Brain wave analysis and diagnosis method based on brain-computer interface
Through the brain wave analysis and diagnosis method based on brain-computer interface, through frequency band division and feature data processing, combined with historical and template feature data, the problem of individual differences affecting the analysis results is solved, and diagnostic accuracy and detection efficiency are improved.
Patent Information
- Application Number
- CN202411142624.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-20
- Publication Date
- 2024-11-22
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing brain wave analysis and diagnosis methods are affected due to large individual differences and fuzzy brain wave characteristics analysis.
Using brain wave analysis and diagnosis method based on brain-computer interface, the frequency band division and feature data are extracted by obtaining brain wave frequency signals in δ, θ, α, β and γ bands, and the frequency band division and feature data are extracted. The model is used for comparison and analysis, combining the patient's history. Feature data and template feature data, the diagnostic results are classified and processed using preset thresholds to generate visual list data.
It improves the accuracy of diagnostic data of brain wave analysis, reduces individual differences, enhances the efficiency of detection of brain problems, and provides more accurate and rapid diagnostic results.
Smart Images

Figure CN119004074A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of brain wave diagnosis, and in particular to a brain wave analysis and diagnosis method based on a brain-computer interface. Background Art
[0002] In today's medical and neuroscience fields, brain-computer interface-based EEG analysis and diagnosis methods are gradually emerging. With the rapid development of science and technology, our exploration of the brain is deepening. As the electrophysiological manifestation of brain activity, brain waves contain rich information. Traditional medical diagnostic methods have limitations in the detection of certain neurological diseases, and the emergence of brain-computer interface technology has brought new opportunities for brain wave analysis. By directly collecting electrical signals generated by the brain, we can obtain the activity state of the brain in a more real-time and accurate manner. Early brain wave research mainly relied on simple EEG equipment, but these devices have deficiencies in signal resolution and acquisition speed. With the advancement of sensor technology, signal processing algorithms and artificial intelligence, brain-computer interfaces can capture more subtle and complex brain wave changes. This technology not only has potential in disease diagnosis, such as epilepsy, Parkinson's disease, Alzheimer's disease, etc., but also provides a new perspective for the evaluation of mental illnesses, such as depression, schizophrenia, etc.
[0003] However, the existing EEG analysis and diagnosis methods may affect the accuracy of the analysis results due to large individual differences and vague analysis of EEG characteristics. To this end, we propose a EEG analysis and diagnosis method based on brain-computer interface. Summary of the invention
[0004] The purpose of the present invention is to provide a method for analyzing and diagnosing brain waves based on a brain-computer interface.
[0005] In order to solve the problems raised in the above background technology, the present invention provides the following technical solutions: a method for brain wave analysis and diagnosis based on brain-computer interface, comprising the following steps:
[0006] S100. Obtaining the pre-processed δ, θ, α, β and γ band brain wave frequency signals, the time of brain wave frequency monitoring and the patient's personal information;
[0007] S200. Divide different frequency bands into multiple equal-distance divisions, extract feature data for each division, and transmit the feature data for each division to the model for comparison and analysis.
[0008] As a further solution of the present invention: the specific process of frequency band division is:
[0009] S210. Obtain frequency signals of different frequency bands, and convert the frequency signals into spectrum, and at the same time cut the spectrum into multiple spectrum segments according to a specific time, and transmit the multiple spectrum segments to the model for analysis in chronological order;
[0010] S220. Add specific time so that the characteristics of multiple frequency spectra are increased, and then transmit the increased frequency spectra to the model in chronological order for analysis, thereby obtaining analysis and diagnosis data of the model.
[0011] As a further solution of the present invention: the diagnostic result data is obtained by a formula, and the specific formula is as follows:
[0012]
[0013] Among them, Q i represents the current patient's diagnosis result data, m represents different EEG frequency band data, and f a Indicates the model's processing operation on the spectrum, L i represents the i-th division of the spectrum feature data, L represents the spectrum feature data, and then the spectrum feature diagnosis result data of the model with different division times is obtained.
[0014] As a further solution of the present invention: by analyzing the energy distribution and peak frequency of the patient's diagnostic data multiple times, the model obtains the patient's current brain wave characteristic data and quickly analyzes the current patient's brain wave division.
[0015] As a further solution of the present invention: the different frequencies of the patient's brain waves are measured and integrated, and i∈(1, 2, ..., n), and the characteristics of different frequencies are diagnosed and analyzed multiple times through the model, using the formula The diagnostic result data of the total number of divisions of the brain wave frequency signal is obtained as W t ;
[0016] Among them, n represents the maximum number of divisions of the patient's brain waves, and then the total number of analysis features is imported into the model.
[0017] As a further solution of the present invention: the model retrieves historical feature data and template feature data of the current patient, and preprocesses the spectrum feature data of this measurement according to the feature data and the historical feature data of the current patient.
[0018] As a further solution of the present invention: the threshold value W is preset in the model 33% , W 66% and W 100% , and W 33% , W 66% and W 100%Respectively represent the feature analysis trends that appear in the features, classify the feature times due to individual differences, and W t Respectively in W 33% , W 66% and W 100% Matching is performed so that the model can save the error due to the patient characteristics analysis, and then W t W 33% , W 66% and W 100% The diagnostic result data is classified into W t Match the characteristic data analyzed by the model, and respectively t Match the different diagnostic result data with the data occupied by the corresponding feature data in the model;
[0019] When W t ≤W 33% When W t Delete the diagnostic result data in the
[0020] When W 33% <W t ≤W 66% When W t Export the diagnostic result data in
[0021] When W 66% <W t <W 100% When W t Export and store the diagnostic result data;
[0022] And convert the stored data and exported data into visual list data.
[0023] As a further solution of the present invention: a temporary storage area is established through the model, and the temporary storage area is used to temporarily store the test data of the same patient, and the historical feature data is retrieved by the model to match the current feature data, and the diagnosis result data of the patient's brain wave characteristics by the model is obtained through a formula. The specific formula is as follows:
[0024] S e =γ1·W t66% +γ2·W t100%
[0025] Among them, S e represents the patient's diagnostic output, γ1 and γ2 represent W t66% and W t100% The corresponding weight coefficient, W t66% Indicates that W 33% <W t ≤W 66%The diagnostic result data between t100% Indicates that W 66% <W t <W 100% The diagnostic result data between them is then converted into visual list data for export.
[0026] By adopting the above technical solution, compared with the prior art, the beneficial effects of the present invention are:
[0027] 1. The present invention divides the frequency bands of δ, θ, α, β and γ, and processes the features in the divided multiple frequency bands, so that the model analyzes the frequency bands of different divisions, which is convenient for the model to analyze the frequency bands of δ, θ, α, β and γ more accurately and mine the data in the frequency spectrum features. The frequency spectrum of the same time period is obtained by multiple diagnostic analysis models, so as to improve the accuracy of the diagnostic data analyzed by the model;
[0028] 2. The present invention uses the diagnostic result data W t The diagnostic result data of each division is integrated and transmitted to the model. The patient's historical feature data in the model is combined with the diagnostic data for analysis to determine the patient's brain improvement status and enrich the patient's feature data. At the same time, the rich EEG feature data is combined with the template feature data, so that the feature data of the EEG analyzed by the current model is more consistent with the patient himself, reducing individual differences and improving the accuracy of the diagnostic results to a certain extent.
[0029] 3. The present invention presets the threshold W in the model 33% , W 66% and W 100% To W t Processing, and then W t Different diagnostic result data can be deleted, stored and exported, and the exported data can be generated into corresponding visual list data, so that medical staff can quickly check the patient's diagnostic results through the visual list data, so that the analytical diagnosis method can deeply explore the patient's brain problems and improve the detection efficiency of the patient's brain. BRIEF DESCRIPTION OF THE DRAWINGS
[0030] Figure 1 A schematic diagram of the steps of the analysis and diagnosis method in an embodiment of the present invention;
[0031] Figure 2 Schematic diagram of the steps of the frequency band division method in an embodiment of the present invention. DETAILED DESCRIPTION
[0032] The specific embodiments of the present invention will be further described below in conjunction with the accompanying drawings. It should be noted that the description of these embodiments is used to help understand the present invention, but does not constitute a limitation of the present invention.
[0033] In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.
[0034] Embodiment 1
[0035] A brain wave analysis and diagnosis method based on brain-computer interface, which installs the brain-computer interface on the patient's head and collects brain wave data from the head. The brain wave analysis and diagnosis method based on brain-computer interface is gradually emerging. With the rapid development of science and technology, our exploration of the brain is deepening. Brain waves, as the electrophysiological manifestation of brain activity, contain rich information. Traditional medical diagnostic methods have limitations in the detection of certain neurological diseases, and the emergence of brain-computer interface technology has brought new opportunities for brain wave analysis. By directly collecting the electrical signals generated by the brain, we can obtain the activity state of the brain more real-time and accurately. Early brain wave research mainly relied on simple electroencephalogram equipment, but these devices have deficiencies in signal resolution and acquisition speed. With the advancement of sensor technology, signal processing algorithms and artificial intelligence, brain-computer interfaces can capture more subtle and complex brain wave changes. This technology not only has potential in disease diagnosis, such as epilepsy, Parkinson's disease, Alzheimer's disease, etc., but also can provide a new perspective for the evaluation of mental illness, such as depression, schizophrenia, etc.
[0036] However, the existing EEG analysis and diagnosis methods may affect the accuracy of the analysis results due to the large individual differences and vague analysis of EEG characteristics.
[0037] Therefore, in order to effectively solve the above problems, the present application proposes a method for brain wave analysis and diagnosis based on brain-computer interface, as shown in the accompanying drawings of the specification. Figure 1-2 As shown, the following steps are included:
[0038] S100. Obtaining the pre-processed δ, θ, α, β and γ band brain wave frequency signals, the time of brain wave frequency monitoring and the patient's personal information;
[0039] S200. Divide different frequency bands into multiple equal-distance divisions, extract feature data for each division, and transmit the feature data for each division to the model for comparison and analysis;
[0040] The specific process of frequency band division is as follows:
[0041] S210. Obtain frequency signals of different frequency bands, and convert the frequency signals into spectrum, and at the same time cut the spectrum into multiple spectrum segments according to a specific time, and transmit the multiple spectrum segments to the model for analysis in chronological order;
[0042] S220. Increasing the specific time so that the characteristics of multiple frequency spectra are increased, and then transmitting the increased frequency spectra to the model in chronological order for analysis, thereby obtaining analysis and diagnosis data of the model;
[0043] The diagnostic result data is obtained through the formula, the specific formula is as follows:
[0044]
[0045] Among them, Q i represents the current patient's diagnosis result data, m represents different EEG frequency band data, and f a Indicates the model's processing operation on the spectrum, L i represents the i-th division of the spectrum feature data, L represents the spectrum feature data, and then the spectrum feature diagnosis result data of the model for different division times is obtained;
[0046] By analyzing the energy distribution and peak frequency of the patient's diagnostic data multiple times, the model obtains the patient's current brain wave characteristic data and quickly analyzes the patient's current brain wave division.
[0047] Specific work flow: The brain wave acquisition device is installed on the patient's head to collect brain wave data from the patient's head, and the brain wave data is preprocessed. The preprocessed brain wave data is processed according to different δ, θ, α, β and γ frequency bands. The frequency bands are split and the feature data after splitting are analyzed and compared, so that the features in the frequency bands are matched multiple times.
[0048] Furthermore, by dividing the frequency bands of δ, θ, α, β and γ and processing the features in the divided multiple spectrums, the model can analyze the differently divided frequency bands, which makes it easier for the model to perform more accurate analysis on the frequency bands of δ, θ, α, β and γ and mine data from the spectrum features. By using multiple diagnostic analysis models to obtain the spectrum of the same time period, the accuracy of the diagnostic data analyzed by the model can be improved.
[0049] Embodiment 2
[0050] Based on the first embodiment, as shown in Figures 1-2 of the specification, the different frequencies of the patient's brain waves are integrated, and i∈(1, 2, ..., n), and the characteristics of different frequencies are diagnosed and analyzed multiple times through the model, using the formula The diagnostic result data of the total number of divisions of the brain wave frequency signal is obtained as W t ;
[0051] Among them, n represents the maximum number of divisions of the patient's EEG waves, and then the total number of analyzed features is imported into the model;
[0052] The model retrieves the historical feature data of the current patient and the template feature data, and preprocesses the spectrum feature data of this measurement according to the feature data and the historical feature data of the current patient.
[0053] Specific workflow: Integrate the result data of the model analysis and diagnosis, then transfer the integrated diagnosis result data to the model, and use the model to connect the personal data of historical patients and the template feature data for combination, so that the model can process the diagnosis result data of the current detected δ, θ, α, β and γ frequency bands, and eliminate the individual difference data in the δ, θ, α, β and γ frequency bands.
[0054] Furthermore, through the diagnosis result data W t The diagnostic result data of each division is integrated and transmitted to the model. The patient's historical feature data in the model is combined with the diagnostic data for analysis to determine the patient's brain improvement status and enrich the patient's feature data. At the same time, the rich EEG feature data is combined with the template feature data, so that the feature data of the EEG analyzed by the current model is more consistent with the patient himself, which reduces individual differences and improves the accuracy of the diagnostic results to a certain extent.
[0055] Embodiment 3
[0056] Based on the second embodiment, as shown in Figures 1-2 of the specification, the threshold W is preset in the model. 33% , W 66% and W 100% , and W 33% , W 66% and W 100% Respectively represent the feature analysis trends that appear in the features, classify the feature times due to individual differences, and W t Respectively in W 33% , W 66% and W 100% Matching is performed so that the model can save the error due to the patient characteristics analysis, and then W t W 33% , W 66% and W 100% The diagnostic result data is classified into W t Match the characteristic data analyzed by the model, and respectively tMatch the different diagnostic result data with the data occupied by the corresponding feature data in the model;
[0057] When W t ≤W 33% When W t Delete the diagnostic result data in the
[0058] When W 33% <W t ≤W 66% When W t Export the diagnostic result data in
[0059] When W 66% <W t <W 100% When W t Export and store the diagnostic result data;
[0060] And convert the stored data and exported data into visual list data;
[0061] A temporary storage area is established through the model, and the temporary storage area is used to temporarily store the test data of the same patient, and the historical feature data is retrieved by the model to match the current feature data, and the model's diagnosis result data on the patient's brain wave characteristics is obtained through the formula. The specific formula is as follows:
[0062] S e =γ1·W t66% +γ2·W t100%
[0063] Among them, S e represents the patient's diagnostic output, γ1 and γ2 represent W t66% and W t100% The corresponding weight coefficient, W t66% Indicates that W 33% <W t ≤W 66% The diagnostic result data between t100% Indicates that W 66% <W t <W 100% The diagnostic result data between them is then converted into visual list data for export.
[0064] Specific workflow: Through the model, W t Process and remove W tThe error data of the diagnosis results are analyzed, and the diagnosis data of different stages are processed by weight coefficients. The diagnosis result data are arranged according to their proportions. Then, the arranged diagnosis result data are generated into a corresponding visualization list, and the visualization list data is exported.
[0065] Furthermore, by presetting the threshold W in the model 33% , W 66% and W 100% To W t Processing, and then W t Different diagnostic result data can be deleted, stored and exported, and the exported data can be generated into corresponding visual list data, so that medical staff can quickly check the patient's diagnostic results through the visual list data, so that the analytical diagnosis method can deeply explore the patient's brain problems and improve the detection efficiency of the patient's brain.
[0066] The above shows and describes the basic principles, main features and advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.
Claims
1. A brain wave analysis and diagnosis method based on brain-computer interface, It is characterized in that The following steps are involved: S100. Obtaining the pre-processed δ, θ, α, β and γ band brain wave frequency signals, the time of brain wave frequency monitoring and the patient's personal information; S200. Divide different frequency bands into multiple equal-distance divisions, extract feature data for each division, and transmit the feature data for each division to the model for comparison and analysis.
2. According to claim 1, a method for analyzing and diagnosing brain waves based on a brain-computer interface, It is characterized in that The specific process of frequency band division is as follows: S210. Obtain frequency signals of different frequency bands, and convert the frequency signals into spectrum, and at the same time cut the spectrum into multiple spectrum segments according to a specific time, and transmit the multiple spectrum segments to the model for analysis in chronological order; S220. Add specific time so that the characteristics of multiple frequency spectra are increased, and then transmit the increased frequency spectra to the model in chronological order for analysis, thereby obtaining analysis and diagnosis data of the model.
3. According to claim 2, a method for analyzing and diagnosing brain waves based on a brain-computer interface, Features: The diagnostic result data is obtained through a formula, and the specific formula is as follows: Among them, Q i represents the current patient's diagnosis result data, m represents different EEG frequency band data, and f a Indicates the model's processing operation on the spectrum, L i represents the i-th division of the spectrum feature data, L represents the spectrum feature data, and then the spectrum feature diagnosis result data of the model with different division times is obtained.
4. According to claim 3, a method for analyzing and diagnosing brain waves based on a brain-computer interface, Features: By analyzing the energy distribution and peak frequency of the patient's diagnostic data multiple times, the model obtains the patient's current brain wave characteristic data and quickly analyzes the patient's current brain wave division.
5. The method for brain wave analysis and diagnosis based on brain-computer interface according to claim 4, Features: The different frequencies of the patient's brain waves are integrated, and i∈(1, 2, ..., n). The model performs multiple diagnostic analyses on the characteristics of different frequencies, using the formula The diagnostic result data of the total number of divisions of the brain wave frequency signal is obtained as W t ; Among them, n represents the maximum number of divisions of the patient's brain waves, and then the total number of analysis features is imported into the model.
6. The method for brain wave analysis and diagnosis based on brain-computer interface according to claim 5, Features: The model retrieves historical feature data and template feature data of the current patient, and preprocesses the frequency spectrum feature data of this measurement according to the feature data and the historical feature data of the current patient.
7. The method for brain wave analysis and diagnosis based on brain-computer interface according to claim 6, Features: The threshold W is preset in the model 33% , W 66% and W 100% , and W 33% , W 66% and W 100% Respectively represent the feature analysis trends that appear in the features, classify the feature times due to individual differences, and W t Respectively in W 33% , W 66% and W 100% Matching is performed so that the model can save the error due to the patient characteristics analysis, and then W t W 33% , W 66% and W 100% The diagnostic result data is classified into W t Match the characteristic data analyzed by the model, and respectively t Match the different diagnostic result data with the data occupied by the corresponding feature data in the model; When W t ≤W 33% When W t Delete the diagnostic result data in the When W 33% <W t ≤W 66% When W t Export the diagnostic result data in When W 66% <W t <W 100% When W t Export and store the diagnostic result data; And convert the stored data and exported data into visual list data.
8. The method for brain wave analysis and diagnosis based on brain-computer interface according to claim 7, Features: A temporary storage area is established through the model, and the temporary storage area is used to temporarily store the test data of the same patient, and the historical feature data is retrieved by the model to match the current feature data, and the model's diagnosis result data on the patient's brain wave characteristics is obtained through the formula. The specific formula is as follows: S e =c 1 ·W t66% +g 2 ·W t100% Among them, S e represents the patient's diagnostic output, γ 1 and γ 2 Respectively represent W t66% and W t100% The corresponding weight coefficient, W t66% Indicates that W 33% <W t ≤W 66% The diagnostic result data between t100% Indicates that W 66% <W t <W 100% The diagnostic result data between them is then converted into visual list data for export.
Citation Information
Patent Citations
Traditional Chinese medicine doctor diagnosis and treatment system based on brain waves
CN116168807A
Perioperative anesthesia depth monitoring system based on deep learning
CN116982937A
Method and device for correcting and enhancing electroencephalogram signals and computer equipment
CN118044827A
Brain wave analysis method for diagnosing depression
CN118319330A
Automatic diagnostic method which classifies signals from multiple patients pathologically or physiologically based on surgical or treatment outcome and system thereof
KR102043236B1
