A system, method, storage medium, and electronic device for assessing consciousness level.

Abstract

This application provides a system, method, storage medium, and electronic device for assessing consciousness level. The system includes: a data acquisition headband for acquiring initial EEG signals and initial cerebral oxygenation signals of a target object after receiving a signal acquisition command from an acquisition control module; an acquisition control module for processing the initial EEG signals and initial cerebral oxygenation signals to obtain digital EEG signals and digital cerebral oxygenation signals; a data processing module for preprocessing the digital EEG signals and digital cerebral oxygenation signals to obtain EEG signal segments and cerebral oxygenation signal segments; acquiring feature parameters based on the EEG signal segments and the cerebral oxygenation signal segments; a consciousness assessment module for fusing the feature parameters to obtain fused features; and inputting the fused features and EEG signal segments into a target consciousness assessment model to obtain the consciousness level of the target object. Embodiments of this application can achieve accurate prediction of the consciousness level of a target object.

Description

Technical Field

[0001] This application relates to the field of consciousness prediction technology, and more specifically, to a system, method, storage medium, and electronic device for evaluating consciousness levels. Background Technology

[0002] Because the consciousness of the target is immeasurable, it is currently difficult to identify effective biomarkers, and therefore the means of consciousness assessment are also very limited.

[0003] Currently, assessing a target's consciousness typically relies on behavioral assessment scales. This involves manually observing, analyzing, and evaluating the target's electroencephalogram (EEG) signals over multiple time periods to determine their level of consciousness. However, this manual assessment method is not only labor-intensive but also results in highly subjective and inaccurate assessments.

[0004] Therefore, how to provide a technical solution for evaluating awareness levels with high accuracy has become an urgent technical problem to be solved. Summary of the Invention

[0005] The purpose of some embodiments of this application is to provide a system, method, storage medium, and electronic device for evaluating consciousness level. The technical solutions of the embodiments of this application can achieve accurate prediction of the consciousness level of a target object with high efficiency.

[0006] In a first aspect, some embodiments of this application provide a system for evaluating consciousness level, comprising: a data acquisition headband and a main controller; the main controller comprising: an acquisition control module, a data processing module, and a consciousness evaluation module; wherein, the data acquisition headband is used to acquire initial EEG signals and initial cerebral oxygenation signals of a target object after receiving a signal acquisition command sent by the acquisition control module; the acquisition control module is used to process the initial EEG signals and the initial cerebral oxygenation signals to obtain digital EEG signals and digital cerebral oxygenation signals; the data processing module is used to preprocess the digital EEG signals and the digital cerebral oxygenation signals to obtain EEG signal segments and cerebral oxygenation signal segments; and to acquire feature parameters based on the EEG signal segments and cerebral oxygenation signal segments; the consciousness evaluation module is used to fuse the feature parameters to obtain fused features; and to input the fused features and the EEG signal segments into a target consciousness evaluation model to obtain the consciousness level of the target object.

[0007] In some embodiments of this application, initial EEG and cerebral oxygenation signals of the target object are collected via a data acquisition headband. These signals are then processed by the acquisition control module and data processing module of the main controller to obtain characteristic parameters. Finally, the characteristic parameters are fused, and the resulting EEG signal segments are input into the target consciousness evaluation model to obtain the target object's consciousness level. This application can assess the consciousness level of a target object using a data acquisition headband and a main controller, eliminating the need for manual intervention, thus improving efficiency and accuracy.

[0008] In some embodiments, the acquisition control module is configured to: amplify, filter, and perform analog-to-digital conversion on the initial EEG signal and the initial cerebral oxygenation signal to obtain the digital EEG signal and the digital cerebral oxygenation signal; and transmit the digital EEG signal and the digital cerebral oxygenation signal to the data processing module via a wireless communication module.

[0009] Some embodiments of this application convert initial EEG signals and initial cerebral blood oxygenation signals through an acquisition control module to obtain corresponding digital signals, providing effective data support for subsequent analysis.

[0010] In some embodiments, the data processing module is used to: filter and denoise the EEG digital signal, remove artifacts, and then segment it according to a sliding step size to obtain EEG signal segments; and filter and eliminate artifacts the cerebral blood oxygenation digital signal, and then segment it according to a sliding step size to obtain cerebral blood oxygenation signal segments.

[0011] Some embodiments of this application use a data processing module to filter and perform other processing on the EEG digital signal and the cerebral blood oxygen digital signal, respectively, to provide effective data support for subsequent accurate analysis.

[0012] In some embodiments, the data processing module is configured to: extract features from the EEG signal segment to obtain EEG feature parameters, wherein the EEG feature parameters include: time-domain features, frequency-domain features, and nonlinear features; calculate the relative concentrations of oxyhemoglobin and deoxyhemoglobin in the cerebral oxygenation signal segment to obtain cerebral oxygen saturation parameters; calculate coupling feature parameters between the EEG signal segment and the cerebral oxygenation signal segment after downsampling the EEG signal segment; and normalize the EEG feature parameters, the cerebral oxygen saturation parameters, and the coupling feature parameters to obtain the feature parameters.

[0013] Some embodiments of this application obtain EEG feature parameters by extracting features from EEG signal segments from different aspects, calculate brain oxygen saturation parameters from brain blood oxygenation signal segments, fuse EEG signal segments and brain blood oxygenation signal segments to obtain coupling feature parameters, and finally normalize the three types of data to obtain feature parameters, which can provide effective data support for subsequent accurate analysis.

[0014] In some embodiments, the system further includes: a display module; the display module is configured to: display the level of consciousness; display EEG data and EEG curves corresponding to the EEG digital signals; and display cerebral blood oxygen data and cerebral blood oxygen curves corresponding to the cerebral blood oxygen digital signals.

[0015] Some embodiments of this application can achieve effective display and monitoring of target object data by setting up a display module.

[0016] In some embodiments, the display module is configured to issue a warning message when the cerebral blood oxygen data exceeds a preset threshold or when the level of consciousness is abnormal.

[0017] Some embodiments of this application provide early warnings of abnormal situations, allowing for timely awareness of changes.

[0018] In some embodiments, the system further includes a training module, which is used to: acquire a training dataset and a test dataset, wherein both the training dataset and the test dataset include: object feature samples and corresponding consciousness level labels for the object feature samples, wherein the object feature samples include: electroencephalogram (EEG) signal sample data, EEG feature samples, brain oxygen saturation samples, and coupling feature samples, and the consciousness level labels include: conscious state, coma state, vegetative state, and minimally conscious state; train an initial consciousness evaluation model using the training dataset to obtain a consciousness evaluation model to be tested; test the consciousness evaluation model to be tested using the test dataset to obtain the target consciousness evaluation model; wherein the target consciousness evaluation model is pre-integrated into the consciousness evaluation module.

[0019] Some embodiments of this application train and test the constructed initial consciousness evaluation model using training and testing datasets to obtain a target consciousness evaluation model that meets the requirements, which can provide model support for subsequent prediction of the consciousness level of the target object.

[0020] In some embodiments, the initial consciousness evaluation model includes: multiple convolutional layers, transposed convolutional layers, pooling layers, fully connected layers, batch normalization layers, dropout layers, and a branch fusion module; the EEG signal sample data is used to input to the multiple convolutional layers, and the fused sample corresponding to the EEG feature sample, brain oxygen saturation sample, and coupled feature sample is input to the transposed convolutional layer.

[0021] Some embodiments of this application can provide effective model support for subsequent training by constructing an initial consciousness evaluation model.

[0022] In some embodiments, the data acquisition headband is connected to the wireless communication module of the acquisition control module; the data acquisition headband includes: an electroencephalogram (EEG) acquisition sensor and a brain oxygenation sensor; the EEG acquisition sensor includes a measuring electrode, a ground electrode, and a reference electrode; the brain oxygenation sensor includes a light source and two photodetectors.

[0023] Some embodiments of this application can achieve accurate data acquisition and wireless data transmission with high efficiency by deploying corresponding sensors on the data acquisition headband and connecting them to the wireless communication module.

[0024] Secondly, some embodiments of this application provide a method for evaluating consciousness level, comprising: after receiving a signal acquisition command sent by the acquisition control module, a data acquisition headband acquires initial EEG signals and initial cerebral oxygenation signals of a target object; the acquisition control module processes the initial EEG signals and the initial cerebral oxygenation signals to obtain digital EEG signals and digital cerebral oxygenation signals; a data processing module preprocesses the digital EEG signals and the digital cerebral oxygenation signals to obtain EEG signal segments and cerebral oxygenation signal segments; characteristic parameters are acquired based on the EEG signal segments and cerebral oxygenation signal segments; a consciousness evaluation module fuses the characteristic parameters to obtain fused features; and the fused features and the EEG signal segments are input into a target consciousness evaluation model to obtain the consciousness level of the target object.

[0025] In some embodiments, the acquisition control module processes the initial EEG signal and the initial cerebral oxygenation signal to obtain digital EEG signals and digital cerebral oxygenation signals, including: amplifying, filtering, and performing analog-to-digital conversion on the initial EEG signal and the initial cerebral oxygenation signal to obtain the digital EEG signals and the digital cerebral oxygenation signals; and transmitting the digital EEG signals and the digital cerebral oxygenation signals to the data processing module through a wireless communication module.

[0026] In some embodiments, the data processing module preprocesses the EEG digital signal and the cerebral blood oxygenation digital signal to obtain EEG signal segments and cerebral blood oxygenation signal segments, including: filtering and denoising the EEG digital signal, removing artifacts, and then segmenting it according to a sliding step size to obtain EEG signal segments; filtering and eliminating artifacts in the cerebral blood oxygenation digital signal, and then segmenting it according to a sliding step size to obtain cerebral blood oxygenation signal segments.

[0027] In some embodiments, obtaining feature parameters based on the EEG signal segment and the cerebral oxygenation signal segment includes: extracting features from the EEG signal segment to obtain EEG feature parameters, wherein the EEG feature parameters include: time-domain features, frequency-domain features, and nonlinear features; calculating the relative concentrations of oxyhemoglobin and deoxyhemoglobin in the cerebral oxygenation signal segment to obtain cerebral oxygen saturation parameters; calculating coupling feature parameters between the EEG signal segment and the cerebral oxygenation signal segment after downsampling the EEG signal segment; and normalizing the EEG feature parameters, the cerebral oxygen saturation parameters, and the coupling feature parameters to obtain the feature parameters.

[0028] In some embodiments, the method further includes: a display module displaying the level of consciousness; displaying EEG data and EEG curves corresponding to the EEG digital signals; and displaying cerebral blood oxygen data and cerebral blood oxygen curves corresponding to the cerebral blood oxygen digital signals.

[0029] In some embodiments, the method further includes: the display module issuing a warning message when the cerebral blood oxygen data exceeds a preset threshold or the level of consciousness is abnormal.

[0030] In some embodiments, the method further includes: acquiring a training dataset and a test dataset, wherein both the training dataset and the test dataset include: object feature samples and corresponding consciousness level labels for the object feature samples, wherein the object feature samples include: electroencephalogram (EEG) signal sample data, EEG feature samples, brain oxygen saturation samples, and coupling feature samples, and the consciousness level labels include: conscious state, coma state, vegetative state, and minimally conscious state; training an initial consciousness evaluation model using the training dataset to obtain a consciousness evaluation model to be tested; and testing the consciousness evaluation model to be tested using the test dataset to obtain the target consciousness evaluation model.

[0031] In some embodiments, the initial consciousness evaluation model includes: multiple convolutional layers, transposed convolutional layers, pooling layers, fully connected layers, batch normalization layers, dropout layers, and a branch fusion module; the EEG signal sample data is used to input to the multiple convolutional layers, and the fused sample corresponding to the EEG feature sample, brain oxygen saturation sample, and coupled feature sample is input to the transposed convolutional layer.

[0032] In some embodiments, the data acquisition headband is connected to the wireless communication module of the acquisition control module; the data acquisition headband includes: an electroencephalogram (EEG) acquisition sensor and a brain oxygenation sensor; the EEG acquisition sensor includes a measuring electrode, a ground electrode, and a reference electrode; the brain oxygenation sensor includes a light source and two photodetectors.

[0033] Thirdly, some embodiments of this application provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, can implement the method described in any embodiment of the first aspect.

[0034] Fourthly, some embodiments of this application provide an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, can implement the method as described in any embodiment of the first aspect.

[0035] Fifthly, some embodiments of this application provide a computer program product, the computer program product including a computer program, wherein the computer program, when executed by a processor, can implement the method described in any embodiment of the first aspect. Attached Figure Description

[0036] To more clearly illustrate the technical solutions of some embodiments of this application, the accompanying drawings used in some embodiments of this application will be briefly described below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 A system diagram for assessing level of consciousness is provided for some embodiments of this application;

[0038] Figure 2 Schematic diagram of the structure of the initial level of consciousness evaluation model provided for some embodiments of this application;

[0039] Figure 3 A flowchart of a method for assessing level of consciousness is provided for some embodiments of this application;

[0040] Figure 4 A schematic diagram of an electronic device provided for some embodiments of this application. Detailed Implementation

[0041] The technical solutions of some embodiments of this application will now be described with reference to the accompanying drawings.

[0042] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0043] To facilitate understanding of this application, the following explanations are provided for some abbreviations:

[0044] EEG stands for Electroencephalogram.

[0045] fNIRS, Functional Near-Infrared Spectroscopy, is a technique for imaging brain function using near-infrared spectroscopy.

[0046] DOC stands for Disorder of Consciousness.

[0047] UWS stands for Unresponsive Wakefulness Syndrome.

[0048] MSC stands for Minimally Conscious State.

[0049] HbR, Reduced hemoglobin, is a type of hemoglobin.

[0050] HbO2, Oxyhemoglobin.

[0051] fMRI, Functional Magnetic Resonance Imaging.

[0052] PET stands for Positron Emission Tomography.

[0053] In related technologies, altered consciousness manifests as wakefulness without clear awareness. The states of such subjects include coma, vegetative state (also known as unresponsive wakefulness syndrome), and minimally conscious state. Vegetative state and minimally conscious state subjects are difficult to distinguish in daily behavior, as well as in neuropathology and physiology. Subjects in a minimally conscious state have greater potential for neural recovery than those in a vegetative state. Therefore, accurate prediction and identification of a subject's level of consciousness is of great significance.

[0054] Currently, existing technologies for assessing the level of consciousness in subjects with DOC (Disease of Constipation) employ two main approaches. One approach is through behavioral scales, such as the Revised Coma Scale-R (CRS-R), which is easy to use and inexpensive. However, it is highly subjective and cannot guarantee the objectivity and accuracy of the assessment results, thus affecting clinical decision-making. Another approach is to use neuroimaging assessment. Currently, neuroimaging methods used for assessing the level of consciousness in DOC subjects include CT, functional magnetic resonance imaging (fMRI), positron emission tomography (PET), and diffusion tensor imaging (DTI). Neuroimaging assessment effectively avoids the influence of subjective factors and can accurately identify subtle changes between multiple assessments, making it valuable in assessing consciousness levels and offering high accuracy. However, it is not easy to implement and is expensive. Additionally, serum biomarkers can be used in some cases, but this technique is not yet mature enough for widespread application.

[0055] As can be seen from the aforementioned related technologies, existing methods for evaluating the consciousness level of target objects cannot simultaneously balance cost and assessment accuracy, resulting in poor adaptability. Furthermore, existing techniques that solely use EEG to assess consciousness suffer from poor anti-interference capabilities and low spatial resolution of the EEG signal, thus failing to yield highly accurate assessment results.

[0056] In view of this, this application combines high spatial resolution fNIRS signals (or brain oxygenation signals) with EEG signals for dual-modal fusion analysis to improve the noise resistance of the entire system, compensate for electromagnetic interference, and provide richer other modal brain features for the target consciousness evaluation model, thereby improving the accuracy and adaptability of consciousness assessment.

[0057] Specifically, some embodiments of this application provide a method for evaluating consciousness level. This method involves acquiring initial electroencephalogram (EEG) signals and initial cerebral oxygenation (COS) signals of a target subject using a data acquisition headband. These signals are then processed and segmented to obtain EEG signal segments and COS signal segments. Characteristic parameters are then calculated and analyzed from these segments. Finally, the fused characteristic parameters and the EEG signal segments are input into a trained target consciousness evaluation model to obtain the target subject's consciousness level. Some embodiments of this application can achieve consciousness assessment of the target subject without human intervention, reducing labor costs while improving the accuracy and efficiency of consciousness level assessment.

[0058] The following is in conjunction with the appendix Figure 1 The overall structure of a system for evaluating the level of consciousness provided by some embodiments of this application is illustrated by way of example.

[0059] like Figure 1As shown, some embodiments of this application provide a system for assessing consciousness level, which includes a data acquisition headband 100 and a main controller 200. The data acquisition headband 100 includes an EEG sensor 110 and a cerebral oxygenation sensor 120. The main controller 200 includes an acquisition control module 210, a data processing module 220, a consciousness assessment module 230, and a display module 240. The acquisition control module 210 includes an EEG controller 211, an infrared light controller 212, and a wireless communication module 213. The data processing module 220 includes a preprocessing module 221, an EEG calculation module 222, and a cerebral oxygenation calculation module 223. The consciousness assessment module 230 includes a feature fusion module 231 and a grading module 232. The display module 240 includes a display 241, a database 242, a power supply 243, and an early warning subsystem 244.

[0060] The following example illustrates... Figure 1 The functions of each module.

[0061] In some embodiments of this application, the data acquisition headband is connected to the wireless communication module of the acquisition control module; the data acquisition headband includes: an electroencephalogram (EEG) acquisition sensor and a brain oxygenation sensor; the EEG acquisition sensor includes a measuring electrode, a ground electrode, and a reference electrode; the brain oxygenation sensor includes a light source and two photodetectors.

[0062] For example, in some embodiments of this application, a portable EEG-fNIRS headband (as a specific example of a data acquisition headband 100) is used for data acquisition from a target object. The EEG-fNIRS headband uses flexible materials to fix the electroencephalogram (EEG) sensor and the blood oxygen saturation sensor to the target object's head to facilitate data acquisition.

[0063] Specifically, the EEG acquisition sensor 110 includes measuring electrode pads, a ground electrode, and a reference electrode for acquiring dual-channel EEG signals. The EEG electrode pads include a patch measuring electrode for detecting EEG signals on the outer side of the forehead above the brow bone, a patch measuring electrode for detecting EEG signals on the inner side of the forehead above the brow bone, a ground electrode located in the center of the forehead, and a reference electrode located at the temple. The brain oxygenation sensor 120 includes a light source and two photodetectors. The light source emits near-infrared light to the target brain via a light-emitting diode, and the photodetectors collect the response signal to obtain the intensity of the near-infrared light after net radiation from the forehead. The light signal is converted into an electrical signal, which is then wirelessly transmitted (e.g., via Bluetooth data transmission) to the infrared light controller 212. Using a shield to cover the brain oxygenation saturation acquisition module can reduce the influence of ambient light. It is understood that, in addition to the data acquisition headband 100 of this application, other devices with similar functions can be used to acquire data; the embodiments of this application are not limited to this.

[0064] In some embodiments of this application, the data acquisition headband 100 is used to acquire the initial electroencephalogram (EEG) signal and initial cerebral blood oxygenation signal of the target object after receiving a signal acquisition instruction sent by the acquisition control module.

[0065] For example, in some embodiments of this application, the wireless communication module 213 of the acquisition control module 210 is connected to the EEG-fNIRS headband and is used to transmit instruction acquisition signals to the sensors of the EEG-fNIRS headband, so that the EEG-fNIRS headband can synchronously acquire the initial EEG signal and the initial fNIRS signal through the EEG acquisition sensor 110 and the cerebral blood oxygen acquisition sensor 120, and transmit the acquired initial EEG signal and the initial fNIRS signal to the acquisition control module 210 of the main controller 200 through the wireless communication module 213.

[0066] Understandably, to ensure the synchronization of the initial EEG signal and the initial fNIRS signal, a photoelectric synchronous detection system can be used in practical applications to synchronously acquire the signals. By synchronously acquiring the initial EEG signal and initial fNIRS signal of the target object using the photoelectric synchronous detection system, the initial EEG signal and initial fNIRS signal of all positions equipped with EEG-fNIRS headbands are acquired at the same time, ensuring the synchronization of signal acquisition.

[0067] In some embodiments of this application, the acquisition control module 210 is used to process the initial EEG signal and the initial cerebral blood oxygenation signal to obtain digital EEG signal and digital cerebral blood oxygenation signal.

[0068] For example, in some embodiments of this application, in order to facilitate subsequent analysis of the initial EEG signal and the initial fNIRS signal, it is necessary to convert them into corresponding digital signals.

[0069] In some embodiments of this application, the acquisition control module 210 is used to amplify, filter, and perform analog-to-digital conversion on the initial EEG signal and the initial cerebral blood oxygenation signal to obtain the digital EEG signal and the digital cerebral blood oxygenation signal; and to send the digital EEG signal and the digital cerebral blood oxygenation signal to the data processing module through a wireless communication module.

[0070] For example, in some embodiments of this application, after receiving the initial EEG signal, the EEG controller 211 amplifies the initial EEG signal through a preamplifier circuit, removes 50Hz power frequency interference through a notch filter, and converts the analog initial EEG signal acquired by the EEG electrodes into the corresponding digital EEG signal through an analog-to-digital converter chip. After receiving the initial fNIRS signal, the infrared light controller 212 amplifies and filters the fNIRS signal through a preamplifier and filter circuit, and then converts the acquired analog fNIRS signal into the corresponding digital brain oxygenation signal through an analog-to-digital converter chip. Finally, the wireless communication module 213 can wirelessly transmit the digital EEG signal and the digital brain oxygenation signal to the data processing module 220.

[0071] It should be understood that in practical applications, in addition to collecting EEG and fNIRS signals, the type of data to be collected can be determined according to the actual situation, such as vital signs information and medical record information of the target object. It is understood that in practical applications, the data to be collected can be selected according to the actual situation, and the embodiments of this application are not limited thereto.

[0072] In some embodiments of this application, the data processing module 220 is used to preprocess the EEG digital signal and the cerebral blood oxygenation digital signal to obtain EEG signal segments and cerebral blood oxygenation signal segments; and to obtain feature parameters based on the EEG signal segments and cerebral blood oxygenation signal segments.

[0073] For example, in some embodiments of this application, the data processing module 220 mainly processes, analyzes and extracts features from the EEG digital signal and the cerebral blood oxygen digital signal. After the data processing module 220 performs filtering, feature extraction, parameter calculation and other operations, it can provide data support for subsequent prediction of consciousness level.

[0074] In some embodiments of this application, the data processing module 220 is used to filter and denoise the EEG digital signal, remove artifacts, and then segment it according to a sliding step size to obtain EEG signal segments; and to filter and eliminate artifacts the cerebral blood oxygenation digital signal and then segment it according to a sliding step size to obtain cerebral blood oxygenation signal segments.

[0075] For example, in some embodiments of this application, the preprocessing module 221 in the data processing module 220 can filter and denoise the EEG digital signal, removing baseline drift and artifacts such as EEG, EMG, and electrosurgical interference, thereby improving the signal-to-noise ratio. Specifically, bandpass filtering can extract the band of interest or the rhythm of the EEG digital signal, notch filtering can remove noise in the interference frequency band such as power frequency interference; baseline drift removal refers to eliminating drift by detrending processing, taking into account the DC component of the EEG digital signal; EEG and ECG artifacts can be removed using a method based on independent component analysis (ICA). Finally, data segmentation is performed through a sliding window of length 60s, with a sliding step size of 1s (as a specific example of the sliding step size), resulting in a series of EEG signal segments (as a specific example of EEG signal segments). The purpose of segmentation is to avoid excessively large amounts of data being calculated at once. Data is segmented using a sliding time window, dividing the data into several segments. A threshold can be set during segmentation; if the amplitude of the EEG digital signal exceeds this threshold, the signal in that segment is considered a motion artifact and not included in the segmented data. The preprocessing module 221 processes the cerebral blood oxygen digital signal in a similar manner to the EEG digital signal processing and segmentation method described above, and will not be elaborated upon here to avoid repetition. It should be understood that the specific values ​​of the sliding time window and sliding step size can be flexibly set in practical applications, and this embodiment does not impose specific limitations here.

[0076] In some embodiments of this application, the data processing module 220 is used to extract features from the EEG signal segments to obtain EEG feature parameters, wherein the EEG feature parameters include: time-domain features, frequency-domain features, and nonlinear features.

[0077] For example, in some embodiments of this application, the EEG calculation module 222 can extract features from the preprocessed segmented EEG data (as a specific example of EEG signal segments) in the time domain, frequency domain, and nonlinear domain, and calculate EEG feature parameters that can be used to characterize the consciousness and cognitive level of the target object.

[0078] Specifically, time-domain features include suppression ratio (SR), frequency-domain features include alpha band relative power and 95% edge frequency (SEF), and nonlinear features include spectral entropy (SE) and permutation Lempel-Ziv complexity (PLZC). Alpha band relative power is the proportion of alpha rhythm (e.g., 8–13 Hz) band power in the total power, reflecting the activity level of the alpha band in the EEG signal. The 95% edge frequency refers to a specific frequency point on the frequency axis of the EEG power spectral density function, where 95% of the total power is below this frequency. It describes the frequency characteristics of the EEG signal and dynamically changes with changes in the level of consciousness.

[0079] Burst suppression in EEG signals typically occurs during brain injury, syncope, or deep anesthesia. SR (Shannon Restriction) measures the degree of brain inhibition: SR = (duration of inhibition / total window duration) × 100%. Spectral entropy, a normalized form of Shannon entropy, is assessed using the power spectrum amplitude component of the time series and quantifies the spectral complexity of the EEG signal. SE (Sequence Entropy) is calculated as follows:

[0080]

[0081] In the formula, P f This represents the frequency value.

[0082] The specific calculation process for PLZC is as follows:

[0083] Perform a permutation process on segmented EEG data s(n) to generate a finite sequence of symbols x(n), which contains a total of m! symbol types. Compute the Lempel-Ziv complexity c(n) on this sequence. The total number of subsequences existing in the sequence x(n) has an upper bound, denoted as L(n):

[0084] L(n)=c(n)[log β c(n)+1]

[0085] Where s(n) is the preprocessed segmented EEG data; m is the number of data points in each motif, and the possible number of motifs is m!; β is the number of symbols in the symbol sequence, which is equal to m! in the symbol arrangement process.

[0086] PLZC can be defined as a normalized c(n):

[0087]

[0088] Where n represents the total length of the symbol sequence. When n is large, PLZC can be simplified to:

[0089]

[0090] In some embodiments of this application, the data processing module 220 is used to calculate the relative concentrations of oxyhemoglobin and deoxyhemoglobin in the brain oxygenation signal segment to obtain brain oxygen saturation parameters.

[0091] For example, in some embodiments of this application, the brain oxygenation calculation module 223 can convert the light intensity attenuation of the brain oxygenation signal segment into hemodynamic parameters by modifying Beer-Lambert's law, and calculate the relative concentration changes of oxyhemoglobin (HbO2) and deoxyhemoglobin (HbR), i.e., brain oxygen saturation parameters.

[0092] In some embodiments of this application, the data processing module 220 is used to calculate the coupling characteristic parameters between the EEG signal segment and the cerebral blood oxygenation signal segment after downsampling the EEG signal segment.

[0093] For example, in some embodiments of this application, in order to better combine the signals of EEG and fNIRS, the correlation coupling value between EEG and fNIRS signals (as a specific example of coupling feature parameters) is calculated as one of the features of consciousness evaluation. The calculation process of the coupling value includes: downsampling the segmented EEG data to ensure the synchronous sampling rate with the cerebral oxygenation signal segment, and calculating the correlation coupling value between EEG and rSO2 (the absolute value of the mixed venous oxygen saturation percentage in the frontal cortex). Specifically, in order to combine EEG and fNIRS signals and find the coupling coefficient, this application constructs a feature fusion algorithm based on canonical correlation analysis of kernel functions to calculate neurovascular coupling and obtain the maximum correlation between different modalities (i.e., EEG and rSO2). According to the different consciousness states of the target object, its correlation exhibits different patterns. The kernel function is used to map the signal to the kernel space, and a criterion function describing their correlation is established between the two sets of feature vectors (i.e., EEG and rSO2). The combined canonical correlation features are extracted through a given feature fusion strategy for classification and recognition, thereby obtaining the coupling value between EEG and rSO2. It is understood that, in addition to the feature fusion algorithm of the canonical correlation analysis of the kernel function provided in this application, other methods, including Spearman correlation analysis, can be selected to calculate the coupling value according to the number of channels of the actual acquired signal. It should be understood that the embodiments of this application are not limited to this.

[0094] In some embodiments of this application, the data processing module 220 is used to normalize the EEG feature parameters, the brain oxygen saturation parameters, and the coupling feature parameters to obtain the feature parameters.

[0095] For example, in some embodiments of this application, the EEG feature parameters, brain oxygen saturation parameters, and brain-computer oxygen coupling features (i.e., coupling values) calculated in the data processing module 220 are normalized. This ensures the accuracy of the final output consciousness level and avoids the impact of differences in features at different scales on the gradient descent process of the initial consciousness evaluation model. The normalization method can be selected according to the actual situation.

[0096] Specifically, since EEG feature parameters, brain oxygen saturation parameters, and brain-computer interface oxygen coupling features are data at different scales, all of these features are normalized using Z-Score standards to ensure that all features can be evaluated on the same dimension. Normalization is performed using the following formula to calculate the corresponding values ​​for EEG feature parameters, brain oxygen saturation parameters, and brain-computer interface oxygen coupling features:

[0097] Zscore=(x-μ) / σ

[0098] Where x represents the EEG characteristic parameter, brain oxygen saturation parameter, or brain-computer oxygen coupling characteristic, μ represents the average value of the EEG characteristic parameter, brain oxygen saturation parameter, or brain-computer oxygen coupling characteristic, σ represents the standard deviation of the EEG characteristic parameter, brain oxygen saturation parameter, or brain-computer oxygen coupling characteristic, and Zscore represents the normalized characteristic value.

[0099] In some embodiments of this application, the consciousness evaluation module 230 is used to fuse the feature parameters to obtain fused features; and input the fused features and the EEG signal segments into the target consciousness evaluation model to obtain the consciousness level of the target object.

[0100] For example, in some embodiments of this application, the feature fusion module 231 can perform feature fusion on normalized EEG feature parameters, brain oxygen saturation parameters, or brain-computer oxygen coupling features to obtain a feature matrix (as a specific example of fused features). The feature matrix and EEG signal segments are input together into the trained target consciousness evaluation model in the grading module 232 to output the consciousness level of the target object. The consciousness level may include: awake, comatose, vegetative state (VS), and minimally conscious state (MCS), etc.

[0101] The following example illustrates the process of obtaining the goal awareness evaluation model.

[0102] It should be noted that the target awareness evaluation model can be obtained by pre-training an initial awareness evaluation model through a training module, and then deploying or integrating it into the awareness evaluation module to facilitate subsequent prediction of the target object's awareness level. The training module can be integrated into the awareness level evaluation system or can be independent of it. The specific configuration can be determined according to actual circumstances, and this application does not impose specific limitations on the embodiments herein.

[0103] Before training the model, in some embodiments of this application, an initial consciousness evaluation model needs to be constructed first. The initial consciousness evaluation model includes: multiple convolutional layers, transposed convolutional layers, pooling layers, fully connected layers, batch normalization layers, dropout layers, and a branch fusion module; EEG signal sample data is used as input to multiple convolutional layers, and the fused samples corresponding to EEG feature samples, brain oxygen saturation samples, and coupled feature samples are input to the transposed convolutional layer.

[0104] For example, in some embodiments of this application, such as Figure 2 The diagram shows the structure of the initial consciousness evaluation model. Figure 2 It can be seen that the initial consciousness evaluation model includes c1 convolutional layers, c2 transposed convolutional layers, c3 pooling layers, c4 fully connected layers, c5 batch-normalization layers, c6 dropout layers, and 1 branch fusion structure (i.e., branch fusion module). Among them, c1, c2, c3, c4, and c5 can be flexibly set according to the amount and characteristics of data. During training, the hyperparameters can be updated using the grid search method to find the optimal model. Specifically, each convolutional layer and transposed convolutional layer extract effective features from the model's input (e.g., EEG data, feature fusion data) through convolutional computation; each pooling layer can choose between max pooling or average pooling to reduce parameter computation while preserving the spatial relationship of features; fully connected layers are used to reduce the dimensionality of high-dimensional data into one-dimensional vectors; each batch-normalization layer distributes the neurons of each neural network layer into a standard normal distribution with a mean of 0 and a variance of 1; each dropout layer randomly removes a portion of neurons from the network according to a set amount to reduce overfitting; the branch fusion structure fuses the features extracted by the convolutional neural network together through vector addition or vector concatenation.

[0105] To enable the evaluation of the consciousness level of a target object, in some embodiments of this application, the consciousness level evaluation system further includes: a training module (not shown in the figure); the training module is used to: acquire a training dataset and a test dataset, wherein both the training dataset and the test dataset include: object feature samples and consciousness level labels corresponding to the object feature samples, wherein the object feature samples include: EEG signal sample data, EEG feature samples, brain oxygen saturation samples, and coupling feature samples, and the consciousness level labels include: conscious state, coma state, vegetative state, and minimally conscious state; train an initial consciousness evaluation model using the training dataset to obtain a consciousness evaluation model to be tested; test the consciousness evaluation model to be tested using the test dataset to obtain the target consciousness evaluation model; wherein the target consciousness evaluation model is pre-integrated into the consciousness evaluation module.

[0106] For example, in some embodiments of this application, EEG and fNIRS signal samples from different objects can be collected via a data acquisition headband 100. These samples are then processed according to the data processing methods described above by the acquisition control module 210 and the data processing module 220 to obtain a series of EEG signal sample data, EEG features (as a specific example of EEG feature samples), cerebral blood oxygenation features (as a specific example of cerebral oxygen saturation samples), and coupling features (as a specific example of coupling feature samples). Then, experts analyze and label the consciousness level of each object. The above sample data processing operations yield a dataset, which is then divided into a training dataset and a test dataset. The initial consciousness evaluation model can be trained using the training and test datasets to determine the weight values ​​and neural network hyperparameters in the final target consciousness evaluation model for subsequent prediction of consciousness levels.

[0107] The specific training process includes: using the training dataset as... Figure 2 The initial consciousness evaluation model is trained by setting an initial learning rate and optimizing its parameters using an optimization algorithm, such as Adam. The test dataset is then used as input to the trained consciousness evaluation model. The structure and parameters of the model are adjusted based on the generated accuracy curve and loss characteristics. The cross-entropy loss function is used to evaluate the difference between the probability distribution of the trained model and the true distribution. This training and testing process is repeated until the target consciousness evaluation model with the optimal accuracy in classifying consciousness states is obtained. This target consciousness evaluation model serves as a consciousness level classification model capable of assessing the consciousness state of DOC objects. It should be understood that the optimization algorithm and loss function can be selected based on the actual training situation, and this application does not impose specific limitations on them.

[0108] In some embodiments of this application, the display module 240 is used to: display the level of consciousness; display EEG data and EEG curves corresponding to the EEG digital signals; and display cerebral blood oxygen data and cerebral blood oxygen curves corresponding to the cerebral blood oxygen digital signals.

[0109] For example, in some embodiments of this application, the display 241 can display the consciousness level of the target object output by the consciousness evaluation module 230. The display 241 can also be connected to the acquisition control module 210 to generate and display data and graphs corresponding to the EEG digital signal and the cerebral blood oxygen digital signal.

[0110] In some embodiments of this application, the display module 240 is used to issue a warning message when the cerebral blood oxygen data exceeds a preset threshold or the level of consciousness is abnormal.

[0111] For example, in some embodiments of this application, the warning subsystem 244 of the display module 240 is further equipped with a parameter data threshold (as a specific example of a preset threshold), and when real-time data (such as cerebral blood oxygen data or consciousness level fluctuations, frequent fluctuations, etc.) exceeds the parameter data threshold, an alarm is automatically activated to issue a sound or light warning (as a specific example of a warning message) to promptly remind the monitoring personnel. For example, when the monitored data is cerebral blood oxygen data, the parameter data threshold can be a percentage value; when the monitored data is consciousness level, the parameter data threshold can be a status level. In addition, the database 242 in the display module 240 can store relevant data such as the target object's basic personal information. The power supply 243 can be a wired power source and can charge a power bank; in case of emergencies, the power stored in the power bank can be used.

[0112] As can be seen from the consciousness level evaluation system described in this application, the data acquisition headband uses flexible material with EEG and cerebral oxygenation sensors. The main controller controls the detection time of the EEG and cerebral oxygenation sensors, enabling real-time acquisition and synchronous reading of two key brain information parameters: EEG and cerebral oxygenation. In application, this allows for accurate prediction of the target's consciousness state over a given period. Simultaneously, the use of a Bluetooth communication module (as a specific example of wireless communication module 213) ensures low power consumption for wireless data transmission, guaranteeing the operation of battery-powered devices. The wireless headband detection method eliminates the limitations of wires and cables on the range and mode of movement, providing a more portable consciousness level assessment for DOC (Digital Oriented Individual) subjects. After synchronously acquiring the forehead EEG signal, it can be analyzed in conjunction with the cerebral oxygenation signal from this area. EEG signals are generated by neuronal firing, while cerebral oxygenation signals are generated by local blood oxygen metabolism in the brain; both can play different roles in consciousness level determination. This invention extracts and jointly analyzes features from two different modalities of brain signals, calculating coupled features. Different weights are trained in the target consciousness assessment model using EEG data and cerebral oxygen saturation data from different brain regions, providing different feature information in different states of consciousness to achieve more accurate consciousness level discrimination. This invention constructs a deep learning model based on multimodal feature fusion (i.e., the initial consciousness assessment model) for the consciousness level classification of DOC objects. The convolutional and deconvolutional layers of the convolutional neural network in the deep learning model automatically learn the abstract representation features of the training dataset and parameters, fully extracting information present in the time series, effectively achieving effective analysis of EEG signals, realizing accurate consciousness level classification, and providing assistance for clinical research.

[0113] The following is in conjunction with the appendix Figure 3 The implementation process of consciousness level evaluation provided by some embodiments of this application is illustrated by way of example.

[0114] Please see the appendix Figure 3 , Figure 3 A flowchart of a method for evaluating consciousness level is provided for some embodiments of this application. The method includes: S310, after receiving a signal acquisition command from the acquisition control module, the data acquisition headband acquires initial EEG signals and initial cerebral oxygenation signals of the target object. S320, the acquisition control module processes the initial EEG signals and initial cerebral oxygenation signals to obtain digital EEG signals and digital cerebral oxygenation signals. S330, the data processing module preprocesses the digital EEG signals and digital cerebral oxygenation signals to obtain EEG signal segments and cerebral oxygenation signal segments; based on the EEG signal segments and cerebral oxygenation signal segments, feature parameters are obtained. S340, the consciousness evaluation module fuses the feature parameters to obtain fused features; the fused features and the EEG signal segments are input into a target consciousness evaluation model to obtain the consciousness level of the target object.

[0115] The above process is illustrated below by example.

[0116] In some embodiments of this application, S320 may include: amplifying, filtering, and performing analog-to-digital conversion on the initial EEG signal and the initial cerebral oxygenation signal to obtain the digital EEG signal and the digital cerebral oxygenation signal; and transmitting the digital EEG signal and the digital cerebral oxygenation signal to the data processing module via a wireless communication module.

[0117] In some embodiments of this application, S330 may include: filtering and denoising the EEG digital signal, removing artifacts, and then segmenting it according to a sliding step size to obtain EEG signal segments; filtering and eliminating artifacts in the cerebral blood oxygenation digital signal, and then segmenting it according to a sliding step size to obtain cerebral blood oxygenation signal segments.

[0118] In some embodiments of this application, S330 may include: extracting features from the EEG signal segment to obtain EEG feature parameters, wherein the EEG feature parameters include: time-domain features, frequency-domain features, and nonlinear features; calculating the relative concentrations of oxyhemoglobin and deoxyhemoglobin in the brain oxygenation signal segment to obtain brain oxygen saturation parameters; downsampling the EEG signal segment and calculating coupling feature parameters between the EEG signal segment and the brain oxygenation signal segment; and normalizing the EEG feature parameters, the brain oxygen saturation parameters, and the coupling feature parameters to obtain the feature parameters.

[0119] In some embodiments of this application, the method for evaluating the level of consciousness may further include: displaying the level of consciousness using a display module; displaying EEG data and EEG curves corresponding to the EEG digital signals; and displaying cerebral blood oxygen data and cerebral blood oxygen curves corresponding to the cerebral blood oxygen digital signals.

[0120] In some embodiments of this application, the method for evaluating the level of consciousness may further include: the display module issuing a warning message when the cerebral blood oxygen data exceeds a preset threshold or the level of consciousness is abnormal.

[0121] In some embodiments of this application, the method for evaluating consciousness level may further include: a training module acquiring a training dataset and a test dataset, wherein both the training dataset and the test dataset include: object feature samples and consciousness level labels corresponding to the object feature samples, wherein the object feature samples include: electroencephalogram (EEG) signal sample data, EEG feature samples, brain oxygen saturation samples, and coupling feature samples, and the consciousness level labels include: conscious state, coma state, vegetative state, and minimally conscious state; training an initial consciousness evaluation model using the training dataset to obtain a consciousness evaluation model to be tested; testing the consciousness evaluation model to be tested using the test dataset to obtain the target consciousness evaluation model; wherein the target consciousness evaluation model is pre-integrated into the consciousness evaluation module.

[0122] In some embodiments of this application, the initial consciousness evaluation model includes: multiple convolutional layers, transposed convolutional layers, pooling layers, fully connected layers, batch normalization layers, dropout layers, and a branch fusion module; the EEG signal sample data is used to input to the multiple convolutional layers, and the fused samples corresponding to the EEG feature samples, brain oxygen saturation samples, and coupled feature samples are input to the transposed convolutional layer.

[0123] In some embodiments of this application, the data acquisition headband is connected to the wireless communication module of the acquisition control module; the data acquisition headband includes: an electroencephalogram (EEG) acquisition sensor and a brain oxygenation sensor; the EEG acquisition sensor includes a measuring electrode, a ground electrode, and a reference electrode; the brain oxygenation sensor includes a light source and two photodetectors.

[0124] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process of the method described above can be referred to the corresponding process in the aforementioned system, and will not be elaborated further here.

[0125] Some embodiments of this application also provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, can perform the operation of any of the methods corresponding to the methods provided in the above embodiments.

[0126] Some embodiments of this application also provide a computer program product, which includes a computer program, wherein when the computer program is executed by a processor, it can implement the operation of any of the methods corresponding to the above embodiments provided in the above embodiments.

[0127] like Figure 4 As shown, some embodiments of this application provide an electronic device 400, which includes a memory 410, a processor 420, and a computer program stored in the memory 410 and executable on the processor 420. When the processor 420 reads the program from the memory 410 via a bus 430 and executes the program, it can implement the methods of any of the above embodiments.

[0128] Processor 420 can process digital signals and may include various computing architectures. For example, it may be a complex instruction set computer architecture, a reduced instruction set computer architecture, or an architecture that implements multiple instruction set combinations. In some examples, processor 420 may be a microprocessor.

[0129] Memory 410 can be used to store instructions executed by processor 420 or data related to the execution of instructions. These instructions and / or data may include code for implementing some or all of the functions of one or more modules described in the embodiments of this application. The processor 420 of this disclosure embodiment can be used to execute instructions in memory 410 to implement the methods shown above. Memory 410 includes dynamic random access memory, static random access memory, flash memory, optical memory, or other memories well known to those skilled in the art.

[0130] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application. It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0131] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0132] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

Claims

1. A system for evaluating the level of consciousness, characterized in that, include: The system includes a data acquisition headband and a main controller; the main controller comprises: an acquisition control module, a data processing module, a consciousness evaluation module, and a training module; wherein, The data acquisition headband is used to acquire the initial EEG signal and initial cerebral blood oxygen signal of the target object after receiving the signal acquisition command sent by the acquisition control module. The acquisition and control module is used to process the initial EEG signal and the initial cerebral blood oxygenation signal to obtain digital EEG signal and digital cerebral blood oxygenation signal. The data processing module is used to preprocess the EEG digital signal and the cerebral blood oxygenation digital signal to obtain EEG signal segments and cerebral blood oxygenation signal segments; and to obtain feature parameters based on the EEG signal segments and cerebral blood oxygenation signal segments. The step of obtaining feature parameters based on the electroencephalogram (EEG) signal segment and the cerebral blood oxygenation signal segment includes: EEG feature parameters are obtained by extracting features from the EEG signal segment, brain oxygen saturation parameters are calculated from the brain blood oxygen signal segment, coupling feature parameters are obtained by fusing the EEG signal segment and the brain blood oxygen signal segment, and the feature parameters are obtained by normalizing the EEG feature parameters, the brain oxygen saturation parameters and the coupling feature parameters. The consciousness evaluation module is used to fuse the feature parameters to obtain fused features; and input the fused features and the EEG signal segments into the target consciousness evaluation model to obtain the consciousness level of the target object. The training module is used to: acquire a training dataset and a test dataset, wherein both the training dataset and the test dataset include: object feature samples and consciousness level labels corresponding to the object feature samples, wherein the object feature samples include: EEG signal sample data, EEG feature samples, brain oxygen saturation samples and coupling feature samples, and the consciousness level labels include: conscious state, coma state, vegetative state and minimal consciousness state; The initial consciousness evaluation model was trained using the training dataset to obtain the consciousness evaluation model to be tested; The target consciousness evaluation model is obtained by testing the test dataset; wherein the target consciousness evaluation model is pre-integrated into the consciousness evaluation module.

2. The system as described in claim 1, characterized in that, The acquisition and control module is used for: The initial EEG signal and the initial cerebral blood oxygenation signal are amplified, filtered, and converted from analog to digital to obtain the digital EEG signal and the digital cerebral blood oxygenation signal. The EEG digital signal and the cerebral blood oxygen digital signal are transmitted to the data processing module via the wireless communication module.

3. The system as described in claim 1 or 2, characterized in that, The data processing module is used for: After filtering, noise reduction, and artifact removal of the EEG digital signal, it is segmented according to the sliding step size to obtain EEG signal segments. After filtering and artifact removal, the digital brain oxygenation signal is segmented according to the sliding step size to obtain brain oxygenation signal segments.

4. The system as described in claim 1 or 2, characterized in that, The data processing module is used for: Feature extraction is performed on the EEG signal segments to obtain EEG feature parameters, wherein the EEG feature parameters include: time-domain features, frequency-domain features, and nonlinear features; The relative concentrations of oxyhemoglobin and deoxyhemoglobin in the brain oxygenation signal segment are calculated to obtain the brain oxygen saturation parameter. After downsampling the EEG signal segment, the coupling characteristic parameters between the EEG signal segment and the cerebral blood oxygenation signal segment are calculated; The EEG characteristic parameters, the brain oxygen saturation parameters, and the coupling characteristic parameters are normalized to obtain the characteristic parameters.

5. The system as described in claim 1, characterized in that, The initial consciousness evaluation model includes: multiple convolutional layers, transposed convolutional layers, pooling layers, fully connected layers, batch normalization layers, dropout layers, and a branch fusion module; the EEG signal sample data is used to input to the multiple convolutional layers, and the fused samples corresponding to the EEG feature samples, brain oxygen saturation samples, and coupled feature samples are input to the transposed convolutional layer.

6. The system as described in claim 1 or 2, characterized in that, The data acquisition headband is connected to the wireless communication module of the acquisition control module; the data acquisition headband includes: an electroencephalogram (EEG) acquisition sensor and a brain oxygenation sensor; the EEG acquisition sensor includes a measuring electrode, a ground electrode, and a reference electrode; the brain oxygenation sensor includes a light source and two photodetectors.

7. A method for evaluating the level of consciousness, characterized in that, A system for assessing level of consciousness as described in any one of claims 1-6, the method comprising: After receiving the signal acquisition command sent by the acquisition control module, the data acquisition headband acquires the initial EEG signal and initial cerebral blood oxygenation signal of the target object; The acquisition and control module processes the initial EEG signal and the initial cerebral blood oxygenation signal to obtain digital EEG signal and digital cerebral blood oxygenation signal; The data processing module preprocesses the EEG digital signal and the cerebral blood oxygenation digital signal to obtain EEG signal segments and cerebral blood oxygenation signal segments; and obtains feature parameters based on the EEG signal segments and cerebral blood oxygenation signal segments. The consciousness evaluation module fuses the feature parameters to obtain fused features; the fused features and the EEG signal segments are input into the target consciousness evaluation model to obtain the consciousness level of the target object.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein the computer program is executed by a processor to perform the method as described in claim 7.

9. An electronic device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the computer program is executed by the processor to perform the method of claim 7.

Images