By constructing a lightweight model with fewer parameters, the accuracy and efficiency of EEG-driven emotion recognition are improved.

The invention discloses a frequency-phase-linear frequency modulation combined high-frequency SSVEP coding and decoding system and method, and the system comprises a stimulation parameter configuration module which configures three different linear frequency modulation rates for each basic frequency, a visual stimulation generation and presentation module which generates a dynamic brightness modulation stimulation sequence of each target, and a dynamic brightness modulation stimulation sequence of each target. The electroencephalogram data preprocessing module preprocesses electroencephalogram data acquired by the electroencephalogram signal acquisition unit, the reference signal generation and feature extraction module reconstructs reference signals, and the feature fusion and decision module assigns values to low-frequency sub-bands and executes correlation aggregation operation and decision to obtain classified target characters. According to the method, multi-target coding is supported in a limited bandwidth, so that the instruction capacity and the user experience of a high-frequency brain-computer interface are improved; and the joint frequency-phase-linear frequency modulation is adopted, so that high-precision identification can be realized without individual training, and the decoding efficiency and robustness are improved.

The invention discloses a brain load assessment method and system based on hemodynamic information and electroencephalogram signals, and relates to the field of brain-computer interface classification systems. The technical problem that how to fuse electroencephalogram signals and hemodynamic information and comprehensively utilize complementary characteristics of the electroencephalogram signals and the hemodynamic information in time and space dimensions to achieve accurate recognition and dynamic evaluation of the mental load state in the prior art is urgently needed to be solved in the current brain-computer interface field is solved. The method comprises the following steps: acquiring EEG data and fNIRS data of a testee; performing preprocessing, selecting a suitable frequency band, removing interference such as high-frequency noise and low-frequency baseline drift, and constructing a corresponding training data set; respectively constructing an EEG single-mode network and an fNIRS single-mode network which are used for extracting related characteristics; and constructing an EEG-fNIRS multi-mode classification network for fusing data of brain imaging modes in the EEG single-mode network and the fNIRS single-mode network, and performing classification evaluation on the hemodynamic information of the subject and the brain power load data of the electroencephalogram by using the EEG-fNIRS multi-mode classification network.

The invention relates to a driver brain-controlled vehicle intention recognition method and system based on neural mechanism guidance. Recording EEG signals of the driver under different vehicle control motor imagery tasks; analyzing the collected EEG data, and identifying a key cortex activation area related to vehicle control; deploying the positions of an fNIRS emission light source and a receiving detector, and obtaining hemodynamic signals; brain region activation analysis is carried out on the collected fNIRS signals, the difference response condition of a key cortex activation region in a motor imagery task is verified, and a motor imagery data set about driver vehicle control is made; constructing a deep learning model, automatically extracting time domain, frequency domain, space and time sequence characteristics of the fNIRS signal, and training the model; the motor imagery brain signals are classified, and the vehicle control intention of the driver is output; according to the method, the physiological interpretability and the model structure rationality of the brain-controlled vehicle system are remarkably improved, and the problem of insufficient generalization caused by blind modeling is avoided.

The invention discloses an aromatic sleep intervention feedback system based on a lightweight electroencephalogram sleep staging model, which relates to the technical field of intelligent sleep monitoring and intervention and comprises an electroencephalogram sensor module, a data processing module and an atomizer module. The electroencephalogram sensor module is used for collecting an original electroencephalogram signal of a prefrontal lobe Fpz point location in a single-point mode, and the original electroencephalogram signal is transmitted to the data processing module through Bluetooth after being preprocessed and subjected to digital-to-analog conversion. The data processing module is used for preprocessing electroencephalogram data and then inputting the electroencephalogram data into the lightweight model for sleep staging, and a dynamic aroma intervention instruction is generated. The atomizer module controls essential oil atomization parameters according to the instruction to realize closed-loop intervention; according to the method, the forehead lobe Fpz point position electroencephalogram signals are collected in a single-point mode, real-time processing is achieved in combination with a lightweight sleep staging model, aroma intervention parameters are dynamically adjusted based on different sleep stages, and the problems that in the prior art, equipment wearing is complex, algorithm delay is large, and the intervention mode is single are solved; the method has the advantages of simplifying the wearing structure, improving the processing efficiency and optimizing the intervention effect.

The invention provides an EEG function connection prediction method based on fMRI deep cross-modal representation learning, and relates to the crossing field of neuroiconography and artificial intelligence. According to the method, a sample pair is constructed through time alignment of EEG and a blood oxygen level dependent signal, an end-to-end deep neural network architecture composed of a time projection unit, a cross-attention fusion encoder and a connection synthesis head is designed, and direct mapping from a BOLD signal to EEG function connection is achieved. The model adopts a composite loss function, and gives consideration to the consistency of the prediction connection matrix with the real EEG function connection in numerical precision and topological mode, thereby achieving the high-fidelity reconstruction of the EEG function connection map at the frequency domain level. According to the invention, the topological structure stability of the brain function network can be effectively maintained. Under the conditions of incomplete EEG data, serious noise interference or complete loss, the frequency domain function connection characteristics can still be stably recovered, and a reliable analysis substitution path is provided for brain function network research.

The invention discloses a flight safety warning method based on pilot operation intention recognition, which comprises the following steps: S1, collecting EEG data of a subject in each stage of simulated flight through an electroencephalogram signal data collection device; s2, carrying out preprocessing operation on the collected EEG data; s3, extracting and fusing the electroencephalogram signals by constructing a convolutional gradient attention network, and mapping fused features to an operation intention of a pilot; s4, calculating safety risk factors according to the deviation between the flight state and the preset flight envelope, the pilot operation intention and the external environment factors; s5, comparing the risk factor with a preset risk threshold, and dividing the risk into a high level, a middle level and a low level; and S6, the cockpit touch alarm system triggers multi-mode alarm in the cockpit according to the risk level, and adjusts the flight state in real time. The flight accident risk caused by human factors can be reduced, and the aviation safety is improved.