Electroencephalogram signal processing method and related device thereof

Abstract

The invention relates to an electroencephalogram signal processing method and a related device thereof. The electroencephalogram signal processing method comprises the steps that a left brain electroencephalogram signal and a right brain electroencephalogram signal of a target object are collected; performing asymmetric spatial feature extraction on the left brain electroencephalogram signal and the right brain electroencephalogram signal to obtain a left brain electroencephalogram signal feature and a right brain electroencephalogram signal feature; and based on a preset nearest neighbor algorithm, comparing the left brain electroencephalogram signal features and the right brain electroencephalogram signal features with electroencephalogram signal features in a preset signal feature library, and determining whether the left brain electroencephalogram signal and the right brain electroencephalogram signal are epilepsy electroencephalogram signals or not. According to the scheme provided by the invention, the epilepsy electroencephalogram signal recognition accuracy can be improved, the signal collaborative analysis capability of the left and right brain regions is enhanced, and the recognition accuracy of the epilepsy state of the patient is improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a method for processing electroencephalogram (EEG) signals and related devices. Background Art

[0002] Epilepsy is a chronic disease caused by abnormal electrical activity of brain neurons, leading to transient brain dysfunction.

[0003] In related technologies, whether a patient has an epileptic seizure is generally determined by analyzing the patient's EEG signals. However, the structure of the functional areas in the brain is not symmetrical, so the acquired EEG signals are not similar, which can easily lead to misjudgment that the patient is in an epileptic state when analyzing the EEG signals, thereby reducing the accuracy of epilepsy detection. Summary of the Invention

[0004] In order to solve or partially solve the problems existing in the related technology, the present application provides a method for processing EEG signals and related devices, which can improve the accuracy of epileptic EEG signal recognition, enhance the ability of collaborative analysis of signals from the left and right brain regions, and improve the accuracy of identifying the patient's epileptic state.

[0005] The first aspect of the present application provides a method for processing EEG signals, which is characterized by comprising: collecting the left brain EEG signals and the right brain EEG signals of the target object; performing asymmetric spatial feature extraction on the left brain EEG signals and the right brain EEG signals respectively to obtain left brain EEG signal features and right brain EEG signal features; based on a preset nearest neighbor algorithm, comparing the left brain EEG signal features and the right brain EEG signal features with EEG signal features in a preset signal feature library to determine whether the left brain EEG signals and the right brain EEG signals are epileptic EEG signals.

[0006] In combination with the first aspect, in a possible implementation of the first aspect, the asymmetric spatial feature extraction is performed on the left brain EEG signal and the right brain EEG signal respectively to obtain the left brain EEG signal features and the right brain EEG signal features, including: performing time-frequency analysis on the left brain EEG signal to obtain the time domain features and frequency domain features of the left brain EEG signal; performing time-frequency analysis on the right brain EEG signal to obtain the time domain features and frequency domain features of the right brain EEG signal.

[0007] In combination with the first aspect, in a possible implementation of the first aspect, the preset nearest neighbor algorithm compares the left brain EEG signal features and the right brain EEG signal features with the EEG signal features in a preset signal feature library to determine whether the left brain EEG signal and the right brain EEG signal are epileptic EEG signals, including: determining the K value of the nearest neighbor algorithm based on the distribution density of epileptic EEG signals in the signal feature library; if the proportion of epilepsy signals in the K nearest neighbors of the left brain EEG signal features or the right brain EEG signal features exceeds a preset threshold, then it is determined to be an epileptic EEG signal.

[0008] In combination with the first aspect, in a possible implementation of the first aspect, determining the K value of the nearest neighbor algorithm based on the distribution density of the epileptic EEG signal in the signal feature library includes: when the distribution density indicates a high-density area, determining that the K value of the nearest neighbor algorithm is less than a first threshold; when the distribution density indicates a low-density area, determining that the K value of the nearest neighbor algorithm is greater than a second threshold.

[0009] In combination with the first aspect, in a possible implementation of the first aspect, the preset nearest neighbor algorithm uses the Mahalanobis distance as a distance metric, and also includes: based on the covariance matrix of the epileptic EEG signals and the non-epileptic EEG signals in the signal feature library, calculating the Mahalanobis distance between the left brain EEG signal features, the right brain EEG signal features and the epileptic EEG signals and the non-epileptic EEG signals in the signal feature library; based on the minimum Mahalanobis distance, voting on the left brain EEG signal features and the right brain EEG signal features to determine whether the left brain EEG signal features and the right brain EEG signal features are epileptic EEG signals.

[0010] In combination with the first aspect, in a possible implementation of the first aspect, the method further includes: performing baseline correction and power frequency interference filtering on the left brain EEG signal and the right brain EEG signal.

[0011] In combination with the first aspect, in a possible implementation method of the first aspect, it also includes: when the left brain EEG signal or the right brain EEG signal is determined to be an epilepsy signal, generating an audible and visual alarm and alarm information; associating the alarm information with the preset data table of the target object to generate a diagnostic recommendation report.

[0012] The second aspect of the present application provides an EEG signal processing device, including an acquisition module for acquiring the left-brain EEG signals and the right-brain EEG signals of a target object; a processing module for performing asymmetric spatial feature extraction on the left-brain EEG signals and the right-brain EEG signals, respectively, to obtain left-brain EEG signal features and right-brain EEG signal features; and a determination module for comparing the left-brain EEG signal features and the right-brain EEG signal features with EEG signal features in a preset signal feature library based on a preset nearest neighbor algorithm to determine whether the left-brain EEG signals and the right-brain EEG signals are epileptic EEG signals.

[0013] In combination with the second aspect, in a possible implementation of the second aspect, the processing module is also used to perform time-frequency analysis on the left brain EEG signal to obtain the time domain characteristics and frequency domain characteristics of the left brain EEG signal; and perform time-frequency analysis on the right brain EEG signal to obtain the time domain characteristics and frequency domain characteristics of the right brain EEG signal.

[0014] In combination with the second aspect, in a possible implementation of the second aspect, the determination module is also used to determine the K value of the nearest neighbor algorithm based on the distribution density of the epileptic EEG signals in the signal feature library; if the proportion of epileptic signals in the K nearest neighbors of the left brain EEG signal feature or the right brain EEG signal feature exceeds a preset threshold, it is determined to be an epileptic EEG signal.

[0015] In combination with the second aspect, in a possible implementation of the second aspect, the determination module is also used to determine that the K value of the nearest neighbor algorithm is less than a first threshold when the distribution density indication is a high-density area; and when the distribution density indication is a low-density area, determine that the K value of the nearest neighbor algorithm is greater than a second threshold.

[0016] In combination with the second aspect, in a possible implementation of the second aspect, the determination module is also used to calculate the Mahalanobis distance between the left brain EEG signal features, the right brain EEG signal features and the epileptic EEG signals and non-epileptic EEG signals in the signal feature library based on the covariance matrix of the epileptic EEG signals and the non-epileptic EEG signals in the signal feature library; based on the minimum Mahalanobis distance, the left brain EEG signal features and the right brain EEG signal features are voted to determine whether the left brain EEG signal features and the right brain EEG signal features are epileptic EEG signals.

[0017] In combination with the second aspect, in a possible implementation of the second aspect, the processing module is further used to perform baseline correction and power frequency interference filtering on the left brain EEG signal and the right brain EEG signal.

[0018] In combination with the second aspect, in a possible implementation of the second aspect, the processing module is also used to generate an audible and visual alarm and alarm information when the left brain EEG signal or the right brain EEG signal is determined to be an epilepsy signal; and associate the alarm information with a preset data table of the target object to generate a diagnostic recommendation report.

[0019] A third aspect of the present application provides an electronic device, including:

[0020] processor; and

[0021] The memory stores executable codes thereon, and when the executable codes are executed by the processor, the processor is caused to execute the method described above.

[0022] A fourth aspect of the present application provides a computer-readable storage medium having executable code stored thereon. When the executable code is executed by a processor of an electronic device, the processor is caused to execute the method described above.

[0023] A fifth aspect of the present application provides a computer program product, comprising a computer program / instruction, which implements the method described above when executed by a processor.

[0024] The technical solution provided by this application may have the following beneficial effects:

[0025] The present application provides an EEG signal processing method and related devices, including: collecting the left-brain EEG signals and the right-brain EEG signals of a target object; performing asymmetric spatial feature extraction on the left-brain EEG signals and the right-brain EEG signals respectively to obtain left-brain EEG signal features and right-brain EEG signal features; based on a preset nearest neighbor algorithm, comparing the left-brain EEG signal features and the right-brain EEG signal features with EEG signal features in a preset signal feature library to determine whether the left-brain EEG signals and the right-brain EEG signals are epileptic EEG signals, which can improve the accuracy of epileptic EEG signal recognition, enhance the collaborative analysis capability of left and right brain region signals, and improve the accuracy of recognition of the patient's epileptic state.

[0026] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The above and other objects, features and advantages of the present application will become more apparent by describing in more detail exemplary embodiments of the present application in conjunction with the accompanying drawings, wherein like reference numerals generally represent like components in the exemplary embodiments of the present application.

[0028] Figure 1 1 is a flow chart of a method for processing an EEG signal according to an embodiment of the present application;

[0029] Figure 2 Schematic diagram of the structure of the brain electrical signal processing device shown in the embodiment of the present application;

[0030] Figure 3 It is a structural diagram of an electronic device shown in an embodiment of the present application. DETAILED DESCRIPTION

[0031] The following describes embodiments of the present application in more detail with reference to the accompanying drawings. Although the accompanying drawings illustrate embodiments of the present application, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments described herein. Rather, these embodiments are provided to make the present application more thorough and complete, and to fully convey the scope of the present application to those skilled in the art.

[0032] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. The singular forms "a," "an," and "the" used in this application and the appended claims are also intended to include plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0033] It should be understood that although the terms "first", "second", "third", etc. may be used in this application to describe various information, this information should not be limited to these terms. These terms are only used to distinguish information of the same type from each other. For example, without departing from the scope of this application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0034] Epilepsy is a chronic disease caused by abnormal electrical activity of brain neurons, leading to transient brain dysfunction.

[0035] Among related technologies, the field of EEG signal analysis has long focused on emotion recognition and brain function research. Generally, the EEG signals of the patient's brain are analyzed to determine whether the patient has an epileptic seizure. However, the structure of the brain is not symmetrical, so the acquired EEG signals are not similar, which can easily lead to the misjudgment of the patient's epileptic state when analyzing the EEG signals. This method is difficult to capture the differential characteristics of left and right brain activities, reducing the accuracy of epilepsy detection.

[0036] In response to the above problems, the embodiments of the present application provide a method for processing EEG signals and related devices, which can improve the accuracy of identifying epileptic EEG signals, enhance the ability to collaboratively analyze signals from left and right brain regions, and improve the accuracy of identifying a patient's epileptic state.

[0037] The technical solutions of the embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0038] Figure 1 It is a flowchart of the method for processing EEG signals shown in an embodiment of the present application.

[0039] See also Figure 1 , a method for processing an EEG signal, comprising:

[0040] S110: Collecting left and right brain EEG signals of the target subject.

[0041] Specifically, electrodes can be set at the head position of the target object, the left electrode collects the EEG signal of the left brain, and the right electrode collects the EEG signal of the right brain, so as to obtain the brain activity status of the left brain and the right brain respectively.

[0042] S120: performing asymmetric spatial feature extraction on the left brain EEG signal and the right brain EEG signal respectively to obtain left brain EEG signal features and right brain EEG signal features.

[0043] Specifically, asymmetric spatial feature extraction refers to independent feature calculation for the left and right brain regions to avoid feature confusion caused by mixing of bilateral signals. Feature extraction can be performed through a channel-by-channel time-frequency analysis method. By independently processing the left and right brain signals, the time-frequency characteristics of each region are retained. After feature extraction of the left brain EEG signals and the right brain EEG signals, the left brain EEG signal features and the right brain EEG signal features are obtained, such as feature vectors containing amplitude and frequency components.

[0044] S130: Based on a preset nearest neighbor algorithm, the left brain EEG signal features and the right brain EEG signal features are compared with EEG signal features in a preset signal feature library to determine whether the left brain EEG signal and the right brain EEG signal are epileptic EEG signals.

[0045] Specifically, the nearest neighbor algorithm is a classification method based on sample similarity. Specifically, it can be a method that dynamically adjusts the number of nearest neighbors. Classification decisions are made by calculating the similarity distance between the left and right brain EEG signal features and the EEG signals in the signal feature library. The left and right brain feature vectors are respectively input into the trained classification model and compared with the similarity of the labeled epilepsy samples in the signal feature library. When the matching degree between the EEG features of either side and the epilepsy sample exceeds the set standard, epilepsy identification is triggered. For example, when the similarity between the left and right brain EEG signal features and multiple EEG signals in the signal feature library is similar, it can be determined that the target subject is in an epileptic state. By lateralizing the brain, abnormal activity in unilateral brain regions can be identified, providing more accurate lateralization information for clinical diagnosis. The nearest neighbor method based on dynamic adjustment of feature distribution is more adaptable to the differences in individual EEG features and improves the accuracy of identifying patients' epileptic states.

[0046] The present application provides an EEG signal processing method, comprising: collecting the left-brain EEG signals and the right-brain EEG signals of a target object; performing asymmetric spatial feature extraction on the left-brain EEG signals and the right-brain EEG signals respectively to obtain left-brain EEG signal features and right-brain EEG signal features; and comparing the left-brain EEG signal features and the right-brain EEG signal features with EEG signal features in a preset signal feature library based on a preset nearest neighbor algorithm to determine whether the left-brain EEG signals and the right-brain EEG signals are epileptic EEG signals, thereby improving the accuracy of epileptic EEG signal recognition, enhancing the collaborative analysis capability of signals from the left and right brain regions, and improving the accuracy of recognition of the patient's epileptic state.

[0047] In one possible implementation, asymmetric spatial feature extraction is performed on the left brain EEG signal and the right brain EEG signal, respectively, to obtain left brain EEG signal features and right brain EEG signal features, including: performing time-frequency analysis on the left brain EEG signal to obtain the time domain features and frequency domain features of the left brain EEG signal; performing time-frequency analysis on the right brain EEG signal to obtain the time domain features and frequency domain features of the right brain EEG signal.

[0048] Specifically, time-frequency analysis refers to the joint analysis of EEG signals in the time and frequency dimensions. Time-frequency analysis of EEG signals can be performed through short-time Fourier transform, and the changing patterns of EEG signals in the time domain and frequency domain can be obtained. After time-frequency analysis, the time domain characteristics and frequency domain characteristics of the left brain EEG signals and the time domain characteristics and frequency domain characteristics of the right brain EEG signals can be obtained. The time domain characteristics may include signal amplitude, variance or zero-crossing rate, etc., and the frequency domain characteristics may generally include rate spectral density, frequency band energy ratio or peak frequency, etc.

[0049] Specifically, after collecting the EEG signals of the left and right brain, the two EEG signals are subjected to time-frequency analysis respectively. For example, for the left brain EEG signal, it is decomposed into multiple sub-bands through wavelet transform, and the energy of each sub-band is calculated as the frequency domain feature. At the same time, the root mean square value of the signal is extracted as the time domain feature. The right brain EEG signal is processed in the same way, and finally a multi-dimensional representation of the left and right brain containing time domain and frequency domain features is obtained. These features can simultaneously reflect the transient waveform abnormalities and energy mutations of the EEG signal in the specific frequency band during epileptic seizures, providing more discriminative input data for subsequent classification, and can more comprehensively capture the problems of sudden high-frequency oscillations and background rhythm disorders in epileptic EEG signals.

[0050] In one possible embodiment, based on a preset nearest neighbor algorithm, the left brain EEG signal features and the right brain EEG signal features are compared with the EEG signal features in a preset signal feature library to determine whether the left brain EEG signals and the right brain EEG signals are epileptic EEG signals, including: determining the K value of the nearest neighbor algorithm based on the distribution density of epileptic EEG signals in the signal feature library; if the proportion of epileptic signals in the K nearest neighbors of the left brain EEG signal features or the right brain EEG signal features exceeds a preset threshold, it is determined to be an epileptic EEG signal.

[0051] Specifically, the distribution density refers to the degree of aggregation of epileptic EEG signals in the signal feature library in the feature space, which can be obtained by calculating the number of samples in a unit volume. It can reflect the sparsity of epileptic EEG signals in different regions. By obtaining the distribution density of epileptic EEG signals in the signal feature library, the K value of the nearest neighbor algorithm can be determined. The K value refers to the number of neighboring samples participating in the voting in the nearest neighbor algorithm. It can be dynamically adjusted according to the local density of the feature space. For example, a smaller K value is selected in a high-density area to avoid noise interference, and a larger K value is selected in a low-density area to enhance classification stability.

[0052] Specifically, in areas where epilepsy samples are densely distributed in the signal feature library, such as the frequency domain feature concentration area corresponding to temporal lobe epilepsy, a smaller K value is used to reduce interference from non-correlated samples; in areas where epilepsy samples are sparse, such as the low-frequency feature area corresponding to frontal lobe epilepsy, a larger K value is used to expand the search range of neighboring samples. By calculating the proportion of epilepsy samples in the K nearest neighbors of the left brain or right brain features, if it exceeds the preset threshold, epilepsy judgment is triggered. For example, when the K value is set to 5 and the proportion of epilepsy samples reaches 4 / 5, the current EEG signal is determined to be an epilepsy signal.

[0053] Furthermore, the samples in the signal feature library can be divided into left brain epilepsy feature samples and right brain epilepsy feature samples. Density modeling is performed on these two sets of feature data to generate two independent density maps: one for the left brain feature space and the other for the right brain feature space. After the left brain EEG signals and right brain EEG signals of the target object are collected in real time and the left brain features and right brain features are extracted: for the left brain feature, its position can be located in the left brain feature density map, and a K is determined for it based on the density distribution of the map. L For right brain features, its position can be located in the right brain feature density map, and a K value can be determined for it based on the density distribution of the map. R For example, if a patient’s epilepsy originates from the left brain, the characteristics of the left brain epilepsy may be very typical and fall into the high-density area, thus obtaining a smaller K L Value, such as K L =5 for fine judgment; while the right brain may only show a less typical synchronous discharge after propagation, and the left brain epilepsy characteristics are less typical and may fall into the low-density area, thus obtaining a larger K R Value, such as K R =15 for judgment, which enables the classification process to accurately match the characteristic patterns of epilepsy signals in the left and right brains. This is better than using the same fixed K value to process left and right brain signals, greatly improving the sensitivity and specificity of focal epilepsy detection.

[0054] Secondly, the K value can be dynamically adjusted according to the patient's status. Before monitoring the patient, the patient's historical medical data (such as the confirmed epilepsy type, seizure frequency, commonly used drugs, etc.) can be obtained, and then the K value can be optimized using this information. For example, if the patient is known to have "left temporal lobe epilepsy", a density map and K value mapping model specially trained with left temporal lobe epilepsy patient data can be preferentially called or weighted, so that the selection of the K value can be related to the characteristic distribution area of ​​the left temporal lobe epilepsy sample; the K value can also be adjusted according to the patient's physiological state, such as awake state and sleep state. For example, many epilepsy types are active during sleep. Abnormal discharges (especially during NREM sleep) will increase significantly or become more typical. The signal feature library can be divided into the wakefulness state sub-library and the sleep state sub-library. When the patient enters the sleep state by analyzing the patient's EEG background rhythm, the K value is switched to the sleep state sub-library and its corresponding density map to determine the K value. Since epilepsy features may be denser during sleep, the K value selected at this time may be smaller overall, for example, the K value is not greater than 8. By dynamically adjusting the K value, the classification process can adapt to the sample distribution characteristics of different regions, effectively distinguishing epilepsy signals from similar features generated by noise or normal physiological activities, thereby improving the robustness of the judgment.

[0055] In one possible embodiment, the K value of the nearest neighbor algorithm is determined based on the distribution density of epileptic EEG signals in the signal feature library, including: when the distribution density indicates a high-density area, determining that the K value of the nearest neighbor algorithm is less than a first threshold; when the distribution density indicates a low-density area, determining that the K value of the nearest neighbor algorithm is greater than a second threshold.

[0056] Specifically, a high-density area refers to an area where epileptic EEG signal samples are concentrated in the feature space, and a low-density area refers to an area where epileptic EEG signal samples are dispersed in the feature space. The high-density area and the low-density area can be divided by setting a threshold for the proportion of sample numbers. For example, when the proportion of epileptic samples in a certain area exceeds 60%, it is determined to be a high-density area, and when the proportion of epileptic samples in a certain area is less than 20%, it is determined to be a low-density area.

[0057] Specifically, the first threshold and the second threshold can be set in advance, for example, the first threshold is set to 5, and the second threshold is set to 15. In the process of epileptic EEG signal recognition, the feature space is first divided into high-density areas and low-density areas by calculating the distribution density of epileptic samples in the signal feature library in the feature space. When the EEG signal feature to be detected falls into the high-density area, a smaller K value is used for nearest neighbor classification, for example, K=3. At this time, it can avoid that too much noise data contained in neighboring samples interferes with the classification results; when the feature to be detected falls into the low-density area, a larger K value is used for nearest neighbor classification, for example, K=20. At this time, it can avoid misjudgment caused by sample sparsity by expanding the range of neighboring samples. This dynamic adjustment mechanism reduces the influence of the K value parameter on the classification performance by matching the sample distribution characteristics of different density areas, thereby effectively improving the accuracy and robustness of epileptic signal recognition.

[0058] In a possible embodiment, the preset nearest neighbor algorithm uses Mahalanobis distance as a distance metric, and also includes: calculating the Mahalanobis distance between the left brain EEG signal features, the right brain EEG signal features and the epileptic EEG signals and non-epileptic EEG signals in the signal feature library based on the covariance matrix of the epileptic EEG signals and the non-epileptic EEG signals in the signal feature library; based on the minimum Mahalanobis distance, voting on the left brain EEG signal features and the right brain EEG signal features to determine whether the left brain EEG signal features and the right brain EEG signal features are epileptic EEG signals.

[0059] Specifically, the Mahalanobis distance can be used to indicate the correlation between features, and can be obtained by calculating the weighted Euclidean distance between the sample point and the category mean vector. The weight is determined by the inverse matrix of the covariance matrix. The covariance matrix is ​​used to characterize the correlation between the various feature dimensions of epileptic EEG signals and non-epileptic EEG signals. It is obtained by statistically analyzing the feature distribution of samples in the signal feature library and calculating the covariance value. Based on the minimum Mahalanobis distance, the left brain EEG signal features and the right brain EEG signal features can be voted. The voting mechanism is to perform majority voting based on the category to which the minimum Mahalanobis distance between the left brain or right brain EEG signal features and the samples in the signal feature library belongs.

[0060] Specifically, the covariance matrix of epileptic EEG signals and non-epileptic EEG signals is first extracted from the signal feature library, and the Mahalanobis distances between the left and right brain EEG signal features and the two types of signals are calculated respectively. For example, for the left brain EEG signal feature, its Mahalanobis distance to the epilepsy category can be calculated by the formula, where the covariance matrix reflects the distribution relationship between epilepsy signal features. Subsequently, the category to which the minimum Mahalanobis distance corresponding to the left and right brain features belongs is selected as the candidate result. If the number of votes for the epilepsy category corresponding to the left or right brain feature exceeds the threshold, it is determined to be an epileptic EEG signal. By introducing the Mahalanobis distance and combining the covariance matrix to dynamically adjust the distance weight, it can more accurately reflect the real distribution differences in the feature space, reduce misjudgments caused by feature redundancy or noise interference, and thus improve the recognition accuracy of epileptic EEG signals.

[0061] In a possible implementation, the method further includes performing baseline correction and power frequency interference filtering on the left brain EEG signals and the right brain EEG signals.

[0062] Specifically, baseline correction eliminates low-frequency baseline drift caused by poor electrode contact or physiological activity during signal acquisition. This can be performed on both left and right brain EEG signals using a high-pass filter. Power frequency interference filtering suppresses fixed-frequency noise interference from the power supply line, and can be handled using a digital notch filter.

[0063] Specifically, after EEG signal acquisition is completed, baseline correction is first performed on the original signals of the left and right brain. For example, a sliding average algorithm with a time window length of 1 second is used to calculate the local mean of each sampling point and subtract it from the original signal to eliminate low-frequency drift caused by breathing or body movement. Then, power frequency interference filtering is applied to the corrected signal. For example, a zero-phase digital notch filter is used with its stopband center frequency set to 50Hz and bandwidth set to 2Hz. A second-order IIR filter is designed through the bilinear transformation method to eliminate periodic interference introduced by the power line. The non-epilepsy-related noise components in the EEG signal are effectively suppressed, providing high-quality input for subsequent feature extraction and classification, reducing the risk of misclassification due to baseline drift or power frequency noise, and improving the recognition reliability of epileptic EEG signals.

[0064] In a possible implementation, the method further includes: generating an audible and visual alarm and alarm information when a left brain EEG signal or a right brain EEG signal is determined to be an epileptic signal; associating the alarm information with a preset data table of the target object to generate a diagnostic recommendation report.

[0065] Specifically, when the EEG signal of the left or right brain is determined to be an epileptic signal, the buzzer immediately emits a warning sound of a set frequency, and the LED light can also flash in a specific color. The alarm information is recorded in real time as a data packet containing a timestamp, signal strength and brain area identification. The target object's age, previous epileptic seizure records and current medication information in the preset data table are retrieved and matched with the brain area positioning data in the alarm information. The matched data is entered into the diagnostic recommendation report, and a diagnostic recommendation report is generated including medication adjustment suggestions, review time and inspection item list.

[0066] The EEG signal processing method of the present application combines asymmetric features with the nearest neighbor algorithm, and realizes accurate identification of epilepsy signals by constructing a dedicated signal feature library, thereby solving the problems of high misjudgment rate and large response delay in traditional methods in epilepsy detection.

[0067] Corresponding to the aforementioned application function implementation method embodiment, the present application also provides an EEG signal processing device, an electronic device and corresponding embodiments.

[0068] Figure 2 Schematic diagram of the structure of the EEG signal processing device shown in an embodiment of the present application.

[0069] See also Figure 2 , an electroencephalogram signal processing device 200, comprising:

[0070] The acquisition module 210 is used to acquire the left brain EEG signals and the right brain EEG signals of the target object.

[0071] The processing module 220 is used to perform asymmetric spatial feature extraction on the left brain EEG signal and the right brain EEG signal respectively to obtain left brain EEG signal features and right brain EEG signal features.

[0072] The determination module 230 is used to compare the left brain EEG signal features and the right brain EEG signal features with the EEG signal features in a preset signal feature library based on a preset nearest neighbor algorithm to determine whether the left brain EEG signal and the right brain EEG signal are epileptic EEG signals.

[0073] The present application provides an EEG signal processing device, comprising: an acquisition module for acquiring left-brain EEG signals and right-brain EEG signals of a target object; a processing module for performing asymmetric spatial feature extraction on the left-brain EEG signals and right-brain EEG signals, respectively, to obtain left-brain EEG signal features and right-brain EEG signal features; a determination module for comparing the left-brain EEG signal features and right-brain EEG signal features with EEG signal features in a preset signal feature library based on a preset nearest neighbor algorithm, to determine whether the left-brain EEG signals and right-brain EEG signals are epileptic EEG signals, thereby improving the accuracy of epileptic EEG signal recognition, enhancing the collaborative analysis capability of left and right brain region signals, and improving the accuracy of recognition of the patient's epileptic state.

[0074] In one possible implementation, the processing module 220 is further configured to perform time-frequency analysis on the left brain EEG signal to obtain the time domain characteristics and frequency domain characteristics of the left brain EEG signal; and to perform time-frequency analysis on the right brain EEG signal to obtain the time domain characteristics and frequency domain characteristics of the right brain EEG signal.

[0075] In a possible embodiment, the determination module 230 is also used to determine the K value of the nearest neighbor algorithm based on the distribution density of epileptic EEG signals in the signal feature library; if the proportion of epileptic signals in the K nearest neighbors of the left brain EEG signal feature or the right brain EEG signal feature exceeds a preset threshold, it is determined to be an epileptic EEG signal.

[0076] In one possible embodiment, the determination module 230 is further used to determine that the K value of the nearest neighbor algorithm is less than a first threshold when the distribution density indicates a high-density area; and to determine that the K value of the nearest neighbor algorithm is greater than a second threshold when the distribution density indicates a low-density area.

[0077] In a possible embodiment, the determination module 230 is further used to calculate the Mahalanobis distance between the left brain EEG signal features, the right brain EEG signal features and the epileptic EEG signals and non-epileptic EEG signals in the signal feature library based on the covariance matrix of the epileptic EEG signals and the non-epileptic EEG signals in the signal feature library; based on the minimum Mahalanobis distance, the left brain EEG signal features and the right brain EEG signal features are voted to determine whether the left brain EEG signal features and the right brain EEG signal features are epileptic EEG signals.

[0078] In a possible implementation, the processing module 220 is further configured to perform baseline correction and power frequency interference filtering on the left brain EEG signals and the right brain EEG signals.

[0079] In one possible embodiment, the processing module 220 is also used to generate an audible and visual alarm and alarm information when the left brain EEG signal or the right brain EEG signal is determined to be an epileptic signal; associate the alarm information with a preset data table of the target object to generate a diagnostic recommendation report.

[0080] Regarding the apparatus in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated again here.

[0081] An embodiment of the present application also provides an electronic device. Figure 3 This is a hardware structure diagram of an embodiment of an electronic device of the present application. The electronic device includes a memory 320 and at least one processor 310. The memory 320 is electrically connected to the at least one processor 310. The memory 320 stores instructions. The at least one processor 310 calls the instructions in the memory 320, causing the electronic device to execute the EEG signal processing method according to any of the aforementioned embodiments of the present application.

[0082] Specifically, the processor 310 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0083] The memory 320 may include a large capacity memory 320 for data or instructions. By way of example and not limitation, the memory 320 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive, or a combination of two or more of these. Where appropriate, the memory 320 may include a removable or non-removable (or fixed) medium. Where appropriate, the memory 320 may be inside or outside the integrated gateway disaster recovery device. In a specific embodiment, the memory 320 is a non-volatile solid-state memory. In a specific embodiment, the memory 320 includes a read-only memory (ROM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically rewritable ROM (EAROM), or a flash memory, or a combination of two or more of these.

[0084] In one example, the control device may further include a communication interface 330 and a bus 340. The processor 310, the memory 320, and the communication interface 330 are connected via the bus 340 and communicate with each other.

[0085] The communication interface 330 is mainly used to implement communication between various modules, devices, units and / or equipment in the embodiments of the present application.

[0086] Bus 340 includes hardware, software or both, and the parts of online data flow billing equipment are coupled to each other. For example, but not limitation, bus can include accelerated graphics port (AGP) or other graphics bus, enhanced industry standard architecture (EISA) bus, front side bus (FSB), hypertransport (HT) interconnection, industry standard architecture (ISA) bus, infinite bandwidth interconnection, low pin count (LPC) bus, memory bus, micro channel architecture (MCA) bus, peripheral component interconnection (PCI) bus, PCI-Express (PCI-X) bus, serial advanced technology attachment (SATA) bus, video electronics standard association local (VLB) bus or other suitable bus or two or more of these combinations. In appropriate cases, bus 340 can include one or more buses. Although the present application embodiment describes and shows specific bus, the application considers any suitable bus or interconnection.

[0087] In addition, in conjunction with the EEG signal processing method in the above embodiments, the present application embodiment may provide a computer-readable storage medium for implementation. The computer-readable storage medium stores instructions that, when executed by a processor, implement any of the EEG signal processing methods in the above embodiments.

[0088] The present application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, a detailed description of known methods is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications and additions, or change the order of the steps after understanding the spirit of the present application.

[0089] The functional blocks shown in the above block diagram can be implemented as hardware, software, firmware or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of the present application are programs or code segments that are used to perform the required tasks. Programs or code segments can be stored in machine-readable media, or transmitted on a transmission medium or a communication link by a data signal carried in a carrier wave. "Machine-readable media" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROMs, flash memories, erasable ROMs (EROMs), floppy disks, CD-ROMs, optical disks, hard disks, optical fiber media, radio frequency (RF) links, etc. The code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0090] It should also be noted that the exemplary embodiments mentioned in this application describe some methods or systems based on a series of steps or devices. However, this application is not limited to the order of the above steps. In other words, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0091] Alternatively, the present application also provides a computer program product that can implement part or all of the steps of the method in the above embodiments. The computer program product includes a computer program / instructions, which implement part or all of the steps of the method in the above embodiments when executed by a processor.

[0092] The above description is only a specific embodiment of the present application. Those skilled in the art will clearly understand that for the convenience and brevity of description, the specific working processes of the systems, modules and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. It should be understood that the scope of protection of the present application is not limited thereto. Any person skilled in the art can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present application, and these modifications or replacements should be included in the scope of protection of the present application.

Claims

1. A method for processing an electroencephalogram signal, characterized in that: include: Collecting left and right brain EEG signals of the target subject; Performing asymmetric spatial feature extraction on the left brain EEG signal and the right brain EEG signal respectively to obtain left brain EEG signal features and right brain EEG signal features; Based on a preset nearest neighbor algorithm, the left brain EEG signal features and the right brain EEG signal features are compared with EEG signal features in a preset signal feature library to determine whether the left brain EEG signal and the right brain EEG signal are epileptic EEG signals.

2. The method according to claim 1, characterized in that The performing asymmetric spatial feature extraction on the left brain EEG signal and the right brain EEG signal respectively to obtain left brain EEG signal features and right brain EEG signal features includes: Performing time-frequency analysis on the left brain electroencephalogram signal to obtain time domain characteristics and frequency domain characteristics of the left brain electroencephalogram signal; Performing time-frequency analysis on the right brain EEG signal to obtain time domain features and frequency domain features of the right brain EEG signal.

3. The method according to claim 1, characterized in that The method of comparing the left brain EEG signal features and the right brain EEG signal features with EEG signal features in a preset signal feature library based on a preset nearest neighbor algorithm to determine whether the left brain EEG signal and the right brain EEG signal are epileptic EEG signals includes: Determining a K value of the nearest neighbor algorithm based on a distribution density of epileptic EEG signals in the signal feature library; If the proportion of epilepsy signals in the K nearest neighbors of the left brain EEG signal feature or the right brain EEG signal feature exceeds a preset threshold, it is determined to be an epileptic EEG signal.

4. The method according to claim 3, characterized in that The determining of the K value of the nearest neighbor algorithm based on the distribution density of the epileptic EEG signal in the signal feature library includes: When the distribution density indication is a high-density area, determining that the K value of the nearest neighbor algorithm is less than a first threshold; When the distribution density indicates a low-density area, it is determined that the K value of the nearest neighbor algorithm is greater than a second threshold.

5. The method according to claim 3, characterized in that The preset nearest neighbor algorithm uses Mahalanobis distance as a distance metric, and also includes: Calculating the Mahalanobis distances between the left brain EEG signal features, the right brain EEG signal features, and the epileptic EEG signals and non-epileptic EEG signals in the signal feature library based on the covariance matrix of the epileptic EEG signals and non-epileptic EEG signals in the signal feature library; Based on the minimum Mahalanobis distance, the left brain EEG signal feature and the right brain EEG signal feature are voted to determine whether the left brain EEG signal feature and the right brain EEG signal feature are epileptic EEG signals.

6. The method according to claim 1, characterized in that Also includes: Baseline correction and power frequency interference filtering are performed on the left brain electroencephalogram signal and the right brain electroencephalogram signal.

7. The method according to claim 1, characterized in that Also includes: When the left brain electroencephalogram signal or the right brain electroencephalogram signal is determined to be an epileptic signal, generating an audible and visual alarm and an alarm message; The alarm information is associated with a preset data table of the target object to generate a diagnosis suggestion report.

8. A device for processing electroencephalogram signals, characterized in that: include: An acquisition module, used to acquire the left brain EEG signals and the right brain EEG signals of the target object; a processing module, configured to perform asymmetric spatial feature extraction on the left brain EEG signal and the right brain EEG signal, respectively, to obtain left brain EEG signal features and right brain EEG signal features; The determination module is used to compare the left brain EEG signal features and the right brain EEG signal features with the EEG signal features in a preset signal feature library based on a preset nearest neighbor algorithm to determine whether the left brain EEG signal and the right brain EEG signal are epileptic EEG signals.

9. An electronic device, characterized in that: include: processor; as well as A memory having executable codes stored thereon, which, when executed by the processor, causes the processor to execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: An executable code is stored thereon, and when the executable code is executed by a processor of an electronic device, the processor is caused to execute the method according to any one of claims 1 to 7.

Images