Vestibular electroencephalographic detection analysis system

Abstract

The application belongs to the technical field of vestibular medicine, and particularly relates to a vestibular electroencephalogram detection and analysis system, which comprises a vestibular multi-modal stimulation module, a multi-modal signal synchronous acquisition module, a vestibular electroencephalogram feature analysis module, a vestibular function corroboration reference module and a result generation and output module, wherein the vestibular multi-modal stimulation module is used for applying vestibular physiological stimulation to a user; the multi-modal signal synchronous acquisition module is used for synchronously and in parallel acquiring multi-modal data signals in a time synchronization manner while the vestibular multi-modal stimulation module applies the vestibular physiological stimulation; the vestibular electroencephalogram feature analysis module is used for adaptively selecting an analysis algorithm to perform vestibular electroencephalogram feature analysis on the multi-modal data; the vestibular function corroboration reference module is used for constructing a corroboration parameter set; and the result generation and output module is used for determining a vestibular electroencephalogram analysis and evaluation result of the user.

Description

Technical Field

[0001] This invention belongs to the field of vestibular medical technology, specifically a vestibular electroencephalogram (EEG) detection and analysis system. Background Technology

[0002] The inner ear vestibular system is a core sensory system for maintaining human motion perception, postural stability, spatial orientation, and balance regulation. Accurate assessment of vestibular function is of significant clinical and research importance for the diagnosis and treatment of dizziness, vertigo, balance disorders, and in fields such as sports medicine and neuromodulation. Current technologies primarily assess vestibular-ocular reflexes (such as nystagmography and video head impulse tests) and vestibular-spinal reflex pathways (such as static / dynamic balance and gait analysis) to indirectly reflect peripheral vestibular function. These methods are technically mature and widely used in clinical practice.

[0003] However, when existing technologies achieve an objective quantitative assessment of the human cerebral cortex's ability to perceive vestibular information and spatially locate itself, they are mainly limited by the core technical bottleneck of not being able to establish a direct mapping relationship between peripheral physiological stimuli and central nervous system responses. This makes it difficult to simultaneously achieve high spatiotemporal resolution resolution of the vestibular-cerebral cortex pathway and specific identification of different vestibular diseases or stimulation patterns in practical applications, thus affecting overall performance. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides a vestibular electroencephalogram (EEG) detection and analysis system to solve the problem that existing technologies cannot establish a direct mapping relationship between peripheral physiological stimuli and central nervous system responses.

[0005] The vestibular electroencephalogram (EEG) detection and analysis system includes: a vestibular multimodal stimulation module, used to apply vestibular physiological stimulation to the user and record the start time, duration and intensity parameters of each stimulus, including: a temperature stimulation unit, a rotation stimulation unit, a head movement stimulation unit and a stimulation timing control unit;

[0006] The multimodal signal synchronous acquisition module is used to acquire multimodal data signals in parallel in a time-synchronous manner while the vestibular physiological stimulation module applies vestibular physiological stimulation. It includes: a multi-lead EEG acquisition unit, a vestibular peripheral reflex acquisition unit, a vestibular symptom behavior labeling unit, and a multimodal synchronous controller.

[0007] The vestibular EEG feature analysis module is used to receive multimodal data acquired by the multimodal signal synchronous acquisition module, and adaptively select an analysis algorithm to perform vestibular EEG feature analysis on the multimodal data according to the vestibular physiological stimulation applied by the vestibular multimodal stimulation module. It includes: EEG preprocessing unit, stimulus locking analysis unit, time-frequency analysis unit, source localization unit and adaptive algorithm selection unit.

[0008] The vestibular function corroborating reference module is used to quantify the vestibular peripheral reflection signals and vestibular symptom behavior marker signals in multimodal data signals. Based on the quantification results, a corroborating parameter set is constructed, including: nystagmus quantification unit, balance and posture analysis unit, symptom quantification unit, and spatiotemporal correlation mapping unit.

[0009] The results generation and output module is used to determine the user's vestibular EEG analysis and evaluation results based on the analysis results of the vestibular EEG feature analysis module and the mutually corroborating reference system constructed by the vestibular function corroborating reference module.

[0010] Preferably, the application of vestibular physiological stimulation to the user is specifically as follows:

[0011] Based on clinical testing needs, the stimulation unit applies standardized vestibular physiological stimulation to users by constructing multiple stimulation paradigms, specifically:

[0012] The temperature stimulation unit generates a temperature difference by injecting a thermostatic gas or liquid into the user's external auditory canal to excite or inhibit the function of the horizontal semicircular canals.

[0013] The rotational stimulation unit applies angular acceleration stimulation to the user through a programmable rotating seat to excite the semicircular canals;

[0014] The head movement stimulation unit stimulates the peripheral vestibular receptors by guiding the user to actively or passively perform head movements in a specific direction and at a specific frequency.

[0015] The stimulation timing control unit precisely controls the timing of each stimulation unit and generates a stimulation timing reference signal.

[0016] Preferably, the parallel acquisition of multimodal data signals in a time-synchronized manner is specifically as follows:

[0017] While applying vestibular physiological stimulation to the vestibular multimodal stimulation module, multimodal data signals are acquired in parallel in a time-synchronized manner. The multimodal data signals include:

[0018] Multi-lead EEG signals were acquired using an EEG recording device with at least 64 leads.

[0019] Peripheral vestibular reflex signals are obtained by acquiring vestibular-ocular reflex data through nystagmus / eye movement recording devices;

[0020] Vestibular symptom behavioral marker signals are obtained by acquiring the user's subjective vertigo score and physiological state data through subjective perception marker devices and vital sign sensors.

[0021] Preferably, the multimodal synchronization controller is as follows:

[0022] The multimodal synchronization controller sends synchronous trigger pulses to the multi-lead EEG acquisition unit, the vestibular peripheral reflex acquisition unit, and the vestibular symptom behavior labeling unit to ensure that all acquisition channels start data acquisition at the same time reference.

[0023] Align the stimulus event timestamps with the data streams of each modality to generate a synchronous multimodal data packet.

[0024] Preferably, the stimulus-locking analysis unit performs stimulus-locking segmentation on the preprocessed EEG signal based on the stimulus timing reference signal, as follows:

[0025] For each stimulus event, the stimulus-locked analysis unit extracts the pre-stimulus baseline period and the post-stimulus response period, and constructs a stimulus-locked EEG fragment set, which includes:

[0026] ;

[0027] in, The range of values ​​is , The base time period length, For the length of the response period, This represents the time when the i-th stimulus event occurs. This represents the preprocessed EEG signal.

[0028] Preferably, the time-frequency analysis unit performs time-frequency transformation on the preprocessed EEG signal to extract energy change features in different frequency bands, as follows:

[0029] Calculate the energy changes in each frequency band, including: frequency band, frequency band, frequency band, Frequency band (13-30Hz) and Frequency band, specifically:

[0030] set up Bandwidth energy is Then we have:

[0031] ;

[0032] in, express Energy in the frequency band Indicates time ,frequency Time-frequency energy distribution at the location;

[0033] according to The calculation method for frequency band energy is to calculate sequentially... frequency band, frequency band, frequency band and Frequency band energy corresponding to a frequency band.

[0034] Preferably, the adaptive algorithm selection unit adaptively selects the corresponding analysis algorithm based on the currently applied vestibular physiological stimulation type, as follows:

[0035] Identify stimulus type labels, including: semicircular canal temperature stimulation, semicircular canal rotation stimulation, and otolith linear acceleration stimulation;

[0036] Based on the stimulus type label, the corresponding analysis algorithm is adaptively selected, specifically as follows:

[0037] When a semicircular canal stimulus is detected, the adaptive algorithm selects the unit to invoke the event-related potential analysis algorithm;

[0038] When otolith stimulation is detected, the adaptive algorithm selects the unit to call the time-frequency energy spectrum analysis algorithm to analyze the energy distribution of the EEG signal in the time-frequency domain;

[0039] When a combined stimulus is detected, the adaptive algorithm selection unit simultaneously invokes the event-related potential analysis algorithm and the time-frequency energy spectrum analysis algorithm to perform joint time-domain and time-frequency domain analysis on the EEG signal.

[0040] Preferably, the quantization process is as follows:

[0041] Quantization processing involves sequentially processing the vestibular peripheral reflex signals and vestibular symptom behavior marker signals from the multimodal data signals through a nystagmus quantification unit, a balance and posture analysis unit, and a symptom quantification unit. Specifically:

[0042] The nystagmus quantification unit processes the eye movement signals acquired by the vestibular peripheral reflection acquisition unit and extracts the nystagmus parameters.

[0043] The balance and attitude analysis unit receives attitude data from the inertial measurement unit and calculates the center of gravity sway trajectory, sway area, and sway speed.

[0044] The symptom quantification unit quantifies the subjective vertigo scores and physiological data collected by the vestibular symptom behavior labeling unit.

[0045] A set of supporting parameters is constructed based on the processing results of the nystagmus quantification unit, balance and posture analysis unit, and symptom quantification unit.

[0046] Preferably, the spatiotemporal correlation mapping unit is used to correlate and map the supporting parameter set with EEG features in the temporal and spatial dimensions, as follows:

[0047] In the time dimension, the temporal correlation between the peak time of EEG characteristics and the peak time of symptom scores was calculated separately.

[0048] In the spatial dimension, calculate the spatial distance error between the source activation peak coordinates and the expected vestibular cortex center coordinates;

[0049] Based on the results of the temporal and spatial correlation analysis, a comprehensive supporting reference vector is constructed.

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] 1. This invention applies standardized vestibular physiological stimuli to users by utilizing a vestibular multimodal stimulation module (including a temperature stimulation unit, a rotation stimulation unit, and a head movement stimulation unit), and records the start time, duration, and intensity parameters of each stimulus. Simultaneously, a multimodal signal synchronous acquisition module acquires multi-lead EEG signals, vestibular peripheral reflex signals, and vestibular symptom behavior marker signals in parallel in a time-synchronized manner. This achieves precise synchronous recording of vestibular peripheral stimulation and central nervous system response, solving the core technical bottleneck of existing technologies that cannot establish a direct mapping relationship between peripheral physiological stimulation and central nervous system response. Thus, in practical applications, it takes into account both high spatiotemporal resolution analysis of the vestibular-cerebral cortex pathway and the ability to specifically identify different vestibular diseases or stimulation patterns.

[0052] 2. This invention utilizes a vestibular EEG feature analysis module to adaptively select the corresponding analysis algorithm based on the type of vestibular physiological stimulation applied by the vestibular multimodal stimulation module (using the event-related potential analysis algorithm for semicircular canal stimulation, the time-frequency energy spectrum analysis algorithm for otolith stimulation, and both algorithms for combined analysis for combined stimulation). This enables differentiated and precise extraction of EEG features evoked by different vestibular receptors, solving the technical problem of lacking specific EEG analysis algorithms for different vestibular receptors in existing technologies. Thus, in practical applications, it improves the accuracy and diagnostic specificity of EEG feature recognition for dizziness and vertigo perception in different vestibular diseases.

[0053] 3. This invention utilizes a vestibular function corroborating reference module to quantify peripheral vestibular reflex signals and vestibular symptom behavior marker signals, constructing a set of corroborating parameters (including nystagmus intensity, slow phase angular velocity, subjective vertical visual deviation angle, and vertigo level). This set of corroborating parameters is then spatiotemporally mapped to EEG characteristics, creating a mutually corroborating reference system. This achieves cross-validation and mutual corroboration between EEG characteristics, peripheral physiological indicators, and subjective symptoms, solving the technical problem of existing EEG analysis results lacking objective reference and being difficult to verify reliability. Therefore, in practical applications, this invention improves the accuracy and clinical reliability of vestibular EEG detection results.

[0054] 4. This invention utilizes a spatiotemporal correlation mapping unit to calculate the temporal correlation between the peak time of EEG features and the peak time of symptom scores, as well as the Pearson correlation coefficient between EEG feature amplitude and slow phase angular velocity in the time dimension. In the spatial dimension, it calculates the spatial distance error between the source activation peak coordinates and the expected vestibular cortex center coordinates. Based on the results of the temporal and spatial correlation analysis, it constructs a comprehensive corroborating reference vector (including the EEG-nystagmus consistency index, the EEG-symptom consistency index, and the EEG-source localization credibility index). This achieves a dual quantitative assessment of EEG features in terms of temporal synchronization and spatial localization accuracy, solving the technical problem in the prior art of balancing the temporal resolution and spatial localization accuracy of EEG signals. Thus, in practical applications, it achieves high spatiotemporal resolution analysis of the vestibular-cerebral cortex pathway.

[0055] 5. This invention utilizes a source localization unit to output a three-dimensional source distribution map, thereby determining the localization of cortical responses induced by vestibular stimulation (including vestibular cortical regions such as the parietal vestibular cortex, posterior insular cortex, and superior temporal gyrus). Furthermore, through a result generation and output module, it generates cortical response localization maps induced by different vestibular receptor stimuli. This enables the spatial localization of the brain's perception of peripheral vestibular receptor information from both ears to objective visualization, moving beyond subjective description. It solves the technical problem in existing technologies where vestibular central perception localization relies on subjective patient descriptions and lacks objective indicators. Therefore, in practical applications, it provides objective spatial localization evidence for the study of the central mechanisms of vestibular diseases.

[0056] 6. This invention utilizes an individualized assessment modeling unit to construct an assessment function, integrates EEG feature vectors, supporting parameter sets, and comprehensive supporting reference vectors for multimodal fusion assessment, and generates structured reports and comparative analysis views through a report generation unit and a visualization output unit. This achieves individualized, visualized, and quantifiable comprehensive assessment of the user's vestibular EEG analysis results, solving the technical problems of scattered, unintegrated, and difficult-to-guide clinical diagnosis and treatment results in existing technologies. Thus, in practical applications, it provides comprehensive, objective, and traceable auxiliary evidence for the diagnosis and rehabilitation of patients with vestibular diseases such as dizziness, vertigo, and balance disorders. Attached Figure Description

[0057] Figure 1 This is a schematic diagram of the overall structure of the vestibular electroencephalogram (EEG) detection and analysis system of the present invention. Detailed Implementation

[0058] The embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and should not be construed as limiting the scope of the invention.

[0059] Example 1

[0060] Reference Figure 1As one embodiment of the present invention, a vestibular electroencephalogram (EEG) detection and analysis system is provided, comprising:

[0061] The vestibular multimodal stimulation module is used to apply at least one preset vestibular physiological stimulus to the user. The vestibular physiological stimulus includes temperature stimulation, rotational stimulation, and head movement stimulation, and records the start time, duration, and intensity parameters of each stimulus. The specific implementation is as follows:

[0062] The vestibular multimodal stimulation module includes a temperature stimulation unit, a rotation stimulation unit, a head movement stimulation unit, and a stimulation timing control unit. Based on clinical testing needs, the stimulation units apply standardized vestibular physiological stimulation to the user by constructing multiple stimulation paradigms, specifically as follows:

[0063] The temperature stimulation unit generates a temperature difference by injecting a thermostatic gas or liquid into the user's external auditory canal to excite or inhibit the function of the horizontal semicircular canals. Specifically:

[0064] The left ear perfusion temperature was set to The perfusion temperature of the right ear was Cold stimuli are at temperatures below body temperature (e.g., 30 degrees Celsius), while hot stimuli are at temperatures above body temperature (e.g., 44 degrees Celsius).

[0065] Temperature stimulation unit records the start time of perfusion End of infusion and duration And generate temperature stimulus intensity parameters. ,in Indicates the infusion temperature. This indicates the user's body temperature.

[0066] The rotational stimulation unit applies angular acceleration stimulation to the user through a programmable rotating seat to excite the semicircular canals, specifically:

[0067] Set the rotational angular velocity to angular acceleration is ;

[0068] Rotational stimulation unit records the rotation initiation time Rotation termination time Maximum angular velocity Maximum angular acceleration And the direction of rotation (clockwise or counterclockwise);

[0069] Based on the rotational stimuli at different stages, the corresponding angular velocity functions are constructed, resulting in:

[0070] when hour, ;

[0071] when hour, ,in The preset constant angular velocity is set by the implementers according to the actual application scenario.

[0072] The head movement stimulation unit stimulates peripheral vestibular receptors by guiding users to actively or passively perform head movements in specific directions and at specific frequencies. The head movement stimulation unit includes a head posture sensor (such as a nine-axis inertial measurement unit) to collect real-time head motion parameters from the user. Specifically:

[0073] Let the head's pose angle in three-dimensional space be ( , , ), representing pitch angle, yaw angle and roll angle respectively;

[0074] The head movement stimulation unit records the start time of head movement. , head movement termination time Head movement direction, head movement amplitude, and head movement frequency ;

[0075] Based on the sine function, the head motion angle function is constructed as follows:

[0076] ;

[0077] in, Indicates the range of head movement. This represents the initial phase, with a value of 0.

[0078] The stimulation timing control unit precisely controls the timing of each stimulation unit and generates a stimulation timing reference signal, specifically as follows:

[0079] The stimulation timing control unit records the global stimulation start time. And generate timestamp labels for each stimulus event. , Indicates the sequence number of the stimulus event;

[0080] The stimulation timing control unit supports single stimulation mode, alternating stimulation mode, and combined stimulation mode;

[0081] In the combined stimulation mode, rotational stimulation and head movement stimulation are applied simultaneously to simulate a complex vestibular input environment;

[0082] The stimulus timing control unit packages the recorded stimulus parameters and timing information into stimulus metadata. The data is then transmitted to the multimodal signal synchronous acquisition module.

[0083] The multimodal signal synchronous acquisition module simultaneously acquires multimodal data signals in a time-synchronized manner while the vestibular multimodal stimulation module applies vestibular physiological stimulation. These multimodal data signals include multi-lead EEG signals acquired using at least a 64-lead EEG recording device; peripheral vestibular reflex signals acquired using a nystagmus / eye movement recording device; and vestibular symptom behavior labeling signals, acquired through a subjective perception labeling device and vital sign sensors to obtain the user's subjective vertigo score and physiological state data. Specifically:

[0084] The multimodal signal synchronous acquisition module includes: a multi-lead EEG acquisition unit, a vestibular peripheral reflex acquisition unit, a vestibular symptom behavior labeling unit, and a multimodal synchronous controller, which are implemented as follows:

[0085] The multi-lead EEG acquisition unit continuously acquires the user's whole brain activity signals using an EEG recording device with at least 64 leads at a preset sampling rate (e.g., 1000Hz).

[0086] The multi-lead EEG acquisition unit is equipped with electrodes arranged according to the international 10-10 or 10-20 system, covering at least the prefrontal, central, parietal, temporal, and occipital regions.

[0087] Set the EEG signal matrix as The transpose of, where Indicates the first The conductive electrode at time The voltage value;

[0088] The multi-lead EEG acquisition unit performs pre-amplification, bandpass filtering (e.g., 0.1-100Hz), and analog-to-digital conversion on the acquired raw EEG signals to generate a digital EEG data stream;

[0089] The vestibular peripheral reflex acquisition unit acquires vestibular oculomotor reflex data through nystagmus / eye-tracking devices (such as video nystagmus meters, eye trackers, or electrode-type eye trackers), specifically:

[0090] The vestibular peripheral reflex acquisition unit acquires horizontal, vertical, and torsional eye movement signals from both eyes in real time, resulting in:

[0091] Set the horizontal eye movement angle as The vertical eye movement angle is Twist the eye movement angle to ;

[0092] For video nystagmus recording, the vestibular peripheral reflection acquisition unit extracts eye movement parameters through pupil-corneal reflection analysis.

[0093] For electrode-based eye-tracking recording, eye movement parameters are extracted using electrooculography analysis.

[0094] Vestibular peripheral reflection acquisition unit calculates slow phase angular velocity It is used to quantify the intensity of the vestibular-ocular reflex.

[0095] The vestibular symptom behavioral labeling unit acquires the user's subjective vertigo score and physiological state data through a subjective perception labeling device and vital sign sensors, specifically:

[0096] The subjective perception labeling device includes a handheld input terminal or voice acquisition device. During or after the stimulus is applied, the user inputs a subjective vertigo intensity score (e.g., a 0-10 visual analog scale score) and a description of the vertigo nature (e.g., a feeling of spinning, swaying, or loss of balance) in real time or retrospectively through this device.

[0097] The subjective vertigo score is set as follows: ;

[0098] Vital signs sensors include heart rate sensors, blood pressure sensors, skin conductance sensors, and blood oxygen saturation sensors, which collect the user's physiological state data in real time.

[0099] Set heart rate systolic blood pressure is Diastolic blood pressure is Skin conductance response level is Blood oxygen saturation was ;

[0100] The vestibular symptom behavior labeling unit constructs symptom behavior vectors from the user's physiological state data. The transpose of .

[0101] The multimodal synchronization controller receives stimulus metadata sent by the stimulus timing control unit. And based on the received stimulus metadata, a global synchronization clock signal is determined, specifically:

[0102] The multimodal synchronization controller sends synchronization trigger pulses to the multi-lead EEG acquisition unit, the vestibular peripheral reflex acquisition unit, and the vestibular symptom behavior labeling unit to ensure that all acquisition channels operate on the same time reference. Start data collection;

[0103] The multimodal synchronization controller aligns the stimulus event timestamps with the data streams of each modality to generate synchronized multimodal data packets. ,in This represents the stimulus parameters that are effective at time t.

[0104] The vestibular EEG feature analysis module receives multimodal data acquired by the multimodal signal synchronous acquisition module and adaptively selects analysis algorithms to perform vestibular EEG feature analysis on the multimodal data based on the vestibular physiological stimulation applied by the vestibular multimodal stimulation module. The analysis algorithms include an event-related potential analysis algorithm for semicircular canal stimulation and a time-frequency energy spectrum analysis algorithm for otolith stimulation, which are implemented as follows:

[0105] The vestibular EEG feature analysis module, based on multimodal data acquired by the multimodal signal synchronous acquisition module and combined with vestibular physiological stimulation applied by the vestibular multimodal stimulation module, adaptively selects an analysis algorithm to perform vestibular EEG feature analysis on the multimodal data, specifically:

[0106] The vestibular EEG feature analysis module includes: an EEG preprocessing unit, a stimulus-locking analysis unit, a time-frequency analysis unit, a source localization unit, and an adaptive algorithm selection unit, which are implemented as follows:

[0107] The EEG preprocessing unit processes the acquired multi-lead EEG signals. Preprocessing operations were performed, including: removal of DC offset, notch filtering (e.g., 50Hz power frequency notch), independent component analysis to remove electrooculogram (EOG) and electromyogram (EMG) artifacts, rereference (e.g., average reference or Laplace reference), and baseline correction. The preprocessed EEG signal was then set as... .

[0108] The stimulus-lock analysis unit performs stimulus-lock segmentation on the preprocessed EEG signal based on the stimulus timing reference signal, specifically as follows:

[0109] For each stimulus event (e.g., the start time of temperature stimulus) The stimulus-locked analysis unit extracts the pre-stimulus baseline time period (e.g., 500ms before stimulation) and the post-stimulus response time period (e.g., 3000ms after stimulation), and constructs a stimulus-locked EEG fragment set, then:

[0110] ;

[0111] in, The range of values ​​is , The base time period length, For the length of the response period, This indicates the time when the i-th stimulus event occurs;

[0112] The stimulus-locked analysis unit averages multiple stimulus-locked segments and calculates event-related potentials, resulting in:

[0113] ;

[0114] Where N is the total number of stimulus events. The calculated event-related potential is given by the time interval [time]. The average EEG response at the location, This represents the EEG signal segment corresponding to the i-th stimulus.

[0115] The time-frequency analysis unit performs time-frequency transformation on the preprocessed EEG signal to extract energy change features in different frequency bands, specifically:

[0116] The time-frequency analysis unit uses methods such as short-time Fourier transform, wavelet transform, or Hilbert-Huang transform to decompose the EEG signal into a time-frequency representation, resulting in:

[0117] Set the frequency to The square of , where f represents the frequency;

[0118] The time-frequency analysis unit calculates the energy changes in each frequency band, including: Frequency band (0.5-4Hz), Frequency band (4-8Hz) Frequency band (8-13Hz) Frequency band (13-30Hz) and Frequency band (30-50Hz);

[0119] set up Bandwidth energy is Then we have:

[0120] ;

[0121] in, express Energy in the frequency band Indicates time ,frequency Time-frequency energy distribution at the location;

[0122] according to The calculation method for frequency band energy is to calculate sequentially... frequency band, frequency band, frequency band and Frequency band energy corresponding to a frequency band.

[0123] The source localization unit, based on the scalp EEG distribution and a head volume conductor model, inverts the distribution of current sources within the cerebral cortex to achieve cortical localization of vestibular evoked responses. The specific implementation is as follows:

[0124] The source localization unit constructs a forward model matrix of EEG signals and, based on observed EEG data, estimates the minimum norm of the source distribution by introducing regularization constraints to determine the coordinate position of each electrode in three-dimensional space. Specifically:

[0125] Set the electrode position matrix as follows ;

[0126] ;

[0127] in, This represents the scalp potential distribution vector (dimension 64×1). This represents the lead field matrix (dimension 64×M, where M is the number of source points). This represents the source current density vector (with dimensions M×1). Represents the regularization parameter;

[0128] The source localization unit outputs a three-dimensional source distribution map through minimum norm estimation. ,in, Represents the spatial coordinates of the brain;

[0129] By analyzing the peak location of source activation, the cortical response induced by vestibular stimulation can be localized, such as the vestibular cortex of the parietal lobe, the posterior insular cortex, and the superior temporal gyrus, among other vestibular-related cortical regions.

[0130] The adaptive algorithm selection unit adaptively selects the corresponding analysis algorithm based on the type of vestibular physiological stimulation currently applied, as follows:

[0131] The adaptive algorithm selection unit first identifies stimulus type labels, including: semicircular canal temperature stimulation, semicircular canal rotation stimulation, and otolith linear acceleration stimulation (achieved through head movement or tilting).

[0132] When semicircular canal stimulation is detected, the adaptive algorithm selects the unit to invoke the event-related potential analysis algorithm, focusing on the amplitude and latency characteristics of latent components such as P1, N1, and P2 in the central region and parietal lobe channels after stimulus locking. P1, N1, and P2 are three "key response nodes" that appear sequentially after the brain receives stimulation, representing "perception-processing-evaluation" respectively.

[0133] The event-related potential analysis algorithm is configured to output feature vectors. Then we have:

[0134] ;

[0135] in, This represents the peak amplitude of component P1. This indicates the latency period (peak occurrence time) of component P1. express Peak amplitude of the component, express The incubation period of the ingredients express Peak amplitude of the component, express The incubation period of the ingredients This represents the output feature vector of the event-related unit;

[0136] When otolith stimulation is detected, the adaptive algorithm selects the unit to invoke the time-frequency energy spectrum analysis algorithm to analyze the energy distribution of the EEG signal in the time-frequency domain, focusing on extracting the energy variation features of the theta and gamma bands, and constructing a time-frequency feature vector, specifically:

[0137] ;

[0138] in, express Average energy after band stimulation express The rate of change of energy in the frequency band This represents the constructed time-frequency feature vector;

[0139] When a combined stimulus (including a combination of rotational and head movement stimuli) is detected, the adaptive algorithm selection unit simultaneously invokes the event-related potential analysis algorithm and the time-frequency energy spectrum analysis algorithm to perform joint time-domain and time-frequency domain analysis of the EEG signal and generate a fused feature vector, then:

[0140] ;

[0141] in, This represents a temporal feature vector extracted based on an event-related potential analysis algorithm, used to characterize the structural features of the neural response latency after stimulus locking. This represents the frequency band energy feature vector extracted based on the time-frequency energy spectrum analysis algorithm, used to characterize the energy distribution and dynamic changes of neural activity in different frequency bands. Represents the fused feature vector;

[0142] It should be noted that the fused feature vector is constructed by concatenating the time-domain features and frequency-domain features at the feature level. Each sub-feature vector is normalized before fusion to eliminate the influence of dimensional differences on the fusion result.

[0143] The vestibular function corroborating reference module quantifies the vestibular peripheral reflex signals and vestibular symptom behavior marker signals in the multimodal data signals. Based on the quantification results, it constructs a corroborating parameter set, which includes nystagmus intensity, slow-phase angular velocity, subjective vertical visual deviation angle, and vertigo level. The constructed corroborating parameter set is then spatiotemporally mapped with the EEG features output by the vestibular EEG feature analysis module. Based on the mapping results, a mutually corroborating reference system is constructed. The specific implementation is as follows:

[0144] The vestibular function supporting reference module includes: a nystagmus quantification unit, a balance and posture analysis unit, a symptom quantification unit, and a spatiotemporal correlation mapping unit, as detailed below:

[0145] The nystagmus quantization unit processes the eye movement signals acquired by the vestibular peripheral reflection acquisition unit to extract nystagmus parameters, specifically:

[0146] The nystagmus quantization unit is used to identify the fast and slow phases of nystagmus, then:

[0147] For horizontal eye movement signals The nystagmus quantization unit determines the slow-phase angular velocity corresponding to the moment the fast phase occurs by detecting high-speed saccadic eye movements. The slow eye movement segment between adjacent fast phases is defined as the slow phase;

[0148] Furthermore, regarding the first slow phase intervals Calculate the slow-phase angular velocity for each slow-phase phase. Then we have:

[0149] ;

[0150] in, , This indicates the total number of detected nystagmus cycles. Indicates the first The slow phase angular velocity of each slow phase interval;

[0151] The nystagmus quantization unit sequentially calculates the average slow phase angular velocity. Maximum slow phase angular velocity nystagmus frequency (in Given the time interval between adjacent nystagmus cycles and the direction of nystagmus (horizontal to the left, horizontal to the right, vertical upward, vertical downward, or torsional), and constructing this as a nystagmus quantization vector, we have:

[0152] ;

[0153] in, This represents the constructed nystagmus quantization vector. Indicates the average slow phase angular velocity. Indicates the maximum slow phase angular velocity. Indicates the frequency of nystagmus. The direction of nystagmus is determined based on the sign of the slow-phase angular velocity and the principal direction component. This includes: horizontal to the left, horizontal to the right, vertical upward, vertical downward, or torsional direction.

[0154] The balance and posture analysis unit analyzes the user's postural stability during and after stimulation, as detailed below:

[0155] The balance and attitude analysis unit receives attitude data from the inertial measurement unit and calculates the center of gravity sway trajectory, sway area, and sway velocity, specifically:

[0156] Let the projected coordinates of the center of gravity on the horizontal plane be... ;

[0157] Calculating the length of the trajectory of the center of gravity swaying, we have:

[0158] ;

[0159] in, This represents the calculated length of the trajectory of the center of gravity swaying.

[0160] And calculate the area of ​​the gravitational sway envelope based on the set of points on the projected trajectory of the center of gravity. And the ratio of the area of ​​movement when eyes are open to the area of ​​movement when eyes are closed. ;

[0161] Based on the calculation results, an attitude stability feature vector is constructed, then:

[0162] ;

[0163] in, This represents the constructed posture stability feature vector, used to comprehensively characterize the user's balance control ability and posture stability characteristics under vestibular stimulation conditions.

[0164] The symptom quantification unit quantifies the subjective vertigo scores and physiological data collected by the vestibular symptom behavior labeling unit, specifically as follows:

[0165] The symptom quantification unit calculates the peak value of the subjective vertigo score after stimulation. Average score And the time when the rating reaches its peak ;

[0166] The symptom quantification unit also analyzes vital sign data and calculates the rate of change in heart rate, resulting in:

[0167] ;

[0168] in, To stimulate the pre-resting heart rate, To determine the average heart rate during the post-stimulation response period, This represents the calculated rate of change in heart rate;

[0169] Simultaneously, the change in skin conductance was calculated using the same method. ;

[0170] Based on the calculation results, a symptom quantification vector is constructed, then:

[0171] ;

[0172] in, This represents the constructed symptom quantification vector.

[0173] The spatiotemporal correlation mapping unit receives EEG features (such as event-related potential features and time-frequency features) output by the vestibular EEG feature analysis module, and correlates and maps the supporting parameter set with the EEG features in the time and space dimensions, as follows:

[0174] Based on the processing results of the nystagmus quantification unit, balance and posture analysis unit, and symptom quantification unit, a set of supporting parameters is constructed, which includes:

[0175] ;

[0176] in, This represents the constructed nystagmus quantization vector. This represents the calculated length of the trajectory of the center of gravity swaying. This represents the area of ​​the envelope of the center of gravity during swaying. This represents the constructed symptom quantification vector. This represents the set of supporting parameters constructed.

[0177] The supporting parameter set and EEG features are correlated and mapped in the temporal and spatial dimensions, specifically as follows:

[0178] In the time dimension, the peak times of EEG characteristics were calculated separately. Peak time of symptom scores The temporal correlation between them is as follows:

[0179] ;

[0180] Simultaneously, calculating the Pearson correlation coefficient between the amplitude of EEG characteristics and the slow-phase angular velocity yields:

[0181] ;

[0182] in, This indicates the temporal correlation between the peak time of EEG characteristics and the peak time of symptom scores. The Pearson correlation coefficient represents the relationship between the amplitude of EEG characteristics and the slow phase angular velocity.

[0183] In the spatial dimension, the spatiotemporal correlation mapping unit will output the three-dimensional source distribution from the source localization unit. Spatial overlap analysis was performed with known vestibular cortex anatomical atlases to calculate the spatial distance error between the source activation peak coordinates and the expected vestibular cortex center coordinates. Then:

[0184] ;

[0185] in, Indicates the source activation peak coordinates. Indicates the expected coordinates of the vestibular cortex center. This represents the spatial distance error between the calculated source activation peak coordinates and the expected vestibular cortex center coordinates.

[0186] Based on the results of the time-space correlation analysis, a comprehensive supporting reference vector is constructed, which yields:

[0187] ;

[0188] in, This represents the constructed comprehensive supporting reference vector. This represents the EEG-nystagmus consistency index in the comprehensive corroborating reference vector. This represents the EEG-symptom concordance index in the comprehensive corroborating reference vector. This represents the EEG-source localization credibility index in the comprehensive corroborating reference vector;

[0189] The EEG-nystagmus consistency index, used to determine whether EEG and nystagmus are consistent, includes:

[0190] ;

[0191] The EEG-symptom concordance index, used to determine whether the timing of EEG readings matches the timing of corresponding symptoms, includes:

[0192] ;

[0193] The EEG-source localization reliability index, used to determine whether the source location of EEG signals is reasonable, includes:

[0194] ;

[0195] in, This represents the space tolerance parameter, and the specific value is set by the implementer based on the actual application scenario.

[0196] Furthermore, the consistency index in the comprehensive supporting reference vector is determined by setting a threshold, specifically as follows:

[0197] This embodiment uses the determination of whether the electroencephalogram (EEG) and nystagmus are consistent as an example. The following is an explanation:

[0198] Set a consistency threshold The set consistency threshold is compared with the EEG-nystagmus consistency index, and the consistency between EEG and nystagmus is determined based on the comparison results. Specifically:

[0199] If the comparison results satisfy the formula This indicates that the brainwave and nystagmus are consistent;

[0200] If the comparison results satisfy the formula This indicates that the brainwave and nystagmus are inconsistent.

[0201] The results generation and output module, based on the analysis results of the vestibular EEG feature analysis module and the mutually corroborating reference system constructed by the vestibular function corroborating reference module, determines the user's vestibular EEG analysis and evaluation results. The vestibular EEG analysis and evaluation results include the localization map of the cerebral cortex response induced by different vestibular receptor stimuli, the consistency analysis results of EEG feature indicators and subjective / objective corroborating indicators, and are implemented as follows:

[0202] The results generation and output module includes: an individualized assessment modeling unit, a report generation unit, and a visualization output unit, as detailed below:

[0203] The individualized assessment modeling unit receives the EEG feature vector output by the vestibular EEG feature analysis module and the supporting parameter set and comprehensive supporting reference vector output by the vestibular function supporting reference module. It then determines the user's comprehensive assessment result through a multimodal fusion model, specifically:

[0204] Constructing the evaluation function, we have:

[0205] ;

[0206] in, This represents the constructed evaluation function. , , This represents the weighting coefficient, which is set by the implementers based on clinical trial data.

[0207] The evaluation function outputs multi-dimensional evaluation results, including: localization maps of cortical responses evoked by different vestibular receptor stimuli, consistency analysis results of EEG characteristic indicators and subjective / objective supporting indicators, as detailed below:

[0208] The three-dimensional source distribution output by the source localization unit in the cortical response localization map. The generated, individualized evaluation modeling unit superimposes J(r) onto the standard anatomical template, marking the activation peak coordinates, activation intensity, and activation region volume (the volume of a continuous region where the activation intensity exceeds 50% of the maximum activation intensity).

[0209] The localization map distinguishes different types of vestibular receptor stimulation, specifically:

[0210] For semicircular canal stimulation, the localization map highlights the parietal vestibular cortex (PIVC) and the posterior insular cortex (PIC).

[0211] For otolith stimulation, the localization map highlights the anterior cingulate cortex and the internal parietal sulcus region in the vestibular cortex network;

[0212] The consistency analysis results between EEG feature indicators and subjective / objective corroborating indicators include: calculating the correlation matrix between EEG features and nystagmus parameters, the correlation matrix between EEG features and vertigo scores, and confidence scores generated based on comprehensive corroborating reference vectors;

[0213] Consistency analysis results are presented in numerical tables and statistical charts, with statistically significant feature pairs (e.g., p-value less than 0.05) marked.

[0214] The report generation unit organizes the assessment results generated by the individualized assessment modeling unit into a structured report. The structured report includes: basic user information and testing date, stimulus protocol description (stimulus type, intensity, and timing), EEG feature analysis results (event-related potential waveforms, time-frequency energy spectra, and source localization 3D reconstruction maps), corroborating reference analysis results (nystagmus waveforms, slow-phase angular velocity trend maps, and vertigo score change curves), consistency analysis results (correlation heatmaps and confidence scores), and clinical interpretation suggestions, as detailed below:

[0215] The visualization output unit presents the structured report generated by the report generation unit to clinicians or researchers via a display device;

[0216] The visualization output unit supports interactive viewing functions, specifically: users can rotate and zoom the source-positioned 3D reconstruction image by clicking with the mouse or touching; and can slide along the timeline to view the time-aligned display of EEG waveforms, nystagmus waveforms, and vertigo scores.

[0217] Reports can be exported as PDF, image sequences, or data tables.

[0218] The visualization output unit also provides a comparative analysis view, which supports comparing the current user's assessment results with a reference database of healthy people of the same age group, generating percentile rankings and abnormality indicator markers.

[0219] It should be noted that, through the collaborative work of various modules, the vestibular EEG detection and analysis system has achieved fully automated analysis of the entire process, from multimodal vestibular stimulation, synchronous signal acquisition, adaptive EEG feature analysis, construction of supporting reference system to generation of comprehensive evaluation results. This enables the human brain to perceive and spatially locate information from the peripheral vestibular receptors of both ears from a subjective assessment to an objective quantification, providing auxiliary objective evidence for the diagnosis and rehabilitation of patients with vestibular diseases such as dizziness, vertigo and balance disorders.

[0220] Furthermore, if the aforementioned function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0221] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0222] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0223] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of protection claimed by the present invention.

Claims

1. A vestibular electroencephalogram (EEG) detection and analysis system, characterized in that: include: The vestibular multimodal stimulation module is used to apply vestibular physiological stimulation to the user and record the start time, duration and intensity parameters of each stimulation, including: temperature stimulation unit, rotation stimulation unit, head movement stimulation unit and stimulation timing control unit; The multimodal signal synchronous acquisition module is used to acquire multimodal data signals in parallel in a time-synchronous manner while the vestibular physiological stimulation module applies vestibular physiological stimulation. It includes: a multi-lead EEG acquisition unit, a vestibular peripheral reflex acquisition unit, a vestibular symptom behavior labeling unit, and a multimodal synchronous controller. The vestibular EEG feature analysis module is used to receive multimodal data acquired by the multimodal signal synchronous acquisition module, and adaptively select an analysis algorithm to perform vestibular EEG feature analysis on the multimodal data according to the vestibular physiological stimulation applied by the vestibular multimodal stimulation module. It includes: EEG preprocessing unit, stimulus locking analysis unit, time-frequency analysis unit, source localization unit and adaptive algorithm selection unit. The vestibular function corroborating reference module is used to quantify the vestibular peripheral reflection signals and vestibular symptom behavior marker signals in multimodal data signals. Based on the quantification results, a corroborating parameter set is constructed, including: nystagmus quantification unit, balance and posture analysis unit, symptom quantification unit, and spatiotemporal correlation mapping unit. The results generation and output module is used to determine the user's vestibular EEG analysis and evaluation results based on the analysis results of the vestibular EEG feature analysis module and the mutually corroborating reference system constructed by the vestibular function corroborating reference module.

2. The vestibular EEG detection and analysis system as described in claim 1, characterized in that: The application of vestibular physiological stimulation to the user is specifically as follows: Based on clinical testing needs, the stimulation unit applies standardized vestibular physiological stimulation to users by constructing multiple stimulation paradigms, specifically: The temperature stimulation unit generates a temperature difference by injecting a thermostatic gas or liquid into the user's external auditory canal to excite or inhibit the function of the horizontal semicircular canals. The rotational stimulation unit applies angular acceleration stimulation to the user through a programmable rotating seat to excite the semicircular canals; The head movement stimulation unit stimulates the peripheral vestibular receptors by guiding the user to actively or passively perform head movements in a specific direction and at a specific frequency. The stimulation timing control unit precisely controls the timing of each stimulation unit and generates a stimulation timing reference signal.

3. The vestibular EEG detection and analysis system as described in claim 2, characterized in that: The parallel acquisition of multimodal data signals in a time-synchronized manner is specifically as follows: While applying vestibular physiological stimulation to the vestibular multimodal stimulation module, multimodal data signals are acquired in parallel in a time-synchronized manner. The multimodal data signals include: Multi-lead EEG signals were acquired using an EEG recording device with at least 64 leads. Peripheral vestibular reflex signals are obtained by acquiring vestibular-ocular reflex data through nystagmus / eye movement recording devices; Vestibular symptom behavioral marker signals are obtained by acquiring the user's subjective vertigo score and physiological state data through subjective perception marker devices and vital sign sensors.

4. The vestibular EEG detection and analysis system as described in claim 3, characterized in that: The multimodal synchronization controller is specifically as follows: The multimodal synchronization controller sends synchronous trigger pulses to the multi-lead EEG acquisition unit, the vestibular peripheral reflex acquisition unit, and the vestibular symptom behavior labeling unit to ensure that all acquisition channels start data acquisition at the same time reference. Align the stimulus event timestamps with the data streams of each modality to generate a synchronous multimodal data packet.

5. The vestibular EEG detection and analysis system as described in claim 4, characterized in that: The stimulus-locking analysis unit performs stimulus-locking segmentation on the preprocessed EEG signal based on the stimulus timing reference signal, as follows: For each stimulus event, the stimulus-locked analysis unit extracts the pre-stimulus baseline period and the post-stimulus response period, and constructs a stimulus-locked EEG fragment set, which includes: ； in, The range of values ​​is , The base period length, For the length of the response period, Let represent the time when the i-th stimulus event occurs. This represents the preprocessed EEG signal.

6. The vestibular EEG detection and analysis system as described in claim 5, characterized in that: The time-frequency analysis unit performs time-frequency transformation on the preprocessed EEG signal to extract energy change features in different frequency bands, as detailed below: Calculate the energy changes in each frequency band, including: frequency band, frequency band, frequency band, frequency band and Frequency band, specifically: set up Bandwidth energy is Then we have: ； in, express Energy in the frequency band Indicates time ,frequency Time-frequency energy distribution at the location; according to The method for calculating band energy is to calculate sequentially... frequency band, frequency band, frequency band and Frequency band energy corresponding to a frequency band.

7. The vestibular EEG detection and analysis system as described in claim 6, characterized in that: The adaptive algorithm selection unit adaptively selects the corresponding analysis algorithm based on the currently applied vestibular physiological stimulation type, as detailed below: Identify stimulus type labels, including: semicircular canal temperature stimulation, semicircular canal rotation stimulation, and otolith linear acceleration stimulation; Based on the stimulus type label, the corresponding analysis algorithm is adaptively selected, specifically as follows: When a semicircular canal stimulus is detected, the adaptive algorithm selects the unit to invoke the event-related potential analysis algorithm; When otolith stimulation is detected, the adaptive algorithm selects the unit to call the time-frequency energy spectrum analysis algorithm to analyze the energy distribution of the EEG signal in the time-frequency domain; When a combined stimulus is detected, the adaptive algorithm selection unit simultaneously invokes the event-related potential analysis algorithm and the time-frequency energy spectrum analysis algorithm to perform joint time-domain and time-frequency domain analysis on the EEG signal.

8. The vestibular EEG detection and analysis system as described in claim 7, characterized in that: The quantization process is as follows: Quantization processing involves sequentially processing the vestibular peripheral reflex signals and vestibular symptom behavior marker signals from the multimodal data signals through a nystagmus quantification unit, a balance and posture analysis unit, and a symptom quantification unit. Specifically: The nystagmus quantification unit processes the eye movement signals acquired by the vestibular peripheral reflection acquisition unit and extracts the nystagmus parameters. The balance and attitude analysis unit receives attitude data from the inertial measurement unit and calculates the center of gravity sway trajectory, sway area, and sway speed. The symptom quantification unit quantifies the subjective vertigo scores and physiological data collected by the vestibular symptom behavior labeling unit. A set of supporting parameters is constructed based on the processing results of the nystagmus quantification unit, balance and posture analysis unit, and symptom quantification unit.

9. The vestibular electroencephalogram (EEG) detection and analysis system as described in claim 8, characterized in that: The spatiotemporal correlation mapping unit is used to correlate and map the supporting parameter set with EEG features in the temporal and spatial dimensions, as follows: In the time dimension, the temporal correlation between the peak time of EEG characteristics and the peak time of symptom scores was calculated separately. In the spatial dimension, calculate the spatial distance error between the source activation peak coordinates and the expected vestibular cortex center coordinates; Based on the results of the temporal and spatial correlation analysis, a comprehensive supporting reference vector is constructed.

Images