Visualization methods, control devices, storage media, and visualization systems for electroencephalography (EEG)

Abstract

This invention provides a method, control device, storage medium, and visualization system for visualizing electroencephalograms (EEGs), belonging to the field of EEG technology. The EEG visualization method includes: acquiring EEG signals; extracting feature indicators from the EEG signals; and generating an EEG feature indicator report based on the extracted feature indicators. Through EEG signal data processing technology, time-domain, frequency-domain, and spatial-domain features of the EEG signals are extracted to obtain feature indicators. Multiple algorithms can be selected for each feature indicator. Different users (e.g., clinical electrophysiologists) can choose different EEG features and corresponding algorithms (or default algorithms) to extract feature indicators, and then visualize the extracted feature indicators to form a customized EEG feature indicator report, helping clinical electrophysiologists obtain more valuable diagnostic information from EEG signals.

Description

Technical Field

[0001] This invention relates to the field of electroencephalography (EEG) technology, and more specifically to an EEG visualization method, control device, storage medium, and visualization system. Background Technology

[0002] Electroencephalography (EEG) is a sensitive indicator for evaluating brain function and is widely used in the diagnosis and research of central nervous system diseases and mental illnesses, as well as in psychology and cognitive science research. Especially for the qualitative and localization of paroxysmal brain dysfunction such as epilepsy, EEG remains an irreplaceable diagnostic technique.

[0003] However, the origin and activity patterns of brain waves still remain largely unknown, and a wealth of information contained within them cannot yet be effectively extracted, analyzed, and interpreted. The scientific and rational application of new technologies to improve EEG analysis and techniques is of great value and assistance in enhancing the diagnostic capabilities of clinical neurophysiologists. Summary of the Invention

[0004] The purpose of this invention is to provide a method for visualizing electroencephalograms (EEGs), which can help improve the work and diagnosis of clinical neurophysiologists.

[0005] To achieve the above objectives, embodiments of the present invention provide a method for visualizing electroencephalograms (EEGs). The method includes: acquiring EEG signals; extracting feature indicators from the EEG signals; and generating an EEG feature indicator report based on the extracted feature indicators.

[0006] Optionally, the characteristic indicators include one or more of the following: amplitude modulation, power spectral density, brain region subband energy distribution, alpha power asymmetry, and brain network characteristics.

[0007] Optionally, the EEG signal can be acquired at a preset sampling rate and a preset number of channels.

[0008] Optionally, after acquiring the EEG signal, the EEG visualization method further includes preprocessing the EEG signal, including: removing the EEG signal acquired through useless channels; performing rereference processing on the EEG signal after removing useless channels; filtering the rereference processed EEG signal at a preset frequency to remove base drift; and removing power frequency noise interference from the EEG signal after removing base drift.

[0009] Optionally, before extracting the feature indicators of the EEG signal, the EEG visualization method further includes: obtaining a bandpass signal from the EEG signal through a bandpass filter, the bandpass signal including the alpha rhythm; and selecting the bandpass signal of the occipital leads.

[0010] Optionally, extracting the modulation and amplitude of the EEG signal includes: performing Alpha rhythm detection on the bandpass signal in the occipital leads to determine the modulation and amplitude of the Alpha rhythm.

[0011] Optionally, determining the regulation and amplitude modulation of the alpha rhythm includes: statistically analyzing whether the frequency change rate of the alpha rhythm is greater than a first threshold th1. α The proportion of th1 α The corresponding Alpha rhythm is determined to be dysregulation; for each Alpha rhythm's spindle-shaped waveform, the peak amplitude in the middle of the spindle-shaped waveform is determined. amp The amplitude difference Δ at both ends amp Peak amplitude amp Ratio amp To indicate the amplitude modulation situation; when the Ratio amp Less than the second threshold th2 α When the corresponding Alpha rhythm is determined to be poorly modulated, the proportion of poorly modulated Alpha rhythms (th2) is statistically analyzed. α ; and statistically analyze the center frequency of the alpha rhythm to obtain the center frequency distribution, within a preset percentage range of center frequencies. α , which serves as the main distribution frequency band range of the Alpha rhythm.

[0012] Optionally, extracting the power spectral density includes: calculating the power spectral density of the bandpass signal for each lead; and determining the distribution frequency band of the periodic component based on the curve corresponding to the calculated power spectral density.

[0013] Optionally, extracting the sub-band energy distribution of the brain region includes: determining a set of sub-bands based on the bandpass signal of the occipital region lead, wherein the sub-bands include delta band, theta band, alpha band, beta band, gamma band, and full band; calculating the signal energy of each sub-band in the set of sub-bands; projecting the signal energy of each sub-band onto a topographic map; and merging the signal energy of each sub-band through brain region classification, wherein the brain region classification includes the prefrontal lobe, frontal lobe, central region, temporal lobe, parietal lobe, and occipital lobe.

[0014] Optionally, extracting the alpha power asymmetry includes: calculating the power spectral density of the bandpass signal in the alpha band; and for each brain region classification, calculating the left-right lateral asymmetry index SI for each brain region classification using the following formula:

[0015]

[0016] Among them, PSD L Power spectral density (PSD) represents the power spectral density of the left brain region. R This represents the power spectral density of the right brain region.

[0017] Optionally, extracting the brain network characteristics includes: extracting the envelope based on the bandpass signal in the Alpha band; calculating the Pearson correlation value of the envelopes of any two leads to obtain a correlation matrix; and converting the correlation matrix into a small-world network.

[0018] Optionally, the EEG feature report may include one or more of the following: graphs and texts about background EEG activity; graphs and texts about the main frequency bands of periodic oscillation components; graphs and texts about alpha power asymmetry; and graphs and texts about brain network characteristics.

[0019] This invention also provides a control device, characterized in that the control device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described electroencephalogram visualization method.

[0020] This invention also provides a machine-readable storage medium, characterized in that the machine-readable storage medium stores instructions that cause a machine to execute the above-described electroencephalogram visualization method.

[0021] This invention also provides an electroencephalogram (EEG) visualization system, which includes the control device and display module described above. The display module is used to display the EEG feature index report generated by the control device.

[0022] Through the above technical solution, this invention utilizes data processing technology to extract time-domain, frequency-domain, and spatial-domain features of electroencephalogram (EEG) signals, obtaining characteristic indicators of the EEG signals. Multiple algorithms are available for each characteristic indicator. Different users (e.g., clinical electrophysiologists) can select different EEG features and corresponding algorithms (or default algorithms) to extract the characteristic indicators. The extracted characteristic indicators are then visualized and output to form a customized EEG characteristic indicator report, helping clinical electrophysiologists obtain more valuable diagnostic information from EEG signals. Furthermore, by quantifying EEG features, the text of routine EEG indicator reports is automatically generated, which can help clinical electrophysiologists improve the efficiency of EEG interpretation.

[0023] Other features and advantages of the embodiments of the present invention will be described in detail in the following detailed description section. Attached Figure Description

[0024] The accompanying drawings are provided to further illustrate embodiments of the present invention and form part of the specification. They are used together with the following detailed description to explain the embodiments of the present invention, but do not constitute a limitation thereof. In the drawings:

[0025] Figure 1 This is a flowchart illustrating the electroencephalogram visualization method provided in an embodiment of the present invention;

[0026] Figures 2a-2f This is a schematic diagram illustrating the example of amplitude modulation adjustment;

[0027] Figure 3 This is a schematic diagram of an example power spectral density;

[0028] Figure 4 This is a schematic diagram of the energy distribution of sub-bands in an example brain region;

[0029] Figure 5 This is a schematic diagram illustrating an example of Alpha power asymmetry;

[0030] Figures 6a-6b This is a schematic diagram illustrating the characteristics of an example brain network;

[0031] Figure 7 This is a schematic diagram of a routine EEG indicator report;

[0032] Figure 8 This is a schematic diagram of an example EEG feature index report. Detailed Implementation

[0033] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit the scope of the present invention.

[0034] Before explaining the embodiments of the present invention in detail, the names involved in the embodiments of the present invention will be defined and explained.

[0035] 1. Adjust amplitude

[0036] Alpha rhythm is a hallmark rhythm in electroencephalograms (EEGs). The International Federation for Electroencephalography and Clinical Neurophysiology (IFSECN, 1974) defines alpha rhythm as: "a rhythm of 8–13 Hz occurring in the posterior head during wakefulness, with the highest voltage generally in the occipital region. The amplitude is variable, often below 50 μV in adults. It is more likely to occur with eyes closed and in a relaxed state. Concentration, especially visual attention and active mental activity, can inhibit it." Alpha rhythm is the most important indicator for analyzing background EEG activity and is closely related to brain function and developmental level. However, alpha rhythm is not correlated with intelligence level, personality, or character; patients with intellectual disabilities may also have a good alpha rhythm.

[0037] Alpha rhythms are mostly rounded or sinusoidal. The frequency of alpha rhythms is closely related to age. Infants and young children have not yet developed alpha rhythms in the occipital region; the first alpha rhythm generally appears around age 3, at around 8 Hz. By age 10, the frequency of alpha rhythms approaches adult levels. The dominant frequency range of adult alpha rhythms is between 9 and 11 Hz, and the alpha rhythm is relatively constant for each individual. After age 60, alpha rhythms slow down; research reports show that the alpha rhythm of a generally healthy centenarian is predominantly 8 Hz. In the same adult individual during the same recording, the frequency variation of a sustained alpha rhythm within the corresponding regions of both hemispheres does not exceed 0.5–1 Hz; this is called frequency modulation and reflects the regularity of brain wave activity. The frequency variation range across the entire head should not exceed 2 Hz. Besides age, the frequency of alpha rhythms is particularly sensitive to alertness levels, cerebral blood flow perfusion levels, and the effects of certain drugs. When a subject experiences mild drowsiness with their eyes closed, the frequency of the alpha rhythm may slow down, and its distribution area may expand towards the forebrain.

[0038] Alpha rhythms generally have the highest amplitude in the occipital region. There may be a slight amplitude difference between the left and right occipital regions, with the right side usually showing a higher amplitude, which is related to the fact that the left side of the skull is thicker than the right. This physiological asymmetry in amplitude is generally between 20% and 30%, and if it exceeds 50%, it may be clinically significant.

[0039] Modulation: The amplitude of a normal alpha rhythm exhibits a spindle-shaped change, gradually increasing and decreasing, known as amplitude modulation, which reflects the stability of brain waves. Each spindle-shaped alpha rhythm lasts for 1-2 seconds, with a few lasting up to 20 seconds, and a small amount of low-amplitude beta activity between two sequences. The younger the child, the less stable their brain waves are, often lacking amplitude modulation. More stable amplitude modulation typically appears after 9-10 years of age. Poor amplitude modulation can manifest as continuous, unchanging rhythmic alpha waves, or as alpha waves with irregular amplitudes and no discernible pattern.

[0040] Regulation: refers to the frequency regulation of EEG, reflecting the regularity of brain electrical activity. In normal adults, the basic rhythmic brainwave frequency in the occipital region is quite stable. Within a short period (e.g., 1-3 seconds) of the same recording, the frequency difference in the same area should not exceed 1 Hz, and the frequency difference in corresponding areas of the two hemispheres should not exceed 0.5 Hz; otherwise, it indicates poor regulation. Children, whose brains are not yet fully developed, lack such stable frequency regulation.

[0041] 2. Power spectral density

[0042] Electrophysiological signals possess both periodic and aperiodic components. Periodic oscillations are associated with numerous physiological, cognitive, behavioral, and disease states. New evidence suggests that the aperiodic component has a hypothetical physiological explanation and varies dynamically with age, task demands, and cognitive state. Nearly a century of research has demonstrated that periodic oscillations reflect a wide range of cognitive, perceptual, and behavioral states. Recent studies indicate that periodic oscillations contribute to the coordination of information transfer between regions. Notably, almost all major neurological and psychiatric disorders are associated with dysfunction of periodic oscillations. Historically, the vast majority of research on periodic oscillations has relied on standard frequency bands, broadly defined as: delta band (1-4 Hz), theta band (4-8 Hz), alpha band (8-13 Hz), beta band (13-30 Hz), low gamma band (30-60 Hz), high-frequency activity band (60-250 Hz), and fast ripple band (200-400 Hz).

[0043] 3. Topographic maps of energy distribution in different frequency bands

[0044] 1) Delta band (1-4Hz): associated with deep sleep.

[0045] 2) Theta band (4-8Hz): It is associated with emotional processing and appears to be a good feature in diagnostic tools, but little information is available about its mechanism.

[0046] 3) Alpha band (8-13Hz): reflects the resting state and relaxation of the brain. The asymmetry of brain activity is related to approach-avoidance mode.

[0047] 4) Beta band (13-30Hz): associated with anticipation, anxiety, and introversion. Beta waves are more associated with anxiety and rumination, which are common in depressed patients.

[0048] 5) Gamma band (30-40Hz): Related to attention and sensory systems, and may be associated with mood fluctuations. A 2018 review of the gamma band by Fitzgerald et al. reported many studies showing increased gamma wave power in unipolar depressed patients compared to healthy controls, suggesting that gamma waves may be related to mood fluctuations, and that appropriate gamma wave power can ensure mood stability, and even link it to treatment prediction.

[0049] 4. Alpha(α) power asymmetry

[0050] In the resting state, alpha power is negatively correlated with brain activity, and in patients with depression, alpha power is significantly greater on the left side of the frontal lobe than on the right. The avoidance-approach system model proposed by Davidson R et al. posits that the left hemisphere belongs to the approach behavior system, associated with positive emotional processing, while the right hemisphere belongs to the avoidance behavior system, associated with negative emotional processing. Frontal asymmetry reflects the characteristic of impaired approach motivation in depression, an inference consistent with the symptoms of anhedonia and loss of interest in depression. This model helps explain some of the results regarding the alpha band in the left hemisphere of depressed individuals. Since the alpha (α) band is associated with a lack of brain activity, increased alpha band activity on the left side indicates reduced activity, potentially suggesting a lack of approach behavior. Some studies have shown that alpha band activity may occur more frequently on the left side of the brain in subjects with depression. Furthermore, anxiety disorders may alter alpha band lateralization, making depression difficult to diagnose. Alpha band lateralization appears to be an effective biomarker for specific symptoms, such as irritability and lethargy, and unipolar depression, which differs from bipolar depression, but it is less sensitive to other symptoms, particularly anxiety. Future research should explore its sensitivity to other external factors for diagnostic prediction.

[0051] 5. Brain Network

[0052] Multiple studies, including SPECT, have found that patients with depression exhibit more pronounced dysfunction in the right hemisphere or non-dominant hemisphere, leading to an imbalance in function between the two hemispheres. Quantitative EEG studies have also shown that right-hemispheric abnormalities are more prominent in patients with depression, while dominant hemisphere dysfunction is more pronounced in patients with mania. Some studies have found that the dominant hemisphere promotes positive emotions, while the non-dominant hemisphere promotes negative emotions.

[0053] Electroencephalogram (EEG) electrodes can acquire the activity of neuronal populations, and this information can be used to analyze the interactions of brain activity between different brain regions and study their network structure. This invention preferably utilizes two features: functional connectivity and small-world characteristics. Functional connectivity (FC) represents the synchronous activity between different brain regions. Small-world (SW) refers to the small-world structure naturally present in the human brain, which consists of highly interconnected small regions linked by several centers.

[0054] Many researchers have studied the characteristics of brain networks and linked them to depression. Orgo et al. found that in a depressive state, there are more random networks in the brain, as well as enhanced functional connectivity, meaning that many neurons (or regions) are activated synchronously. Current research seems to agree on the randomness of brain networks in patients with depression, but regarding functional connectivity, some studies have found stronger connections in healthy subjects, while others have found stronger connections in depressed subjects. Therefore, further research is needed to analyze brain networks related to depression.

[0055] Figure 1 This is a flowchart illustrating the electroencephalogram visualization method provided in this embodiment of the invention. Please refer to it. Figure 1 The method for visualizing the electroencephalogram (EEG) may include the following steps:

[0056] Step S110: Acquire electroencephalogram (EEG) signals.

[0057] Preferably, the electroencephalogram (EEG) signals are acquired at a preset sampling rate and a preset number of channels.

[0058] For example, EEG signals can be acquired using a clinically standard EEG amplifier, with a sampling rate ranging from 200 to 1000 Hz and a channel count ranging from 16 to 32. In outpatient settings, EEG signal acquisition can last for more than 5 minutes, under resting conditions.

[0059] Preferably, after step S110, the EEG visualization method may further include preprocessing the EEG signal, including: 1) removing the EEG signal acquired from useless channels; 2) performing rereference processing on the EEG signal after removing useless channels; 3) filtering the rereference processed EEG signal at a preset frequency to remove base drift; 4) removing power frequency noise interference from the EEG signal after removing base drift.

[0060] For example, 1) Remove useless channels. When acquiring EEG signals, some channels that are not needed later may be acquired, such as electromyography signals, bilateral mastoid points, etc. These can be removed from the acquired EEG signals and do not need to be included in subsequent analysis.

[0061] 2) Rereference. Rereference can include methods such as monopolar rereference, bipolar rereference, Laplacian rereference, or common average rereference. This embodiment of the invention preferably uses common average rereference. That is, the mean of all data (EEG signals) from the whole brain is used as the reference data, and each value of the EEG signal is subtracted from the reference data to obtain the rereferenced EEG signal.

[0062] 3) Base drift removal. To eliminate the effects of data drift, the EEG signal after rereference is high-pass filtered at a preset frequency (e.g., 0.1Hz) to obtain the EEG signal after base drift removal.

[0063] 4) Remove power frequency, for example, remove the 50Hz power frequency. In this embodiment of the invention, a 50Hz notch filter is preferably used to eliminate power frequency noise interference, resulting in an EEG signal after power frequency noise removal.

[0064] Step S120: Extract the feature indicators of the electroencephalogram (EEG) signal.

[0065] The preferred feature indicators in the embodiments of the present invention include one or more of the following: amplitude modulation adjustment, power spectral density, brain region subband energy distribution, alpha power asymmetry, and brain network characteristics.

[0066] For the definitions and explanations of the aforementioned characteristic indicators, please refer to the preceding text.

[0067] Preferably, prior to step S120, the method for visualizing the electroencephalogram (EEG) further includes: obtaining a bandpass signal from the EEG signal using a bandpass filter, the bandpass signal including the alpha rhythm; and selecting the bandpass signal from the occipital leads.

[0068] For example, the preprocessed EEG signal is filtered through an 8-13Hz bandpass filter to obtain a bandpass signal that includes the alpha rhythm (also known as the α rhythm). The bandpass signal of the occipital region lead (e.g., 01 or 02) is selected because the left and right hemispheres of the EEG signal can remain symmetrical, so one of the two leads can be selected.

[0069] Preferably, extracting the modulation and amplitude modulation of the EEG signal includes: performing Alpha rhythm detection on the bandpass signal in the occipital leads to determine the modulation and amplitude modulation of the Alpha rhythm.

[0070] For example, alpha rhythm detection can be performed on the bandpass signal in the epigastric leads, for instance, using a supplementary detection algorithm. Figure 2a As shown, each point represents an Alpha rhythm. Figure 2a In the diagram, the horizontal axis represents time, and the vertical axis represents frequency. Figure 2b In the diagram, the horizontal axis represents time, and the rate of change of frequency.

[0071] More preferably, determining the regulation and amplitude modulation of the alpha rhythm includes: statistically analyzing whether the frequency change rate of the alpha rhythm is greater than a first threshold th1. α The proportion of th1 α The corresponding Alpha rhythm is determined to be dysregulation; for each Alpha rhythm's spindle-shaped waveform, the peak amplitude in the middle of the spindle-shaped waveform is determined. amp The amplitude difference Δ at both ends amp Peak amplitude amp Ratio amp To indicate the amplitude modulation situation; when the Ratio amp Less than the second threshold th2 α When the corresponding Alpha rhythm is determined to be poorly modulated, the proportion of poorly modulated Alpha rhythms (th2) is statistically analyzed. α ; and statistically analyze the center frequency of the alpha rhythm to obtain the center frequency distribution, within a preset percentage range of center frequencies. α , which serves as the main distribution frequency band range of the Alpha rhythm.

[0072] Following the above example, this embodiment of the invention uses the rate of change of the alpha rhythm to represent the adjustment status. For example... Figure 2b As shown, each point represents Figure 2a The corresponding rate of change of the alpha rhythm frequency. To distinguish malregulated alpha rhythms, this embodiment of the invention uses 0.33 Hz / s as the threshold th1 for distinguishing malregulated alpha rhythms. α Statistically greater than th1 α R1 α Regarding the amplitude modulation of the alpha rhythm, this embodiment of the invention uses the peak amplitude (midpoint) of the spindle-shaped waveform of each alpha rhythm. amp The amplitude difference Δ between the two ends amp Ratio of the highest peak amplitudeamp To represent amplitude modulation (as shown in the following expression), such as Figure 2c As shown, each point represents Figure 2a The amplitude of the corresponding Alpha rhythm.

[0073]

[0074] When the alpha rhythm exhibits a sustained, unchanging amplitude, it is called dysregulation, which can be addressed by adjusting the ratio. amp To reflect, i.e., Ratio amp A very small value indicates poor regulation.

[0075] To differentiate poorly modulated alpha rhythms, set the ratio. amp The threshold is th2 α When Ratio amp Below th2 α When, it indicates a poor amplitude modulation ratio. amp Rhythm, the proportion of statistically poor amplitude modulation R2 α Determine the frequency band distribution range of the alpha rhythm. Figure 2a In this process, each alpha rhythm has a corresponding center frequency. By statistically analyzing the center frequencies of all alpha rhythms, the distribution of center frequencies is obtained, such as... Figure 2d As shown. In this embodiment of the invention, a center frequency range representing a preset percentage (e.g., 80% or more) of the total number of alpha rhythms is selected. α This refers to the main distribution frequency band range of the Alpha rhythm.

[0076] like Figure 2e As shown, the regulation of alpha rhythms is statistically analyzed to determine the proportion of dysregulation; for example... Figure 2f As shown, the amplitude modulation of the alpha rhythm is statistically analyzed to determine the proportion of poor amplitude modulation.

[0077] like Figures 2a-2e The amplitude modulation settings shown have an Alpha rhythm distributed in the 9-10Hz range, with a modulation ratio of 26.88% and an amplitude modulation ratio of 5.38%.

[0078] Preferably, extracting the power spectral density includes: calculating the power spectral density of the bandpass signal for each lead; and determining the distribution frequency band of the periodic component based on the curve corresponding to the calculated power spectral density.

[0079] Power spectral density (PSD) is the distribution of spectral energy per unit time.

[0080] As an example, the PSD of the EEG signal for each lead is calculated using the following formula:

[0081]

[0082] in, The Fourier transform form of signal x(t):

[0083]

[0084] This allows us to obtain the PSD curve for each lead, such as... Figure 3 As shown, each line represents a different lead, with the horizontal axis representing frequency and the vertical axis representing power spectral density. The periodic oscillation component is determined. The dominant frequency range of the EEG signal's PSD curve is detected to determine the distribution band F0 of the periodic component. The frequency distribution range of each band is as follows: delta (1-4Hz), theta (4-8Hz), alpha (8-13Hz), beta (13-30Hz), gamma (30-40Hz).

[0085] Preferably, extracting the sub-band energy distribution of the brain region includes: determining a set of sub-bands based on the bandpass signal of the occipital region lead, wherein the sub-bands include delta band, theta band, alpha band, beta band, gamma band, and full band; calculating the signal energy of each sub-band in the set of sub-bands; projecting the signal energy of each sub-band onto a topographic map; and merging the signal energy of each sub-band through brain region classification, wherein the brain region classification includes the prefrontal lobe, frontal lobe, central region, temporal lobe, parietal lobe, and occipital lobe.

[0086] Continuing with the example above, the EEG signal is passed through a bandpass filter to obtain a series of sub-bands, including delta (1-4Hz), theta (4-8Hz), alpha (8-13Hz), beta (13-30Hz), gamma (30-40Hz), and the full band (1-40Hz). The energy of the EEG signal in each sub-band is calculated using the following formula:

[0087]

[0088] Where x(t) represents the electroencephalogram (EEG) signal.

[0089] The energy of the EEG signal in each sub-band is projected onto the topographic map, such as... Figure 4 As shown on the right, this represents the topographic distribution of EEG energy in the sub-band throughout the brain.

[0090] The signal energy from all leads is combined according to, for example, six brain regions: prefrontal, frontal, central, temporal, parietal, and occipital. The left and right brain regions are shown separately. Figure 4 The left side shows the average EEG signal energy of the sub-band in each brain region. The diagonal line represents the left brain region, and the gray bar represents the right brain region.

[0091] Preferably, extracting the Alpha power asymmetry includes: calculating the power spectral density of the bandpass signal in the Alpha band; and for each brain region classification, calculating the left-right lateral asymmetry index SI for each brain region classification using the following formula:

[0092]

[0093] Among them, PSD L Power spectral density (PSD) represents the power spectral density of the left brain region. R This represents the power spectral density of the right brain region.

[0094] Following the example above, the power spectral density (PSD) is calculated using the EEG signal in the Alpha band obtained from the example. The asymmetry index (SI) for the left and right brain regions is calculated using equation (5) for each of the six brain regions. When SI is greater than 0, it indicates that the energy on the left side of the brain region is greater than that on the right; conversely, it indicates that the energy on the left side of the brain region is less than that on the right. Figure 5 As shown, forehead asymmetry reflects characteristics of impaired proximity motivation systems. Based on alpha power asymmetry, it was found that the alpha power on the left side of the forehead is greater than that on the right side.

[0095] Preferably, extracting the brain network characteristics includes: extracting the envelope based on the bandpass signal in the Alpha band; calculating the Pearson correlation value of the envelopes of any two leads to obtain a correlation matrix; and converting the correlation matrix into a small-world network.

[0096] Continuing with the previous example, the envelope of the EEG signal in the Alpha band obtained from the example is extracted. Then, the Pearson correlation coefficient (corr) of the envelopes of each pair of leads is calculated, representing the synchronicity of the signals between leads. The formula for calculating corr is as follows:

[0097]

[0098] Where X represents a signal from one lead, Y represents a signal from another lead, and μ represents the mean.

[0099] The above calculations yield a square matrix whose dimension equals the number of leads. A larger value for `corr` indicates higher signal synchronization between leads. Figure 6a As shown.

[0100] To aid in observing synchronized activity between brain regions, this embodiment of the invention preferably converts the correlation matrix into a "small-world" display. The correlation matrix is ​​binarized, retaining only those correlation values ​​greater than thresh, resulting in matrix mat0, where thresh is represented by the following formula:

[0101] thresh = mean corr +std corr (7)

[0102] Where, mean corr std represents the average value of the correlation matrix. corr This represents the standard deviation of the correlation matrix. For example... Figure 6b As shown, each node represents the corresponding lead, and the lines connecting the leads represent the correlation values. The larger the correlation value, the thicker the line.

[0103] The "small-world" visualization method reveals the balance between the left and right hemispheres of the brain. Specifically, the more symmetrical the left and right hemispheres are in the "small-world" network, the more balanced the functions between the two hemispheres; conversely, disparity indicates dysfunction in one hemisphere. This symmetry can be further quantified using mat0.

[0104] Step S130: Generate an EEG feature index report based on the extracted feature indicators.

[0105] The preferred embodiment of the present invention includes one or more of the following electroencephalogram (EEG) feature index reports: 1) textual and graphical representations of background EEG activity; 2) textual and graphical representations of the main frequency bands of periodic oscillation components; 3) textual and graphical representations of alpha power asymmetry; and 4) textual and graphical representations of brain network characteristics.

[0106] For example, 1) Automatically generate text and graphics about background EEG activity. For instance, the proportion of dysmodulation in the alpha rhythms from step S120 can be used to automatically generate text and graphics about background EEG activity. The proportion of dysmodulation in the alpha rhythms of the background activity can be used to evaluate the patient's EEG modulation. The automatic text format is, for example, "Based on the modulation of alpha rhythms, it was found that the alpha rhythms are distributed in the Range..." α (Hz), the poor adjustment ratio is R1 α The amplitude modulation defect rate is R2 α Automatic graphic formatting, for example, is... Figures 2a-2fAs shown. The modulation of the background activity alpha rhythm is quantified based on the rate of frequency change, while the amplitude modulation is quantified based on the amplitude difference of the rhythm waveform.

[0107] 2) Automatically generate text and graphics describing the main frequency bands of periodic oscillation components. For example, when evaluating the range of periodic oscillation components in EEG signals using PSD curves, the automatic text format could be something like, "Based on power spectral density, the main periodic oscillation components are in the F0 band"; the automatic graphic format could be something like... Figure 3 As shown.

[0108] 3) Automatically generate text and graphics about Alpha power asymmetry. For example, using Equation (5), when SI is greater than 0, it indicates that the energy on the left side of the brain region is greater than that on the right side, and vice versa. The automatic text format is, for example, “Forehead asymmetry reflects the characteristics of impaired proximity motivation systems. Based on Alpha power asymmetry, it was found that the Alpha power on the left side of the forehead is greater than (or less than) that on the right side”; the automatic graphic format is, for example,… Figure 5 As shown.

[0109] 4) Automatically generate text and graphics about brain network characteristics. For example, using mat0 to quantify symmetry, the automatic text format is something like "Based on functional connectivity and small-world networks of brain regions, we found that the functions are relatively balanced (or unbalanced) between the two hemispheres"; the automatic graphic format is something like... Figure 6a and 6b As shown.

[0110] Existing routine EEG indicator reports and their components, such as Figure 7 As shown, the component indicators extract their data without any processing, and the structure of the indicator report is immutable and the display is not intuitive. Embodiments of the present invention can generate an EEG feature indicator report composed of the above four sets of images and text (in any combination) according to the worker's selection, for example... Figure 8 As shown, feature indicators of the electroencephalogram (EEG) signal are extracted, and an EEG feature indicator report is generated based on the extracted feature indicators.

[0111] This invention can also provide user-defined routine EEG indicator reports. Different algorithms can be used to extract feature indicators for the same EEG signal, and different hospitals' clinical neurophysiologists have different needs for the output content of EEG reports, thus generating different EEG feature indicator reports. For example, to meet the needs of practical application scenarios, different routine EEG indicator reporting systems will provide multiple algorithm libraries corresponding to each feature indicator. For example, for brain network characteristics, the system can provide algorithms including functional connectivity, graph theory analysis, Granger causality analysis, etc. Different users (e.g., hospitals) can select the output indicators and corresponding algorithms of the EEG report according to their needs, such as brain region symmetry, brain region functional connectivity, etc. Specifically, brain region symmetry can be visualized using the algorithm of bilateral brain region alpha power difference index, and brain region functional connectivity can be visualized using the time-domain Pearson correlation algorithm, such as... Figure 8 As shown.

[0112] In addition, embodiments of the present invention can provide basic patterns for different categories of feature indicators (e.g., the algorithm for extracting feature indicators described in step 120). The basic EEG feature indicator report may include: 1) Alpha rhythm of background activity; 2) Power spectral density (PSD); 3) Subband energy distribution of brain regions; 4) Alpha power asymmetry; 5) Brain network characteristics.

[0113] This invention utilizes data processing technology to extract time-domain, frequency-domain, and spatial-domain features from electroencephalogram (EEG) signals, yielding characteristic indicators. Multiple algorithms are available for each characteristic indicator. Different users (e.g., clinical electrophysiologists) can select different EEG features and corresponding algorithms (or the default algorithm) to extract these indicators. The extracted indicators are then visualized to create a customized EEG characteristic indicator report, helping clinical electrophysiologists extract more valuable diagnostic information from EEG signals. Furthermore, by quantifying EEG features, the invention automatically generates textual reports of routine EEG indicators, improving the efficiency of EEG interpretation for clinical electrophysiologists.

[0114] This invention provides a control device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the computer program to implement the electroencephalogram visualization method described in steps S110-S130.

[0115] The processor contains a kernel, which retrieves the corresponding program units from memory. One or more kernels can be configured, and the visualization method of the electroencephalogram (EEG) described in steps S110-S130 can be implemented by adjusting the kernel parameters.

[0116] The memory may include non-permanent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0117] This invention also provides a machine-readable storage medium storing instructions that cause a machine to execute the electroencephalogram visualization method described in steps S110-S130.

[0118] This invention provides a visualization system for electroencephalograms (EEGs). The EEG visualization system includes the aforementioned control device and display module. The display module is used to display the EEG feature index report generated by the control device.

[0119] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0120] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0121] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0122] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0123] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0124] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0125] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0126] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, 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 that element.

[0127] The above are merely embodiments of this application and are not intended to limit the scope 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 the claims of this application.

Claims

1. A method for visualizing electroencephalograms (EEGs), characterized in that, The visualization methods for the electroencephalogram include: Collect electroencephalogram (EEG) signals; The EEG signals acquired through useless channels are removed, and the EEG signals after removing useless channels undergo rereference processing. The EEG signals after rereference processing are then preprocessed. The EEG signals acquired via the useless channels include electromyography (EMG) signals and bilateral mastoid point signals. The step of rereference processing the EEG signals after removing the useless channels includes: The mean of all the EEG signals in the whole brain is used as reference data, and the reference data is subtracted from each value of each EEG signal to obtain the EEG signal after rereference processing. The feature indices of the EEG signal after rereference processing are extracted, wherein the feature indices include amplitude modulation regulation, power spectral density, brain region sub-band energy distribution, alpha power asymmetry, and brain network characteristics; and Based on the extracted feature indicators, generate an EEG feature indicator report; The electroencephalogram (EEG) visualization method further includes, prior to the extraction of feature indicators of the EEG signal after rereference processing: A bandpass signal, including the alpha band, is obtained from the rereferenced EEG signal using a bandpass filter. Select the bandpass signal of the occipital region lead; The amplitude modulation of the EEG signal after rereference processing includes: Alpha band detection is performed on the bandpass signal of the pillow zone lead to determine the adjustment and amplitude modulation of the alpha band; Determining the adjustment and amplitude modulation of the Alpha band includes: The frequency change rate of the Alpha band is greater than the first threshold. The proportion of this ratio corresponds to the Alpha band being determined as poorly regulated. For each Alpha band, the amplitude of the highest peak in the middle of the spindle-shaped waveform is... Amplitude difference at both ends The highest peak amplitude proportion , to indicate the amplitude adjustment status; When the Less than the second threshold When the corresponding Alpha frequency band is determined to be of poor amplitude modulation, the proportion of Alpha frequency bands with poor amplitude modulation is statistically analyzed; and The center frequencies of the Alpha band are statistically analyzed to obtain the center frequency distribution. The center frequency range of a preset percentage is taken as the main distribution frequency band range of the Alpha band. Extracting the sub-band energy distribution of the brain region includes: Based on the bandpass signal of the pillow zone leads, a set of sub-frequency bands is determined, including the delta band, theta band, alpha band, beta band, and gamma band. Calculate the signal energy of each sub-band in the set of sub-bands; Projecting the signal energy of each sub-band onto the topographic map; and The signal energy of each sub-band is merged by brain region classification, which includes the frontal lobe, central region, temporal lobe, parietal lobe, and occipital lobe.

2. The method for visualizing electroencephalograms according to claim 1, characterized in that, The electroencephalogram (EEG) signals are acquired at a preset sampling rate and a preset number of channels.

3. The method for visualizing electroencephalograms according to claim 1, characterized in that, The EEG signal after rereference processing is preprocessed, including: The EEG signal after rereference processing is filtered at a preset frequency to remove basis drift; Power frequency noise interference is removed from the EEG signal after the base drift removal and rereference processing.

4. The method for visualizing electroencephalograms according to claim 1, characterized in that, Extracting the power spectral density includes: Calculate the power spectral density of the bandpass signal for each lead; Based on the curve corresponding to the calculated power spectral density, the distribution frequency band of the periodic component is determined.

5. The method for visualizing electroencephalograms according to claim 1, characterized in that, Extracting the Alpha power asymmetry includes: Calculate the power spectral density of the bandpass signal in the Alpha band; and For each brain region category, the left-right lateral asymmetry index for each category is calculated using the following formula. : in, This represents the power spectral density of the left brain region. This represents the power spectral density of the right brain region.

6. The method for visualizing electroencephalograms according to claim 1, characterized in that, Extracting the brain network characteristics, including: Extract the envelope based on the bandpass signal in the Alpha band; Calculate the Pearson correlation value of the envelopes of any two leads to obtain the correlation matrix; and The correlation matrix is ​​then converted into a small-world network.

7. The method for visualizing electroencephalograms according to claim 1, characterized in that, The EEG characteristic index report includes: Graphics and text about Alpha power asymmetry; Images and text about the characteristics of brain networks.

8. A control device, characterized in that, The control device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the electroencephalogram visualization method according to any one of claims 1-7.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores instructions that cause a computer to perform the method for visualizing an electroencephalogram according to any one of claims 1-7.

10. A visualization system for electroencephalograms (EEGs), characterized in that, The EEG visualization system includes the control device and display module as described in claim 8. The display module is used to display the electroencephalogram (EEG) feature index report generated by the control device.

Images