Brain wave monitoring visualization method and system

Abstract

The invention discloses a brain wave monitoring visualization method and system, and the method comprises the following steps: obtaining original brain electrical signals of a user at different time points and task scenes, and carrying out filtering, denoising and segmented cleaning to obtain a preprocessing data set; performing frequency domain analysis on the data set, and extracting waveband power distribution; generating thermodynamic diagram data based on the power value, dynamically superposing the thermodynamic diagram data to the visual field of the user through an augmented reality technology, and displaying the activity intensity of the brain area; and accumulating data for a long time to construct a personalized baseline database, comparing the current power with a scene threshold value in real time, generating warning feedback by associating the thermodynamic diagram when the power is abnormal, and optimizing the baseline database. According to the method, the abstract electroencephalogram signals are converted into the visual thermodynamic diagrams, personalized baselines and dynamic feedback are combined, the problems that traditional monitoring data is difficult to understand, depends on group threshold misjudgment and is lack of dynamic optimization are solved, and the intuition, accuracy and practicability of electroencephalogram monitoring are improved.

Description

Technical Field

[0001] This invention belongs to the field of brainwave glasses technology, and relates to a brainwave monitoring visualization method and system. Background Technology

[0002] Electroencephalography (EEG) monitoring, a non-invasive neuroscience technology, reflects brain neural activity by collecting electrical signals from the scalp surface and is widely used in clinical neurological disease diagnosis, cognitive science research, and consumer-grade health management. However, current EEG monitoring technologies still have significant shortcomings in data visualization and personalized adaptation, specifically as follows: Traditional EEG monitoring systems typically present data in the form of waveforms, power spectra, or discrete numerical values. Ordinary users find it difficult to intuitively understand the distribution of activity intensity in different brain regions through abstract waveforms or numbers. For example, the power values ​​of the δ / θ / α / β / γ bands output by clinical EEG instruments are usually displayed in tables or static charts, requiring professional interpretation. While consumer-grade EEG headbands simplify this to single indicators such as "attention scores," they still fail to reflect the specificity of brain regions, leading to an "information gap" in users' perception of their own EEG status.

[0003] Furthermore, existing technologies generally use fixed thresholds based on population statistics (such as mean β-wave power ± 2 standard deviations) as the criteria for anomaly detection, ignoring physiological differences between individuals. For example, some users may have an α-wave inhibition rate of 30% when focused, while others may have only 15%. Using population thresholds can easily lead to misjudgments, such as mislabeling normal states as abnormal or missing detections. If an abnormal state does not reach the population threshold, it may not be identified, affecting the accuracy of monitoring. Summary of the Invention

[0004] The present invention provides a method and system for visualizing electroencephalogram (EEG) monitoring. By preprocessing and frequency domain analysis of EEG signals to generate a visual heatmap, combined with augmented reality dynamic display and a personalized baseline database constructed from long-term data, the method aims to present complex EEG activity intuitively.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: A method for visualizing electroencephalogram (EEG) monitoring includes the following steps: S1. Obtain raw EEG signal data of the user at different time points and in different task scenarios, and perform filtering, noise reduction and signal segmentation cleaning on the raw EEG signal data to obtain the first EEG dataset; S2. Perform frequency domain analysis on the first EEG dataset to decompose the EEG components of different frequency bands, calculate the power value of each frequency band, and obtain the frequency band power distribution results; S3. Based on the frequency band power distribution results, generate first heat map data for visualization using preset color mapping rules; S4. The first heatmap data is superimposed onto the user's visual interface to dynamically update and display the activity intensity distribution of different areas of the brain, forming a second heatmap display effect; S5. Based on long-term data accumulation, construct a baseline database of the user's personal EEG. In real time, compare the current EEG power value with the normal power range threshold of the corresponding scene. If the threshold is exceeded, trigger an abnormality marker. Associate the abnormal information with the second heat map data to generate a third heat map data with warning labels and update the abnormal feedback.

[0006] Furthermore, the process of acquiring raw EEG signal data from users at different time points and in different task scenarios, and then filtering, denoising, and segmenting the raw EEG signal data to obtain a first EEG dataset includes the following steps: S11. Bandpass filtering technology is used to remove high-frequency noise and low-frequency drift from the original EEG signal to obtain the denoised signal; S12. The denoised signal is segmented according to a preset time window, segments with abnormal amplitude are removed, and the time-related information of the task scenario is combined to label them as "focused", "relaxed" and "fatigued" states to obtain the first EEG dataset.

[0007] Furthermore, the step of performing frequency domain analysis on the first EEG dataset to decompose the EEG components into different frequency bands, calculating the power value of each frequency band, and obtaining the frequency band power distribution results includes the following steps: S21. Apply frequency domain analysis to the segmented signals in the first EEG dataset to convert the time domain signals into frequency domain spectra; S22. Extract the components of each frequency band according to the standard EEG frequency band, and calculate the absolute power value and relative power value of each frequency band; S23. Abnormal power values ​​are filtered out by a preset threshold to obtain the power distribution result of the frequency band.

[0008] Furthermore, the step of generating first heatmap data for visualization based on the frequency band power distribution results using preset color mapping rules includes the following steps: S31. Standardize the power values ​​of each electrode point in the frequency band power distribution results to unify the power value range of different electrode points; S32. Divide the standardized power values ​​into high power group and low power group according to the preset power threshold, and map them to the preset warm color range and cool color range respectively; S33. Interpolate the regions between adjacent electrode points to generate spatially continuous first thermal map data.

[0009] Furthermore, the step of overlaying the first heatmap data onto the user's visual interface to dynamically update and display the activity intensity distribution of different brain regions, forming a second heatmap display effect, includes the following steps: S41. Match and calibrate the coordinate information of the first heat map data with the spatial coordinate system of the user's field of vision to ensure that the display position of the heat map corresponds to the display area of ​​the brain; S42. Set a dynamic refresh rate, receive and process the changing data of frequency band power distribution results in real time, and synchronously update the color distribution of the heat map to visualize the brain activity intensity in real time; S43. If the display clarity is lower than the preset threshold, adjust the brightness and contrast parameters of the display device to improve the visual recognition of the heat map; S44. Sharpen the boundaries of adjacent areas of the heat map to enhance the clarity of the outlines of different power areas, thus forming the second heat map display effect.

[0010] Furthermore, the construction of a user's personal EEG baseline database based on long-term data accumulation includes the following steps: S51. By continuously collecting EEG signal data from users in various task scenarios, statistical analysis is performed on the collected power values ​​of each frequency band to obtain the baseline parameters of the power of each frequency band, and a user's personal EEG baseline database is constructed based on the baseline parameters.

[0011] Furthermore, the real-time comparison of the current EEG power value with the normal power range threshold for the corresponding scene, and the triggering of an anomaly marker if the threshold is exceeded, includes the following steps: S52. Acquire the current EEG power value in real time and compare it with the normal power range threshold of the corresponding task scenario in the baseline database; S53. If the current power value exceeds the threshold, the distribution features of the abnormal region are extracted, and abnormal information containing the distribution features is generated.

[0012] Furthermore, the step of associating the anomaly information with the second heatmap data to generate third heatmap data with warning indicators, and updating the displayed anomaly feedback, includes the following steps: S54. Spatially correlate the abnormal information with the second heatmap data, mark warning signs in the abnormal area according to the preset severity rules, and generate the third heatmap data; S55. Update the third heatmap data via the display device and store the anomaly record to the baseline database; S56. Based on the abnormal records, optimize the normal power range threshold of the corresponding scenario in the baseline database using data analysis methods.

[0013] A brainwave monitoring visualization system, used to execute the aforementioned brainwave monitoring visualization method, comprising: The EEG signal acquisition module is used to acquire raw EEG signal data of users at different time points and in different task scenarios; The preprocessing module, connected to the EEG signal acquisition module, is used to filter and denoise the raw EEG signal data and perform signal segmentation and cleaning to output the first EEG dataset. The frequency domain analysis module, connected to the preprocessing module, is used to perform frequency domain analysis on the first EEG dataset, decompose the EEG components of different frequency bands, calculate the power value of each frequency band, and output the frequency band power distribution results. A heatmap generation module, connected to the frequency domain analysis module, is used to generate first heatmap data for visualization based on the frequency band power distribution results and using preset color mapping rules. A visualization module, connected to the heatmap generation module, is used to overlay the first heatmap data onto the user's visual interface, dynamically update and display the activity intensity distribution of different areas of the brain, and form a second heatmap display effect. The baseline database module is used to store a user's personal EEG baseline database built through long-term data accumulation. The baseline database contains normal power range thresholds for various task scenarios. An anomaly detection module, connected to the frequency domain analysis module, baseline database module, and visualization display module, is used to compare the current EEG power value with the normal power range threshold of the corresponding scene in the baseline database in real time. If the threshold is exceeded, an anomaly marker is triggered and an anomaly information is generated. The feedback optimization module, connected to the anomaly detection module, the baseline database module, and the visualization display module, is used to associate the anomaly information with the second heat map data to generate third heat map data with warning labels, update and display the anomaly feedback through the visualization display module, and store the anomaly record to the baseline database module to optimize the normal power range threshold.

[0014] The beneficial effects of this invention are as follows: By preprocessing and frequency domain analysis of complex EEG signals and transforming them into visualized heatmaps, which are then dynamically overlaid onto the user's field of vision using augmented reality technology, this invention effectively solves the problem of abstract data presentation and difficulty for users to intuitively understand brain activity in traditional EEG monitoring. Through long-term data accumulation, a personal EEG baseline database is built, avoiding misjudgments or missed detections caused by relying on group statistical thresholds, significantly improving the personalization and accuracy of anomaly detection. Simultaneously, the dynamic feedback mechanism linking abnormal states with heatmaps not only provides users with "visible" brain state cues but also feeds back into the baseline database through anomaly records, forming a continuous improvement chain of monitoring, feedback, and optimization. This allows the system to better align with individual user characteristics as usage time increases, further enhancing the practicality and reliability of monitoring. Overall, this invention achieves a leap from "professional interpretation" to "user self-perception" of EEG data, providing more intuitive and accurate technical support for scenarios such as mental health management and concentration training. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the method steps of the present invention.

[0016] Figure 2 This is a schematic diagram of the steps of the method for obtaining the frequency band power distribution structure according to the present invention.

[0017] Figure 3 This is a schematic diagram of the steps of the method for obtaining the first heat map data according to the present invention.

[0018] Figure 4 This is a schematic diagram illustrating the display effect of the second thermal map formed by the present invention.

[0019] Figure 5 This is a schematic diagram of the steps of the method for updating abnormal feedback in this invention. Detailed Implementation

[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains; the terminology used herein in the specification is for the purpose of describing particular embodiments only and is not intended to limit the invention; the terms "comprising" and "having," and any variations thereof, in the specification, claims, and foregoing drawings are intended to cover non-exclusive inclusion. The terms "first," "second," etc., in the specification, claims, or foregoing drawings are used to distinguish different objects and not to describe a particular order.

[0021] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0022] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0023] This invention provides an appendix Figures 1-5 In this embodiment of the invention, a method and system for visualizing brainwave monitoring is provided. By preprocessing and frequency domain analysis of brainwave signals to generate a visual heatmap, and combining augmented reality dynamic display with a personalized baseline database constructed from long-term data, the method aims to present complex brainwave activities intuitively. Specific Implementation Example 1 A method for visualizing electroencephalogram (EEG) monitoring includes the following steps: S1. Obtain raw EEG signal data of the user at different time points and in different task scenarios, and perform filtering, noise reduction and signal segmentation cleaning on the raw EEG signal data to obtain the first EEG dataset; S2. Perform frequency domain analysis on the first EEG dataset to decompose the EEG components of different frequency bands, calculate the power value of each frequency band, and obtain the frequency band power distribution results; S3. Based on the frequency band power distribution results, generate first heat map data for visualization using preset color mapping rules; S4. The first heatmap data is superimposed onto the user's visual interface to dynamically update and display the activity intensity distribution of different areas of the brain, forming a second heatmap display effect; S5. Based on long-term data accumulation, construct a baseline database of the user's personal EEG. In real time, compare the current EEG power value with the normal power range threshold of the corresponding scene. If the threshold is exceeded, trigger an abnormality marker. Associate the abnormal information with the second heat map data to generate a third heat map data with warning labels and update the abnormal feedback.

[0025] Specifically, users wear EEG glasses to perform tasks (scenarios can be categorized as focused work, relaxation with eyes closed, and fatigue). The system visualizes EEG monitoring through the following steps: First, raw EEG signals from different time points and scenarios are collected, filtered and denoised (removing environmental interference and physiological artifacts), segmented and cleaned (removing abnormal segments), and the scenario state is labeled to obtain the first EEG dataset; Second, frequency domain analysis is applied to the first EEG dataset to decompose it into EEG band components such as δ, θ, α, β, and γ, calculate the power value of each frequency band, and remove outliers to obtain the frequency band power distribution results reflecting the characteristics of EEG activity in different scenarios; Then, based on the power distribution results, a preset color mapping rule (high power warm color, low power cool color) is used to generate the second EEG dataset. The system first generates heatmap data (e.g., warm tones in the prefrontal cortex during focused scenes, and warm tones in the occipital lobe during relaxed scenes). Then, it overlays the first heatmap onto the visual interface via an AR display module, dynamically updating the activity intensity of each brain region (e.g., the warm tones in the prefrontal cortex expand during focused scenes, and the transitional tones appear in the parietal lobe during fatigue). Finally, based on long-term data accumulation, it statistically analyzes the normal power patterns of each scene to construct a personal EEG baseline database. If the power of a certain brain region exceeds the baseline threshold during monitoring, an abnormality marker is triggered, and a third heatmap with warning indicators is generated by associating it with the heatmap (e.g., abnormal areas are highlighted and flashed). The abnormal status is updated and displayed via AR, and abnormal records are stored to optimize the baseline threshold. Ultimately, this provides students with "visible" brain state feedback, helping them adjust their learning pace and improve efficiency.

[0026] This embodiment achieves end-to-end monitoring of EEG data from "acquisition" to "visualized early warning" through a complete process including data acquisition, preprocessing, frequency domain analysis, heatmap generation, AR display, baseline construction, and anomaly feedback. Each step is logically interconnected: preprocessing provides high-quality data for analysis; frequency domain analysis quantifies brain activity characteristics; heatmaps transform abstract data into intuitive colors; display provides real-time dynamic feedback; and baseline construction and anomaly feedback enhance the personalization and accuracy of monitoring through long-term data. The final result provides users with "visible" feedback on their brain state (such as an expansion of warm tones in the prefrontal cortex when focused, and flashing warning signs in the parietal lobe when fatigued), assisting them in proactively adjusting their work rhythm and improving work efficiency.

[0027] Furthermore, this application features environmental and behavioral monitoring functions, which collect data in real time on the user's environment, including light intensity, sitting posture (via gyroscope / accelerometer), daily wearing time, infrared distance detection (anti-collision warning), and outdoor activity time (GPS positioning). It also includes a built-in microphone to collect ambient noise levels in decibels, with data saved every minute for long-term behavioral analysis. Simultaneously, the EEG monitoring function uses four high-precision dry electrodes (distributed in the Fp1 / Fp2 area of ​​the forehead and the T3 / T4 area of ​​the temporal lobe) built into the glasses to continuously collect raw EEG signals at a certain sampling rate. This allows for real-time analysis of focus (β-wave power ratio), fatigue level (θ-wave power ratio), and emotional state (α / β-wave balance), generating segmented signals at intervals for real-time processing. Preprocessed EEG data, environmental data, and behavioral tags are encrypted and transmitted to the cloud backend via wireless network (Wi-Fi / 4G / 5G). The AI ​​system performs multimodal data fusion analysis: it associates EEG indicators with environmental parameters to construct a user status profile, applies an LSTM model to predict fatigue risk (e.g., triggering an alert when the theta wave proportion is >30% and the sitting posture tilt is >15°), and identifies the optimal focus environment through cluster analysis (e.g., a 20% increase in focus when noise is <50dB). Finally, it automatically generates daily / weekly node reports containing EEG indicator trend charts, statistics of environmental influencing factors, and optimization suggestions, which are synchronized to the mobile app (dynamic heat map + line graph) and the web backend (detailed data table) for users to view. Specific Implementation Example 2 The process of acquiring raw EEG signal data from users at different time points and in different task scenarios, and then filtering, denoising, and segmenting the raw EEG signal data to obtain a first EEG dataset includes the following steps: S11. Bandpass filtering technology is used to remove high-frequency noise and low-frequency drift from the original EEG signal to obtain the denoised signal; The user wears an EEG monitoring device, such as EEG glasses, and assumes that the user completes the following tasks sequentially within 30 minutes: Scenario 1 (0-10 minutes): Focused reading (cognitive task, requires concentration); Scene 2 (10-15 minutes): Close your eyes and relax (no external interference, in a natural and relaxed state); Scenario 3 (15-30 minutes): Simulate fatigue (decreased attention after continuous reading, resulting in yawning, increased physical movements, etc.).

[0029] The device continuously collects raw EEG signals from the scalp surface at a set sampling rate. The initial signals include environmental electromagnetic interference (such as 50Hz power supply noise), high-amplitude artifacts (amplitude approximately ±100μV) generated by the user's blinking, and electromyographic interference (high-frequency noise).

[0030] The EEG glasses acquire raw EEG signal data of the user in the above three task scenarios, forming a continuous raw signal stream. The initial signal exhibits obvious noise due to environmental interference and physiological artifacts (such as blinking and muscle activity). For example, 50Hz power supply noise causes high-frequency sawtooth waves to be superimposed on the signal, blinking causes the local signal amplitude to rise sharply to ±100μV (far exceeding the normal EEG signal amplitude range of ±20-50μV), and muscle activity (such as frowning) introduces high-frequency noise of 30-100Hz.

[0031] By applying bandpass filtering technology to the raw EEG signal, the passband of the bandpass filter can be set to 1-40Hz (covering the entire EEG frequency band of δ, θ, α, β, and γ), effectively filtering out high-frequency noise (such as 50Hz power supply interference) and low-frequency drift (such as baseline fluctuations caused by poor electrode contact), significantly improving the stability of the signal baseline. In the denoised signal, the characteristics of α waves (in a relaxed scenario with eyes closed) and β waves (in a focused reading scenario) are significantly enhanced.

[0032] S12. The denoised signal is segmented according to a preset time window, segments with abnormal amplitude are removed, and the time-related information of the task scenario is combined to label them as "focused", "relaxed" and "fatigued" states to obtain the first EEG dataset.

[0033] Specifically, the denoised signal is divided into segments according to a preset time window of 2 seconds (at a sampling rate of 500Hz, each segment contains 500Hz×2s=1000 sampling points), resulting in a continuous set of signal segments (a total of 30 minutes×60 seconds / 2 seconds=900 segments). The preset time can be changed according to actual needs.

[0034] Furthermore, an amplitude threshold of ±75μV was set (the upper limit of normal EEG signal amplitude determined based on historical data statistics). Amplitude detection was performed on each signal segment: if the amplitude of a sampling point in a signal segment exceeded ±75μV (e.g., the amplitude of a signal segment reached ±90μV due to blinking), it was judged as an abnormal segment and removed. After cleaning, about 300 effective segments remained (removal rate of about 66.7%), and the amplitude of the effective segments were all within the range of ±20-70μV.

[0035] Furthermore, based on the timeline of the task scenario (0-10 minutes for "focus", 10-15 minutes for "relaxation", and 15-30 minutes for "fatigue"), the effective segments after cleaning were labeled with the following statuses: effective segments of 0-10 minutes were labeled as "focus" (approximately 100 segments); effective segments of 10-15 minutes were labeled as "relaxation" (approximately 50 segments); and effective segments of 15-30 minutes were labeled as "fatigue" (approximately 150 segments).

[0036] Through the above filtering, segmentation, cleaning, and labeling operations, a first EEG dataset containing 300 valid signal segments was finally obtained. Each segment was associated with a scene label of "focus," "relaxation," or "fatigue," and the signal quality met the requirements for subsequent frequency domain analysis (no obvious noise interference, stable amplitude range).

[0037] This embodiment achieves the transformation of raw EEG signals into a high-quality first EEG dataset through a complete process of acquisition, filtering and noise reduction, segmentation and cleaning, and scene annotation. Bandpass filtering effectively removes environmental noise, segmentation and cleaning eliminates physiological artifacts, and scene annotation provides clear state correlation information for subsequent analysis, ensuring data reliability and scene specificity. Specific Implementation Example 3 The step of performing frequency domain analysis on the first EEG dataset to decompose the EEG components into different frequency bands, calculating the power value of each frequency band, and obtaining the frequency band power distribution results includes the following steps: S21. Apply frequency domain analysis to the segmented signals in the first EEG dataset to convert the time domain signals into frequency domain spectra; Specifically, Fast Fourier Transform (FFT) was applied to each 2-second segment signal (500Hz sampling rate, 1000 sampling points per segment) in the first EEG dataset, and a Hanning window function was used to reduce spectral leakage, converting the voltage-time waveform in the time domain into a power-frequency spectrum in the frequency domain, clearly showing the energy distribution of different frequency components.

[0039] S22. Extract the components of each frequency band according to the standard EEG frequency band, and calculate the absolute power value and relative power value of each frequency band; Specifically, the frequency domain spectrum is truncated according to the standard EEG frequency bands (δ: 0.5-4Hz, θ: 4-8Hz, α: 8-13Hz, β: 13-30Hz, γ: >30Hz), and the components of each band are extracted; combined with scene tags, the corresponding advantageous frequency bands are extracted (e.g., α waves are extracted for relaxation scenes, β waves for focus scenes, and θ waves for fatigue scenes). Calculate the absolute power (integrated spectral area, unit μV² / Hz) and relative power (the percentage of power in a certain frequency band to the total power) for each band component; for example, in the relaxation scene, the absolute power of the alpha wave is 20μV² / Hz, the total power is 50μV² / Hz, and the relative power is 40%; in the focus scene, the absolute power of the beta wave is 15μV² / Hz, the total power is 40μV² / Hz, and the relative power is 37.5%.

[0040] S23. Abnormal power values ​​are filtered out by a preset threshold to obtain the power distribution result of the frequency band.

[0041] Specifically, an abnormal power value exceeding the mean ± 2.5 times the standard deviation of the power in each frequency band under each scenario is used as a threshold to remove abnormal power values. For example, in the fatigue scenario, the theta wave power of a certain segment is 25 μV² / Hz (mean 12 μV² / Hz, standard deviation 5 μV² / Hz, upper threshold 24.5 μV² / Hz), and it is removed because it exceeds the threshold. After the above processing, the frequency band power distribution results for each scenario are obtained; for example, in the relaxation scenario, the alpha wave accounts for 40% (dominant frequency band), in the focus scenario, the beta wave accounts for 37.5% (dominant frequency band), and in the fatigue scenario, the theta wave accounts for 34.3% (dominant frequency band), intuitively reflecting the characteristics of EEG activity under different cognitive states. Specific Implementation Example 4 The step of generating first heatmap data for visualization based on the frequency band power distribution results and using preset color mapping rules includes the following steps: S31. Standardize the power values ​​of each electrode point in the frequency band power distribution results to unify the power value range of different electrode points; Specifically, a minimum-maximum normalization method is used, where the normalized value = (original power value - minimum power value) / (maximum power value - minimum power value). This method maps the original power values ​​of each electrode point to a uniform range of 0-1. For example, the original power range of the prefrontal cortex β wave is 0.2-0.8, and after normalization, the value is 1.0 for the focused state and 0.0 for the relaxed state. The original power range of the occipital cortex α wave is 0.3-0.9, and after normalization, the value is 1.0 for the relaxed state and 0.0 for the focused state. This ensures that the power value range of different electrode points is consistent, which facilitates subsequent color mapping.

[0043] S32. Divide the standardized power values ​​into high power group and low power group according to the preset power threshold, and map them to the preset warm color range and cool color range respectively; Specifically, based on a preset power threshold of 0.6, the standardized power values ​​are divided into a high-power group (>0.6, mapped to warm colors) and a low-power group (≤0.6, mapped to cool colors). The high-power group corresponds to areas of strong brain activity and uses warm colors such as red (RGB(255,0,0)) and orange (RGB(255,165,0)). The low-power group corresponds to areas of weak activity and uses cool colors such as blue (RGB(0,0,255)) and cyan (RGB(0,255,255)). For example, in a relaxed state, a standardized power of 0.85 for occipital alpha waves (>0.6) is mapped to red, and a standardized power of 0.3 for prefrontal beta waves (≤0.6) is mapped to blue. In a focused state, a standardized power of 0.75 for prefrontal beta waves (>0.6) is mapped to orange.

[0044] S33. Interpolate the regions between adjacent electrode points to generate spatially continuous first thermal map data.

[0045] Specifically, a bilinear interpolation algorithm is applied to the regions between adjacent electrode points to calculate the power value of the intermediate region based on the electrode point coordinates, generating a smooth color gradient. For example, the region between the C3 (0.7, orange) and C4 (0.4, blue) electrode points in the parietal lobe forms an orange-blue gradient through interpolation; the region between the O1 (red) and O2 (red) electrode points in the occipital lobe forms a continuous red high-power region after interpolation, ensuring the spatial continuity of the heatmap. After standardization, group mapping, and interpolation processing, spatially continuous first heatmap data is generated. Examples show that in a relaxed state, the occipital lobe is a concentrated red area (high power alpha waves) and the prefrontal lobe is blue (low power beta waves); in a focused state, the prefrontal lobe is a concentrated orange area (high power beta waves) and the occipital lobe is cyan (low power alpha waves), intuitively reflecting the differences in activity intensity of different brain regions in different states. Specific Implementation Example 5 The step of overlaying the first heatmap data onto the user's visual interface and dynamically updating the distribution of activity intensity in different brain regions to form a second heatmap display effect includes the following steps: S41. Match and calibrate the coordinate information of the first heat map data with the spatial coordinate system of the user's field of vision to ensure that the display position of the heat map corresponds to the display area of ​​the brain; Specifically, the user's head posture and visual field coordinates are obtained through the inertial measurement unit (IMU) and visual positioning module built into the EEG glasses. The preset coordinates of each electrode point in the first heat map (such as Fp1 in the prefrontal lobe and O1 in the occipital lobe) are aligned with the brain model or virtual brain region markers in the user's field of vision. The offset is adjusted by the user's manual confirmation (such as "confirm that the red area of ​​the occipital lobe coincides with the position of the back of the head"), and finally the heat map and the actual brain region are accurately matched.

[0047] S42. Set a dynamic refresh rate, receive and process the changing data of frequency band power distribution results in real time, and synchronously update the color distribution of the heat map to visualize the brain activity intensity in real time; Specifically, the dynamic refresh rate is set to 60Hz (data is received every 1 / 60th of a second), and the latest frequency band power values ​​are obtained from the frequency domain analysis module in real time (e.g., the prefrontal cortex beta wave power increases from 0.75 to 0.8). The color of the corresponding brain region is adjusted according to the power change (e.g., 0.75 beta wave power corresponds to orange, and 0.8 corresponds to red). When the user starts reading, the prefrontal cortex gradually changes from orange to red, reflecting the dynamic changes in the level of attention in real time.

[0048] S43. If the display clarity is lower than the preset threshold, adjust the brightness and contrast parameters of the display device to improve the visual recognition of the heat map; Specifically, the clarity of the heatmap is detected by the SSIM structural similarity index (initially 60%). If it is lower than the preset threshold (80%), the brightness of the display device is adjusted (from 100 cd / m² to 120 cd / m²) and the contrast ratio is adjusted (from 500:1 to 600:1). After optimization, the clarity is improved to 85%, and the boundary between red (occipital lobe) and orange (prefrontal lobe) is clearer, allowing users to clearly distinguish the activity intensity of different brain regions.

[0049] S44. Sharpen the boundaries of adjacent areas of the heat map to enhance the clarity of the outlines of different power areas, thus forming the second heat map display effect.

[0050] Specifically, the Sobel edge detection algorithm is applied to identify the boundaries of different colored regions (such as the boundary between the red occipital lobe and the blue temporal lobe). Gradient enhancement (increasing brightness by 20%) is performed on the boundary pixels, and the gradient areas are smoothed (reducing jagged edges). After processing, the boundary between the high-activity area (red) and the low-activity area (blue) changes from a blurry gradient to a clear outline, allowing users to intuitively see a clear division of activity intensity. After calibration, dynamic updates, clarity optimization, and boundary sharpening, a second heatmap display effect is formed. When the eyes are closed and relaxed, the occipital lobe shows a clear red concentrated area (high alpha wave power), with a clear boundary with the blue low-activity area of ​​the temporal lobe. When focusing on reading, the prefrontal cortex gradually changes from orange to red (beta wave power increase), forming a "front-back" dual active area with the red area of ​​the occipital lobe. Users can intuitively perceive the transition of their attention state through the dynamic color changes in their visual field. Specific Implementation Example Six The process of building a user's personal electroencephalogram (EEG) baseline database based on long-term data accumulation includes the following steps: S51. By continuously collecting EEG signal data from users in various task scenarios, statistical analysis is performed on the collected power values ​​of each frequency band to obtain the baseline parameters of the power of each frequency band, and a user's personal EEG baseline database is constructed based on the baseline parameters.

[0052] Specifically, users can wear EEG glasses to record EEG signals for 30 consecutive days (8 hours per day, 240 hours in total) during typical scenarios such as "focused work," "relaxation and rest," and "post-exercise fatigue" (3 hours, 1 hour, and 1 hour per day, respectively). Scene tags and timestamps are collected simultaneously, generating approximately 432,000 2-second segmented signals. After data cleaning and noise removal, approximately 380,000 valid segments are retained. Frequency domain analysis is then used to calculate the power values ​​(absolute and relative power) of the δ, θ, α, β, and γ frequency bands for each scenario, and the data is grouped and statistically analyzed according to scenario. The mean (μ) and standard deviation (σ) of the frequency band power are calculated. Based on the statistical results, the normal power range thresholds for each frequency band in each scenario are determined using the mean ± 2 times the standard deviation (covering approximately 95% of normal data). For example, the normal range for β waves in a focused work scenario is 12-24 μV² / Hz, and for α waves in a relaxed rest scenario, it is 14-30 μV² / Hz. Finally, the mean, standard deviation, and normal range thresholds for each scenario are stored in a structured manner to form a personal EEG baseline database for each user, providing a personalized reference standard for subsequent abnormality detection. The preset values ​​such as the time can be set according to actual needs. Specific Implementation Example 7 The real-time comparison of the current EEG power value with the normal power range threshold for the corresponding scene, and the triggering of an abnormality marker if the threshold is exceeded, includes the following steps: S52. Acquire the current EEG power value in real time and compare it with the normal power range threshold of the corresponding task scenario in the baseline database; S53. If the current power value exceeds the threshold, the distribution features of the abnormal region are extracted, and abnormal information containing the distribution features is generated.

[0054] Users wear EEG glasses to perform tasks. The system obtains the current EEG power value in real time through the frequency domain analysis module (a segmented signal every 2 seconds), and retrieves the normal power range threshold for the "focused work" scenario from the personal EEG baseline database based on the user's task tag. When comparing the current power value with the threshold, if it is found to be lower than the lower limit of the threshold, it is judged as abnormal. Subsequently, the abnormal region features (such as the location of the abnormal brain region, the direction of the power deviation, and the abnormal coverage area) are extracted to generate abnormal information including the abnormal time, abnormal region, degree of deviation, and coverage area, providing data support for subsequent warnings.

[0055] Further, as an example, a user wears an EEG acquisition device to give a 30-minute meeting presentation ("focused work" scenario). The system uses a frequency domain analysis module to obtain the absolute power value of the beta wave (10 μV² / Hz) of the current segment signal (15 minutes) in real time, and retrieves the normal range threshold of beta waves (12-24 μV² / Hz) for the "focused work" scenario from the user's EEG baseline database. The comparison shows that the current power (10 μV² / Hz) is lower than the lower limit of the threshold (12 μV² / Hz), which is judged as abnormal. The abnormal region features (prefrontal Fp1 / Fp2 electrode points, deviation value -2 μV² / Hz, covering 3% of electrode points) are extracted to generate abnormal information including abnormal time (15 minutes of the meeting), region (prefrontal Fp1 / Fp2), deviation value (-2 μV² / Hz), and area (3%), providing data support for subsequent warnings. Specific Implementation Example 8 The step of associating the anomaly information with the second heatmap data to generate third heatmap data with warning indicators, and updating the anomaly feedback display, includes the following steps: S54. Spatially correlate the abnormal information with the second heatmap data, mark warning signs in the abnormal area according to the preset severity rules, and generate the third heatmap data; S55. Update the third heatmap data via the display device and store the anomaly record to the baseline database; S56. Based on the abnormal records, optimize the normal power range threshold of the corresponding scenario in the baseline database using data analysis methods.

[0057] Specifically, when a user wears AR EEG glasses for evening study, the system detects abnormal prefrontal cortex beta wave power (below the normal threshold for the "focused learning" scenario) and generates an abnormal record containing information such as time, region, and severity (mild). After spatially associating the abnormal region (prefrontal cortex) with the corresponding brain region location in the second heatmap data, a warning label is marked according to preset rules (mild cases are marked with a red border), generating a third heatmap data. The third heatmap is rendered in real time through the AR device, and a red border warning for the prefrontal cortex is displayed in the user's field of vision, intuitively perceiving the state of distracted attention. The abnormal information (time, scenario, region, severity, etc.) is stored in a structured format (such as JSON) in the "abnormal record" sub-table of the baseline database. The data analysis module is periodically triggered to adjust the normal power threshold for the "focused learning" scenario based on historical abnormal records (such as multiple mild abnormalities within this week) (such as lowering the lower limit), making the baseline database more consistent with the user's recent state and improving the accuracy of subsequent monitoring. Specific Implementation Example Nine A brainwave monitoring visualization system, used to execute the aforementioned brainwave monitoring visualization method, comprising: The EEG signal acquisition module is used to acquire raw EEG signal data of users at different time points and in different task scenarios; The preprocessing module, connected to the EEG signal acquisition module, is used to filter and denoise the raw EEG signal data and perform signal segmentation and cleaning to output the first EEG dataset. The frequency domain analysis module, connected to the preprocessing module, is used to perform frequency domain analysis on the first EEG dataset, decompose the EEG components of different frequency bands, calculate the power value of each frequency band, and output the frequency band power distribution results. A heatmap generation module, connected to the frequency domain analysis module, is used to generate first heatmap data for visualization based on the frequency band power distribution results and using preset color mapping rules. A visualization module, connected to the heatmap generation module, is used to overlay the first heatmap data onto the user's visual interface, dynamically update and display the activity intensity distribution of different areas of the brain, and form a second heatmap display effect. The baseline database module is used to store a user's personal EEG baseline database built through long-term data accumulation. The baseline database contains normal power range thresholds for various task scenarios. An anomaly detection module, connected to the frequency domain analysis module, baseline database module, and visualization display module, is used to compare the current EEG power value with the normal power range threshold of the corresponding scene in the baseline database in real time. If the threshold is exceeded, an anomaly marker is triggered and an anomaly information is generated. The feedback optimization module, connected to the anomaly detection module, the baseline database module, and the visualization display module, is used to associate the anomaly information with the second heat map data to generate third heat map data with warning labels, update and display the anomaly feedback through the visualization display module, and store the anomaly record to the baseline database module to optimize the normal power range threshold.

[0059] This invention can be used in a wide range of general-purpose or special-purpose computer system environments or configurations.

[0060] Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments that include any of the above systems or devices.

[0061] This invention can be described in the general context of computer-executable instructions that are executed by a computer, such as program modules.

[0062] Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This invention can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via communication networks.

[0063] In a distributed computing environment, program modules can reside on local and remote computer storage media, including storage devices.

[0064] Specifically, those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by instructing related hardware through computer-readable instructions. These computer-readable instructions can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. The aforementioned storage medium can be a non-volatile storage medium such as a magnetic disk, optical disk, or read-only memory (ROM), or random access memory (RAM).

[0065] It should be understood that although the steps in the flowcharts in the accompanying drawings are shown sequentially as indicated by the arrows, these steps are not necessarily performed in the order indicated by the arrows. Unless otherwise expressly stated herein, there is no strict order in which these steps are performed, and they may be performed in other orders.

[0066] Moreover, at least some steps in the flowchart of the attached figure may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. Their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0067] Obviously, the embodiments described above are only some embodiments of the present invention, and not all embodiments. The accompanying drawings show preferred embodiments of the present invention, but do not limit the scope of the invention. The present invention can be implemented in many different forms; rather, these embodiments are provided to provide a thorough and complete understanding of the disclosure of the present invention.

[0068] Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing specific embodiments or make equivalent substitutions for some of the technical features. Any equivalent structures made using the content of this specification and drawings, whether directly or indirectly applied to other related technical fields, are similarly within the scope of protection of this patent.

Claims

1. A method for visualizing electroencephalogram (EEG) monitoring, characterized in that, Includes the following steps: S1. Obtain raw EEG signal data of the user at different time points and in different task scenarios, and perform filtering, noise reduction and signal segmentation cleaning on the raw EEG signal data to obtain the first EEG dataset; S2. Perform frequency domain analysis on the first EEG dataset to decompose the EEG components of different frequency bands, calculate the power value of each frequency band, and obtain the frequency band power distribution results; S3. Based on the frequency band power distribution results, generate first heat map data for visualization using preset color mapping rules; S4. The first heatmap data is superimposed onto the user's visual interface to dynamically update and display the activity intensity distribution of different areas of the brain, forming a second heatmap display effect; S5. Based on long-term data accumulation, construct a baseline database of the user's personal EEG. In real time, compare the current EEG power value with the normal power range threshold of the corresponding scene. If the threshold is exceeded, trigger an abnormality marker. Associate the abnormal information with the second heat map data to generate a third heat map data with warning labels and update the abnormal feedback.

2. The brainwave monitoring visualization method according to claim 1, characterized in that, The process of acquiring raw EEG signal data from users at different time points and in different task scenarios, and then filtering, denoising, and segmenting the raw EEG signal data to obtain a first EEG dataset includes the following steps: S11. Bandpass filtering technology is used to remove high-frequency noise and low-frequency drift from the original EEG signal to obtain the denoised signal; S12. The denoised signal is segmented according to a preset time window, segments with abnormal amplitude are removed, and the time-related information of the task scenario is combined to label them as "focused", "relaxed" and "fatigued" states to obtain the first EEG dataset.

3. The brainwave monitoring visualization method according to claim 1, characterized in that, The step of performing frequency domain analysis on the first EEG dataset to decompose the EEG components into different frequency bands, calculating the power value of each frequency band, and obtaining the frequency band power distribution results includes the following steps: S21. Apply frequency domain analysis to the segmented signals in the first EEG dataset to convert the time domain signals into frequency domain spectra; S22. Extract the components of each frequency band according to the standard EEG frequency band, and calculate the absolute power value and relative power value of each frequency band; S23. Abnormal power values ​​are filtered out by a preset threshold to obtain the power distribution result of the frequency band.

4. The brainwave monitoring visualization method according to claim 1, characterized in that, The step of generating first heatmap data for visualization based on the frequency band power distribution results and using preset color mapping rules includes the following steps: S31. Standardize the power values ​​of each electrode point in the frequency band power distribution results to unify the power value range of different electrode points; S32. Divide the standardized power values ​​into high power group and low power group according to the preset power threshold, and map them to the preset warm color range and cool color range respectively; S33. Interpolate the regions between adjacent electrode points to generate spatially continuous first thermal map data.

5. The brainwave monitoring visualization method according to claim 1, characterized in that, The step of overlaying the first heatmap data onto the user's visual interface and dynamically updating the distribution of activity intensity in different brain regions to form a second heatmap display effect includes the following steps: S41. Match and calibrate the coordinate information of the first heat map data with the spatial coordinate system of the user's field of vision to ensure that the display position of the heat map corresponds to the display area of ​​the brain; S42. Set a dynamic refresh rate, receive and process the changing data of frequency band power distribution results in real time, and synchronously update the color distribution of the heat map to visualize the brain activity intensity in real time; S43. If the display clarity is lower than the preset threshold, adjust the brightness and contrast parameters of the display device to improve the visual recognition of the heat map; S44. Sharpen the boundaries of adjacent areas of the heat map to enhance the clarity of the outlines of different power areas, thus forming the second heat map display effect.

6. The brainwave monitoring visualization method according to claim 1, characterized in that, The process of building a user's personal electroencephalogram (EEG) baseline database based on long-term data accumulation includes the following steps: S51. By continuously collecting EEG signal data from users in various task scenarios, statistical analysis is performed on the collected power values ​​of each frequency band to obtain the baseline parameters of the power of each frequency band, and a user's personal EEG baseline database is constructed based on the baseline parameters.

7. The brainwave monitoring visualization method according to claim 6, characterized in that, The real-time comparison of the current EEG power value with the normal power range threshold for the corresponding scene, and the triggering of an abnormality marker if the threshold is exceeded, includes the following steps: S52. Acquire the current EEG power value in real time and compare it with the normal power range threshold of the corresponding task scenario in the baseline database; S53. If the current power value exceeds the threshold, the distribution features of the abnormal region are extracted, and abnormal information containing the distribution features is generated.

8. The brainwave monitoring visualization method according to claim 7, characterized in that, The step of associating the anomaly information with the second heatmap data to generate third heatmap data with warning indicators, and updating the anomaly feedback display, includes the following steps: S54. Spatially correlate the abnormal information with the second heatmap data, mark warning signs in the abnormal area according to the preset severity rules, and generate the third heatmap data; S55. Update the third heatmap data via the display device and store the anomaly record to the baseline database; S56. Based on the abnormal records, optimize the normal power range threshold of the corresponding scenario in the baseline database using data analysis methods.

9. A brainwave monitoring visualization system, used to execute the brainwave monitoring visualization method according to claims 1 to 8, characterized in that, include: The EEG signal acquisition module is used to acquire raw EEG signal data of users at different time points and in different task scenarios; The preprocessing module, connected to the EEG signal acquisition module, is used to filter and denoise the raw EEG signal data and perform signal segmentation and cleaning to output the first EEG dataset. The frequency domain analysis module, connected to the preprocessing module, is used to perform frequency domain analysis on the first EEG dataset, decompose the EEG components of different frequency bands, calculate the power value of each frequency band, and output the frequency band power distribution results. A heatmap generation module, connected to the frequency domain analysis module, is used to generate first heatmap data for visualization based on the frequency band power distribution results and using preset color mapping rules. A visualization module, connected to the heatmap generation module, is used to overlay the first heatmap data onto the user's visual interface, dynamically update and display the activity intensity distribution of different areas of the brain, and form a second heatmap display effect. The baseline database module is used to store a user's personal EEG baseline database built through long-term data accumulation. The baseline database contains normal power range thresholds for various task scenarios. An anomaly detection module, connected to the frequency domain analysis module, baseline database module, and visualization display module, is used to compare the current EEG power value with the normal power range threshold of the corresponding scene in the baseline database in real time. If the threshold is exceeded, an anomaly marker is triggered and an anomaly information is generated. The feedback optimization module, connected to the anomaly detection module, the baseline database module, and the visualization display module, is used to associate the anomaly information with the second heat map data to generate third heat map data with warning labels, update and display the anomaly feedback through the visualization display module, and store the anomaly record to the baseline database module to optimize the normal power range threshold.

Images