A method, system and medium for analyzing electroencephalogram signals for cognitive function detection

Abstract

The present invention provides an electroencephalogram (EEG) signal analysis method, system and medium for cognitive function detection, which relate to the technical field of electroencephalogram (EEG) signal analysis, and include the following steps: preprocessing collected EEG signals; extracting signal features from the preprocessed EEG signals; analyzing the feature correlation of time domain features and frequency domain features based on signal features of normal volunteers and volunteers with cognitive impairment; selecting signal features with the largest feature correlation as evaluation features, and analyzing the evaluation features; judging whether a user has cognitive dysfunction; the present invention is used to solve the problem that the existing electroencephalogram (EEG) signal analysis technology, when used for cognitive function analysis, still does not consider whether different features have the greatest impact on cognitive function when considering influencing factors, resulting in insufficient accuracy in evaluating cognitive function.

Description

Technical Field

[0001] The present invention relates to the technical field of electroencephalogram (EEG) signal analysis, and in particular to an EEG signal analysis method, system and medium for cognitive function detection. Background Art

[0002] EEG signal analysis technology refers to a method and tool used to study and interpret human brain electrical activity. It involves recording and analyzing EEG signals measured from the human scalp. EEG signals are an electrical signal that records the activity of brain neurons and are measured through electrode arrays or sensor arrays placed on the scalp.

[0003] Existing EEG signal analysis technology usually involves analyzing the time domain characteristics, frequency domain characteristics and time-frequency domain characteristics of EEG signals in the application of cognitive function detection, and the existing EEG signal analysis technology usually analyzes all of them as evaluation indicators, but different characteristics have different relevance to different functions in human brain activities. Some functions are more closely related to time domain characteristics, while some functions are more closely related to frequency domain characteristics or time-frequency domain characteristics. If they are combined for analysis blindly, the accuracy of the evaluation results will be reduced. For example, in the Chinese patent with authorization announcement number CN113925517B, a method, device and medium for identifying cognitive impairment based on EEG signals are disclosed. The scheme adopts a combination of time domain characteristics, frequency domain characteristics and synchronization characteristics for analysis. In high-precision research such as human brain activity, the more factors considered, the better, but the stronger the correlation, the better. When the existing EEG signal analysis technology is used for cognitive function analysis, it still has the problem of not considering whether different characteristics have the greatest impact on cognitive function when considering influencing factors, resulting in insufficient evaluation accuracy for cognitive function. Summary of the invention

[0004] The present invention aims to solve one of the technical problems in the prior art to at least a certain extent, by collecting EEG signal data of subjects using an electroencephalogram instrument, preprocessing the collected EEG signals, extracting time domain features and frequency domain features from the preprocessed signals, dividing the subjects into normal volunteers and cognitively impaired volunteers according to their own cognitive functions, analyzing the feature correlation of time domain features and frequency domain features based on the signal characteristics of normal volunteers and cognitively impaired volunteers, selecting the signal features with the largest feature correlation as evaluation features, analyzing the evaluation features to obtain a cognitive impairment evaluation line, collecting the user's evaluation features and marking them as features to be analyzed, and finally detecting and analyzing the features to be analyzed based on the cognitive impairment evaluation line to determine whether the user has cognitive dysfunction, so as to solve the problem that the existing EEG signal analysis technology still has the problem of not considering whether different features have the greatest impact on cognitive function when considering influencing factors when used for cognitive function analysis, resulting in insufficient accuracy in evaluating cognitive function.

[0005] In order to achieve the above object, in a first aspect, the present invention provides an electroencephalogram signal analysis method for cognitive function detection, comprising the following steps:

[0006] Using an electroencephalogram instrument to collect the EEG signal data of the subject, and preprocessing the collected EEG signal;

[0007] Extracting signal features from the preprocessed EEG signal, wherein the signal features include time domain features and frequency domain features;

[0008] The subjects were divided into normal volunteers and cognitively impaired volunteers according to their own cognitive functions, and the characteristic correlation of time domain features and frequency domain features was analyzed based on the signal characteristics of the normal volunteers and cognitively impaired volunteers;

[0009] The signal feature with the largest feature correlation is selected as the evaluation feature, and the evaluation feature is analyzed to obtain the cognitive impairment assessment line;

[0010] Collect the user's evaluation features, mark them as features to be analyzed, and perform detection and analysis on the features to be analyzed to determine whether the user has cognitive dysfunction.

[0011] Furthermore, the EEG signal data of the subject is collected by an EEG instrument, and the collected EEG signal is preprocessed, including the following sub-steps:

[0012] Use an electroencephalogram instrument to collect the subject's brain electrical signals;

[0013] The EEG signal is filtered by a bandpass filter to remove noise and obtain a preprocessed signal.

[0014] Further, extracting signal features from the preprocessed EEG signal, wherein the signal features include time domain features and frequency domain features, includes the following sub-steps:

[0015] Extracting time domain features of the preprocessed signal, wherein the time domain features include average amplitude, variance, maximum value and minimum value;

[0016] The frequency domain features of the preprocessed signal are extracted by Fourier transform, and the frequency domain features include power spectrum density and spectrum bandwidth.

[0017] Furthermore, the subjects are divided into normal volunteers and cognitive impairment volunteers according to their own cognitive functions, and the feature correlation of time domain features and frequency domain features is analyzed based on the signal features of the normal volunteers and the cognitive impairment volunteers, including the following sub-steps:

[0018] The subjects were divided into normal volunteers and cognitively impaired volunteers according to their own cognitive functions, and their signal characteristics were collected;

[0019] Based on the time domain characteristics of normal volunteers and volunteers with cognitive impairment, the correlation between time domain characteristics and cognitive function was analyzed;

[0020] The correlation between frequency domain characteristics and cognitive function was analyzed based on the frequency domain characteristics of normal volunteers and volunteers with cognitive impairment.

[0021] Furthermore, the subjects are divided into normal volunteers and cognitively impaired volunteers according to their own cognitive functions, and their signal features are collected, including the following sub-steps:

[0022] The volunteers include the first number of volunteers with cognitive impairment and volunteers with normal cognitive function;

[0023] Volunteers with normal cognitive function were labeled as normal volunteers, and volunteers with cognitive impairment were labeled as cognitive impairment volunteers;

[0024] The time domain features and frequency domain features of the EEG signals of normal volunteers are collected and marked as normal time domain and normal frequency domain respectively;

[0025] The time domain features and frequency domain features of the EEG signals of volunteers with cognitive impairment were collected and marked as abnormal time domain and abnormal frequency domain respectively.

[0026] Furthermore, analyzing the correlation between the time domain characteristics and cognitive function based on the time domain characteristics of normal volunteers and volunteers with cognitive impairment includes the following sub-steps:

[0027] Setting a time domain index calculation formula, substituting a normal time domain into the time domain index calculation formula to calculate a normal time domain index, substituting an abnormal time domain into the time domain index calculation formula to calculate an abnormal time domain index, wherein the normal time domain index and the abnormal time domain index are collectively referred to as time domain index;

[0028] The time domain indicator calculation formula is configured as: , where T is the time domain index, Aa is the average amplitude, is the variance, Amax is the maximum value, and Amin is the minimum value;

[0029] A plane rectangular coordinate system is established with the average amplitude as the X-axis and the time domain index as the Y-axis, named as the time domain analysis chart, and the time domain normal index and the time domain abnormal index are entered into the time domain analysis chart according to the average amplitude;

[0030] Mark the coordinate points corresponding to the normal time domain indicators in the time domain analysis graph as normal time domain coordinate points, and mark the coordinate points corresponding to the abnormal time domain indicators as abnormal time domain coordinate points;

[0031] Linear regression is performed on the time domain normal coordinate points and the time domain abnormal coordinate points in the time domain analysis diagram, and the two straight lines obtained are named the normal time domain straight line and the abnormal time domain straight line respectively;

[0032] The minimum value of the normal time domain straight line is marked as the normal time domain endpoint 1, and the maximum value of the normal time domain straight line is marked as the normal time domain endpoint 2; the minimum value of the abnormal time domain straight line is marked as the abnormal time domain endpoint 1, and the maximum value of the abnormal time domain straight line is marked as the abnormal time domain endpoint 2;

[0033] Connect the normal time domain endpoint 1 with the abnormal time domain endpoint 1, and connect the normal time domain endpoint 2 with the abnormal time domain endpoint 2 to obtain a trapezoid, which is named as the time domain correlation reference;

[0034] The area of ​​the temporal correlation reference is calculated and named as temporal correlation.

[0035] Furthermore, analyzing the correlation between frequency domain features and cognitive function based on the frequency domain features of normal volunteers and volunteers with cognitive impairment includes the following sub-steps:

[0036] Setting a frequency domain index calculation formula, substituting the normal frequency domain into the frequency domain index calculation formula to calculate the normal frequency domain index, substituting the abnormal frequency domain into the frequency domain index calculation formula to calculate the abnormal frequency domain index, the normal frequency domain index and the abnormal frequency domain index are collectively referred to as the frequency domain index;

[0037] The frequency domain index calculation formula is configured as: ; Where A is the frequency domain index, PSD is the power spectrum density, Smax is the maximum value of the spectrum bandwidth, and Smin is the minimum value of the spectrum bandwidth;

[0038] A plane rectangular coordinate system is established with power spectrum density as the X-axis and frequency domain index as the Y-axis, named as frequency domain analysis chart, and the normal frequency domain index and the abnormal frequency domain index are entered into the frequency domain analysis chart according to the power spectrum density;

[0039] Mark the coordinate points corresponding to the normal frequency domain indicators in the frequency domain analysis graph as normal frequency domain coordinate points, and mark the coordinate points corresponding to the abnormal frequency domain indicators as abnormal frequency domain coordinate points;

[0040] Linear regression is performed on the normal frequency domain coordinate points and the abnormal frequency domain coordinate points in the frequency domain analysis diagram, and the two straight lines obtained are named the normal frequency domain straight line and the abnormal frequency domain straight line respectively;

[0041] The minimum value of the normal frequency domain straight line is marked as the normal frequency domain endpoint 1, and the maximum value of the normal frequency domain straight line is marked as the normal frequency domain endpoint 2; the minimum value of the abnormal frequency domain straight line is marked as the abnormal frequency domain endpoint 1, and the maximum value of the abnormal frequency domain straight line is marked as the abnormal frequency domain endpoint 2;

[0042] Connect the first end point of the normal frequency domain with the first end point of the abnormal frequency domain, and connect the second end point of the normal frequency domain with the second end point of the abnormal frequency domain to obtain a trapezoid, which is named frequency domain correlation reference;

[0043] The area of ​​the frequency domain correlation reference is calculated and named as frequency domain correlation.

[0044] Furthermore, the signal feature with the largest feature correlation is selected as the evaluation feature, and the evaluation feature is analyzed to obtain the cognitive impairment evaluation line, which includes the following sub-steps:

[0045] Compare the time domain correlation with the frequency domain correlation. If the time domain correlation is greater than or equal to the frequency domain correlation, then output the time domain correlation signal; if the time domain correlation is less than the frequency domain correlation, then output the frequency domain correlation signal.

[0046] If the time domain related signal is output, the time domain feature is used as the evaluation feature; if the frequency domain related signal is output, the frequency domain feature is used as the evaluation feature;

[0047] The assessment characteristics were analyzed to obtain the cognitive impairment assessment line.

[0048] Further, the assessment characteristics are analyzed to obtain the cognitive impairment assessment line, which includes the following sub-steps:

[0049] The time domain analysis graph or the frequency domain analysis graph corresponding to the evaluation feature is marked as an evaluation analysis graph, and the trapezoid in the evaluation analysis graph is named as an evaluation trapezoid;

[0050] Get the midpoint of the upper base of the evaluation trapezoid, mark it as the upper midpoint, get the midpoint of the lower base of the evaluation trapezoid, mark it as the lower midpoint;

[0051] Connect the upper midpoint with the lower midpoint and label the resulting straight line as the cognitive impairment assessment line;

[0052] The evaluation analysis chart is divided into two parts by the cognitive impairment evaluation line. The part close to the normal straight line is named the normal cognitive area, and the part close to the abnormal straight line is named the cognitive impairment area. The normal straight line includes a normal time domain straight line and a normal frequency domain straight line, and the abnormal straight line includes an abnormal time domain straight line and an abnormal frequency domain straight line.

[0053] Furthermore, collecting the user's evaluation features, marking them as features to be analyzed, and performing detection and analysis on the features to be analyzed to determine whether the user has cognitive dysfunction includes the following sub-steps:

[0054] Collecting user evaluation features and calculating evaluation indicators, wherein the evaluation indicators are time domain indicators or frequency domain indicators;

[0055] Based on the user's evaluation characteristics and evaluation indicators, find the corresponding coordinate point in the evaluation analysis diagram and name it as the user evaluation point;

[0056] Find out whether the user evaluation point is in the normal cognitive area or the cognitive impairment area. If the user evaluation point is in the normal cognitive area, output a normal cognitive signal; if the user evaluation point is in the cognitive impairment area, output a cognitive impairment signal;

[0057] If a normal cognitive signal is output, it indicates that the user's cognitive function is normal; if a cognitive impairment signal is output, it indicates that the user's cognitive function is impaired.

[0058] In a second aspect, the present invention provides an electroencephalogram signal analysis system for cognitive function detection, comprising a data preprocessing module, a feature extraction module, a correlation analysis module, a cognitive analysis module and a cognitive evaluation module; the data preprocessing module, the feature extraction module, the correlation analysis module and the cognitive evaluation module are respectively connected to the cognitive analysis module data;

[0059] The data preprocessing module is used to collect the EEG signal data of the subject using an EEG instrument and preprocess the collected EEG signal;

[0060] The feature extraction module is used to extract signal features from the preprocessed EEG signal, and the signal features include time domain features and frequency domain features;

[0061] The association analysis module is used to divide the subjects into normal volunteers and cognitive impairment volunteers according to their own cognitive functions, and analyze the feature correlation of time domain features and frequency domain features based on the signal features of the normal volunteers and cognitive impairment volunteers;

[0062] The cognitive analysis module is used to select the signal feature with the greatest feature correlation as the evaluation feature, establish a cognitive function detection model, analyze the evaluation feature, and obtain a cognitive impairment evaluation line;

[0063] The cognitive assessment module is used to collect assessment features of the user, mark them as features to be analyzed, detect and analyze the features to be analyzed, and determine whether the user has cognitive dysfunction.

[0064] In a third aspect, the present application provides a storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps in the above method are performed.

[0065] Beneficial effects of the present invention: The present invention pre-processes the collected EEG signals, extracts time domain features and frequency domain features from the pre-processed signals, divides the subjects into normal volunteers and cognitively impaired volunteers according to their own cognitive functions, and then analyzes the time domain correlation and frequency domain correlation based on the signal features of the normal volunteers and the cognitively impaired volunteers. The advantages are that the time domain correlation reflects the difference in time domain features of the EEG signals of the normal volunteers and the cognitively impaired volunteers, and the frequency domain correlation reflects the difference in frequency domain features of the EEG signals of the normal volunteers and the cognitively impaired volunteers. The greater the difference, the more reference value it has, thereby improving the accuracy and rationality of the cognitive function analysis of the EEG signals.

[0066] The present invention selects the signal feature with the largest feature correlation as the evaluation feature, analyzes the evaluation feature, obtains the cognitive impairment evaluation line, divides the cognitive normal area and the cognitive impairment area based on the cognitive impairment evaluation line, collects the user's evaluation features, marks them as features to be analyzed, and finally detects and analyzes the features to be analyzed based on the cognitive impairment evaluation line to determine whether the user has cognitive dysfunction. The advantage is that the analysis of EEG signals is more precise rather than quantitative, and the results obtained by using more parameters for analysis are actually not as accurate as the results obtained by having a higher correlation between the parameters and the analysis target. Therefore, selecting signal features with higher correlation for analysis can more accurately judge the user's cognitive function, thereby improving the accuracy and effectiveness of EEG signal cognitive function analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:

[0068] Figure 1 is a functional block diagram of the system of the present invention;

[0069] Figure 2 is a schematic diagram of a preprocessed signal of the present invention;

[0070] Figure 3is a schematic diagram of a frequency domain waveform of the present invention;

[0071] Figure 4 It is a time domain analysis diagram of the present invention;

[0072] Figure 5 It is a schematic diagram of a normal time domain straight line and an abnormal time domain straight line of the present invention;

[0073] Figure 6 It is a frequency domain analysis diagram of the present invention;

[0074] Figure 7 It is a schematic diagram of a normal frequency domain straight line and an abnormal frequency domain straight line of the present invention;

[0075] Figure 8 is a schematic diagram of a cognitive impairment assessment line of the present invention;

[0076] Fig. 9 The figure is a flow chart of the steps of the method of the present invention. DETAILED DESCRIPTION

[0077] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0078] Example 1, please refer to Figure 1 As shown, the present invention provides an electroencephalogram signal analysis system for cognitive function detection, comprising a data preprocessing module, a feature extraction module, a correlation analysis module, a cognitive analysis module and a cognitive evaluation module; the data preprocessing module, the feature extraction module, the correlation analysis module and the cognitive evaluation module are respectively connected to the cognitive analysis module data;

[0079] The data preprocessing module is used to collect the EEG signal data of the subject using an EEG instrument and preprocess the collected EEG signal;

[0080] The data preprocessing module is configured with a data preprocessing strategy, which includes:

[0081] Use an electroencephalogram instrument to collect the subject's brain electrical signals;

[0082] See also Figure 2 As shown, the EEG signal is filtered by a bandpass filter to remove noise and obtain a preprocessed signal;

[0083] In practical applications, the existing electroencephalogram instrument is used to collect the EEG signals of the subjects, who are volunteers, and the EEG signals are filtered by the existing bandpass filtering technology to obtain the preprocessed signals such as Figure 2 shown.

[0084] The feature extraction module is used to extract signal features from the preprocessed EEG signal, and the signal features include time domain features and frequency domain features;

[0085] The feature extraction module is configured with feature extraction strategies, which include:

[0086] Extracting the time domain features of the preprocessed signal, the time domain features include average amplitude, variance, maximum value and minimum value;

[0087] See also Figure 3 As shown, the frequency domain features of the preprocessed signal are extracted by Fourier transform, and the frequency domain features include power spectrum density and spectrum bandwidth;

[0088] In practical applications, the time domain features of the preprocessed signal are extracted, and the average amplitude is 23μV and the variance is 324μV. 2 , the maximum value is 99μV, the minimum value is -98μV, and the preprocessed signal is converted from a time domain waveform to a frequency domain waveform through Fourier transform. Figure 3 The frequency domain waveform shown in the figure has a power spectrum density of 19μV. 2 / Hz, the spectrum bandwidth is [10Hz,28Hz].

[0089] The correlation analysis module is used to divide the subjects into normal volunteers and cognitive impairment volunteers according to their own cognitive functions, and analyze the feature correlation of time domain features and frequency domain features based on the signal characteristics of the normal volunteers and cognitive impairment volunteers; it includes a volunteer information collection unit, a time domain correlation analysis unit and a frequency domain correlation analysis unit;

[0090] The volunteer information collection unit is used to divide the subjects into normal volunteers and cognitively impaired volunteers according to their own cognitive functions, and collect their signal characteristics;

[0091] The volunteer information collection unit is configured with a volunteer information collection strategy, which includes:

[0092] Recruiting volunteers from the society, including a large number of volunteers with cognitive impairment and volunteers with normal cognitive function;

[0093] Volunteers with normal cognitive function were labeled as normal volunteers, and volunteers with cognitive impairment were labeled as cognitive impairment volunteers;

[0094] The time domain features and frequency domain features of the EEG signals of normal volunteers are collected and marked as normal time domain and normal frequency domain respectively;

[0095] The time domain features and frequency domain features of the EEG signals of volunteers with cognitive impairment are collected and marked as abnormal time domain and abnormal frequency domain respectively;

[0096] In practical applications, the first number is set only to ensure that enough data is collected to provide data support for subsequent analysis. In this embodiment, the first number is set to 100.

[0097] The time domain correlation analysis unit is used to analyze the correlation between the time domain characteristics and cognitive function based on the time domain characteristics of normal volunteers and volunteers with cognitive impairment;

[0098] The time domain correlation analysis unit is configured with a time domain correlation analysis strategy, which includes:

[0099] Set a time domain index calculation formula, substitute the normal time domain into the time domain index calculation formula to calculate the time domain normal index, substitute the abnormal time domain into the time domain index calculation formula to calculate the time domain abnormal index, the time domain normal index and the time domain abnormal index are collectively referred to as the time domain index;

[0100] The time domain indicator calculation formula is configured as: , where T is the time domain index, Aa is the average amplitude, is the variance, Amax is the maximum value, Amin is the minimum value, and both the maximum and minimum values ​​are amplitudes;

[0101] In practical applications, the time domain features extracted in this embodiment are from normal volunteers, with an average amplitude of 23 μV and a variance of 324 μV. 2 , the maximum value is 99μV, the minimum value is -98μV, and the time domain normal index is 985. In this embodiment, only this time domain normal index is calculated to demonstrate the specific analysis process. The other time domain indicators will be directly displayed in the form of a time domain analysis diagram;

[0102] See also Figure 4 As shown in the figure, a plane rectangular coordinate system is established with the average amplitude as the X-axis and the time domain index as the Y-axis, named as the time domain analysis diagram, and the time domain normal index and the time domain abnormal index are entered into the time domain analysis diagram according to the average amplitude;

[0103] Mark the coordinate points corresponding to the normal time domain indicators in the time domain analysis graph as normal time domain coordinate points, and mark the coordinate points corresponding to the abnormal time domain indicators as abnormal time domain coordinate points;

[0104] See also Figure 5As shown, linear regression is performed on the time domain normal coordinate points and the time domain abnormal coordinate points in the time domain analysis diagram, and the two straight lines obtained are named the normal time domain straight line and the abnormal time domain straight line respectively;

[0105] The minimum value of the normal time domain straight line is marked as the normal time domain endpoint 1, and the maximum value of the normal time domain straight line is marked as the normal time domain endpoint 2; the minimum value of the abnormal time domain straight line is marked as the abnormal time domain endpoint 1, and the maximum value of the abnormal time domain straight line is marked as the abnormal time domain endpoint 2;

[0106] Connect the normal time domain endpoint 1 with the abnormal time domain endpoint 1, and connect the normal time domain endpoint 2 with the abnormal time domain endpoint 2 to obtain a trapezoid, which is named as the time domain correlation reference;

[0107] Calculate the area of ​​the temporal correlation reference, named temporal correlation;

[0108] In practical applications, the time domain analysis diagram is constructed as follows Figure 4 As shown, linear regression is performed to obtain the normal time domain straight line and the abnormal time domain straight line as shown in Figure 5 As shown in the figure, the distribution of each endpoint is also as shown in Figure 5 As shown, the time domain correlation is 42656 by calculating the area of ​​the time domain correlation reference;

[0109] The frequency domain correlation analysis unit is used to analyze the correlation between frequency domain features and cognitive function based on the frequency domain features of normal volunteers and volunteers with cognitive impairment;

[0110] The frequency domain correlation analysis unit is configured with a frequency domain correlation analysis strategy, which includes:

[0111] Set the frequency domain index calculation formula, substitute the normal frequency domain into the frequency domain index calculation formula to calculate the normal frequency domain index, substitute the abnormal frequency domain into the frequency domain index calculation formula to calculate the abnormal frequency domain index, the normal frequency domain index and the abnormal frequency domain index are collectively referred to as the frequency domain index;

[0112] The frequency domain indicator calculation formula is configured as: ; Where A is the frequency domain index, PSD is the power spectrum density, Smax is the maximum value of the spectrum bandwidth, and Smin is the minimum value of the spectrum bandwidth;

[0113] In practical applications, the power spectral density is 19μV 2 / Hz, the spectrum bandwidth is [10Hz, 28Hz], that is, PSD=19, Smax=28, Smin=10, and the calculated frequency domain normal index is 342. In this embodiment, only this frequency domain normal index is calculated to demonstrate the specific analysis process, and the other frequency domain indexes will be directly displayed in the form of frequency domain analysis diagrams;

[0114] See also Figure 6 As shown, a plane rectangular coordinate system is established with power spectrum density as the X-axis and frequency domain index as the Y-axis, named as frequency domain analysis diagram, and the normal frequency domain index and the abnormal frequency domain index are entered into the frequency domain analysis diagram according to the power spectrum density;

[0115] Mark the coordinate points corresponding to the normal frequency domain indicators in the frequency domain analysis graph as normal frequency domain coordinate points, and mark the coordinate points corresponding to the abnormal frequency domain indicators as abnormal frequency domain coordinate points;

[0116] See also Figure 7 As shown, linear regression is performed on the normal frequency domain coordinate points and the abnormal frequency domain coordinate points in the frequency domain analysis diagram, and the two straight lines obtained are named the normal frequency domain straight line and the abnormal frequency domain straight line respectively;

[0117] The minimum value of the normal frequency domain straight line is marked as the normal frequency domain endpoint 1, and the maximum value of the normal frequency domain straight line is marked as the normal frequency domain endpoint 2; the minimum value of the abnormal frequency domain straight line is marked as the abnormal frequency domain endpoint 1, and the maximum value of the abnormal frequency domain straight line is marked as the abnormal frequency domain endpoint 2;

[0118] Connect the first end point of the normal frequency domain with the first end point of the abnormal frequency domain, and connect the second end point of the normal frequency domain with the second end point of the abnormal frequency domain to obtain a trapezoid, which is named frequency domain correlation reference;

[0119] Calculate the area of ​​the frequency domain correlation reference, named frequency domain correlation;

[0120] In practical applications, the frequency domain analysis diagram is constructed as follows Figure 6 As shown, linear regression is performed on it to obtain the normal frequency domain straight line and the abnormal frequency domain straight line as shown in Figure 7 As shown in the figure, the distribution of each endpoint is also as shown in Figure 7 As shown, the frequency domain correlation is 11448.75 obtained by calculating the area of ​​the frequency domain correlation reference.

[0121] The cognitive analysis module is used to select the signal feature with the largest feature correlation as the evaluation feature, establish a cognitive function detection model, analyze the evaluation feature, and obtain the cognitive impairment evaluation line; the cognitive analysis module includes an evaluation feature determination unit and an evaluation line analysis unit;

[0122] The evaluation feature determination unit is configured with an evaluation feature determination strategy, and the evaluation feature determination strategy includes:

[0123] Compare the time domain correlation with the frequency domain correlation. If the time domain correlation is greater than or equal to the frequency domain correlation, then output the time domain correlation signal; if the time domain correlation is less than the frequency domain correlation, then output the frequency domain correlation signal.

[0124] If the time domain related signal is output, the time domain feature is used as the evaluation feature; if the frequency domain related signal is output, the frequency domain feature is used as the evaluation feature;

[0125] The assessment characteristics were analyzed to obtain the cognitive impairment assessment line;

[0126] In practical applications, the time domain correlation is found to be greater than the frequency domain correlation through comparison, and a time domain correlation signal is output, and the time domain feature is used as an evaluation feature.

[0127] The evaluation line analysis unit is configured with an evaluation line analysis strategy, which includes:

[0128] The time domain analysis graph or the frequency domain analysis graph corresponding to the evaluation feature is marked as an evaluation analysis graph, and the trapezoid in the evaluation analysis graph is named as an evaluation trapezoid;

[0129] Get the midpoint of the upper base of the evaluation trapezoid, mark it as the upper midpoint, get the midpoint of the lower base of the evaluation trapezoid, mark it as the lower midpoint;

[0130] See also Figure 8 As shown, connect the upper midpoint with the lower midpoint and mark the resulting straight line as the cognitive impairment assessment line;

[0131] The evaluation analysis chart is divided into two parts by the cognitive impairment evaluation line. The part close to the normal straight line is named the cognitive normal area, and the part close to the abnormal straight line is named the cognitive impairment area. The normal straight line includes the normal time domain straight line and the normal frequency domain straight line, and the abnormal straight line includes the abnormal time domain straight line and the abnormal frequency domain straight line.

[0132] In practical applications, the time domain analysis diagram of the last step is marked as the evaluation analysis diagram. Figure 5 Marked as the evaluation analysis diagram, the cognitive impairment evaluation line is obtained by connecting Figure 8 As shown, Figure 8 Among them, those above the cognitive impairment assessment line are all in the cognitive abnormal area, and those below the cognitive impairment assessment line are all in the cognitive normal area.

[0133] The cognitive assessment module is used to collect the user's assessment features, mark them as features to be analyzed, detect and analyze the features to be analyzed, and determine whether the user has cognitive dysfunction;

[0134] The cognitive assessment module is configured with cognitive assessment strategies, which include:

[0135] Collecting user evaluation features and calculating evaluation indicators, where the evaluation indicators are time domain indicators or frequency domain indicators;

[0136] Based on the user's evaluation characteristics and evaluation indicators, find the corresponding coordinate point in the evaluation analysis diagram and name it as the user evaluation point;

[0137] Find out whether the user evaluation point is in the normal cognitive area or the cognitive impairment area. If the user evaluation point is in the normal cognitive area, output a normal cognitive signal; if the user evaluation point is in the cognitive impairment area, output a cognitive impairment signal;

[0138] If the cognitive normal signal is output, the user is notified, indicating that the user's cognitive function is normal; if the cognitive impairment signal is output, the user is notified, indicating that the user's cognitive function is impaired;

[0139] In actual applications, the user's evaluation indicators are collected: the average amplitude is 32μV and the variance is 523μV 2 , the maximum value is 98μV, the minimum value is -96μV, and the calculated evaluation index is 1773.4. The calculation result retains one decimal place, that is, the user evaluation point is (32,1773.4), which is in the cognitive impairment area. A cognitive impairment signal is output to notify the user, indicating that the user's cognitive function is impaired; if the user evaluation point is on the cognitive impairment evaluation line, the user's evaluation characteristics are re-acquired and evaluated again.

[0140] Example 2, please refer to Fig. 9 As shown, a method for analyzing EEG signals for cognitive function detection comprises the following steps:

[0141] Step S1, using an electroencephalogram instrument to collect the subject's electroencephalogram signal data, and preprocessing the collected electroencephalogram signal; Step S1 includes the following sub-steps:

[0142] Step S101, using an electroencephalogram instrument to collect the subject's electroencephalogram signal;

[0143] Step S102, filtering the EEG signal with a bandpass filter to remove noise and obtain a preprocessed signal;

[0144] Step S2, extracting signal features from the preprocessed EEG signal, the signal features including time domain features and frequency domain features; Step S2 includes the following sub-steps:

[0145] Step S201, extracting time domain features of the preprocessed signal, the time domain features including average amplitude, variance, maximum value and minimum value;

[0146] Step S202, extracting frequency domain features of the preprocessed signal by Fourier transform, where the frequency domain features include power spectrum density and spectrum bandwidth;

[0147] Step S3, dividing the subjects into normal volunteers and cognitive impairment volunteers according to their own cognitive functions, and analyzing the feature correlation of time domain features and frequency domain features based on the signal features of the normal volunteers and the cognitive impairment volunteers; Step S3 includes the following sub-steps:

[0148] Step S301, dividing the subjects into normal volunteers and cognitively impaired volunteers according to their own cognitive functions, and collecting their signal features;

[0149] Step S301 includes the following sub-steps:

[0150] Step S3011, recruiting volunteers from the society, wherein the volunteers include a first number of volunteers with cognitive impairment and volunteers with normal cognitive function;

[0151] Step S3012, marking volunteers with normal cognitive function as normal volunteers, and marking volunteers with cognitive impairment as cognitive impairment volunteers;

[0152] Step S3013, collecting the time domain features and frequency domain features of the EEG signals of normal volunteers, and marking them as normal time domain and normal frequency domain respectively;

[0153] Step S3014, collecting the time domain features and frequency domain features of the EEG signals of the cognitively impaired volunteers, and marking them as abnormal time domain and abnormal frequency domain respectively;

[0154] Step S302, analyzing the correlation between the time domain characteristics and cognitive function based on the time domain characteristics of normal volunteers and volunteers with cognitive impairment;

[0155] Step S302 includes the following sub-steps:

[0156] Step S3021, setting a time domain index calculation formula, substituting the normal time domain into the time domain index calculation formula to calculate the time domain normal index, substituting the abnormal time domain into the time domain index calculation formula to calculate the time domain abnormal index, the time domain normal index and the time domain abnormal index are collectively referred to as the time domain index;

[0157] The time domain indicator calculation formula is configured as: , where T is the time domain index, Aa is the average amplitude, is the variance, Amax is the maximum value, and Amin is the minimum value;

[0158] Step S3022, establish a plane rectangular coordinate system with the average amplitude as the X-axis and the time domain index as the Y-axis, name it as a time domain analysis chart, and enter the time domain normal index and the time domain abnormal index into the time domain analysis chart according to the average amplitude;

[0159] Step S3023, marking the coordinate points corresponding to the normal time domain indicators in the time domain analysis graph as normal time domain coordinate points, and marking the coordinate points corresponding to the abnormal time domain indicators as abnormal time domain coordinate points;

[0160] Step S3024, performing linear regression on the time domain normal coordinate points and the time domain abnormal coordinate points in the time domain analysis graph, respectively, and obtaining two straight lines, respectively named as the normal time domain straight line and the abnormal time domain straight line;

[0161] Step S3025, marking the minimum value of the normal time domain straight line as the normal time domain endpoint 1, marking the maximum value of the normal time domain straight line as the normal time domain endpoint 2; marking the minimum value of the abnormal time domain straight line as the abnormal time domain endpoint 1, marking the maximum value of the abnormal time domain straight line as the abnormal time domain endpoint 2;

[0162] Step S3026, connecting the normal time domain endpoint 1 with the abnormal time domain endpoint 1, and connecting the normal time domain endpoint 2 with the abnormal time domain endpoint 2 to obtain a trapezoid, which is named as a time domain correlation reference;

[0163] Step S3027, calculating the area of ​​the time domain correlation reference, named as the time domain correlation;

[0164] Step S303, analyzing the correlation between frequency domain features and cognitive function based on the frequency domain features of normal volunteers and volunteers with cognitive impairment;

[0165] Step S303 includes the following sub-steps:

[0166] Step S3031, setting a frequency domain index calculation formula, substituting the normal frequency domain into the frequency domain index calculation formula to calculate the normal frequency domain index, substituting the abnormal frequency domain into the frequency domain index calculation formula to calculate the abnormal frequency domain index, the normal frequency domain index and the abnormal frequency domain index are collectively referred to as the frequency domain index;

[0167] The frequency domain indicator calculation formula is configured as: ; Where A is the frequency domain index, PSD is the power spectrum density, Smax is the maximum value of the spectrum bandwidth, and Smin is the minimum value of the spectrum bandwidth;

[0168] Step S3032, establishing a plane rectangular coordinate system with power spectrum density as the X-axis and frequency domain index as the Y-axis, named as frequency domain analysis graph, and entering the frequency domain normal index and the frequency domain abnormal index into the frequency domain analysis graph according to the power spectrum density;

[0169] Step S3033, marking the coordinate points corresponding to the normal frequency domain index in the frequency domain analysis graph as normal frequency domain coordinate points, and marking the coordinate points corresponding to the abnormal frequency domain index as abnormal frequency domain coordinate points;

[0170] Step S3034, performing linear regression on the normal frequency domain coordinate points and the abnormal frequency domain coordinate points in the frequency domain analysis graph, and obtaining two straight lines, which are named as the normal frequency domain straight line and the abnormal frequency domain straight line respectively;

[0171] Step S3035, marking the minimum value of the normal frequency domain straight line as the normal frequency domain endpoint 1, marking the maximum value of the normal frequency domain straight line as the normal frequency domain endpoint 2; marking the minimum value of the abnormal frequency domain straight line as the abnormal frequency domain endpoint 1, marking the maximum value of the abnormal frequency domain straight line as the abnormal frequency domain endpoint 2;

[0172] Step S3036, connecting the normal frequency domain endpoint 1 with the abnormal frequency domain endpoint 1, and connecting the normal frequency domain endpoint 2 with the abnormal frequency domain endpoint 2 to obtain a trapezoid, which is named frequency domain correlation reference;

[0173] Step S3037, calculating the area of ​​the frequency domain correlation reference, named as frequency domain correlation;

[0174] Step S4, selecting the signal feature with the largest feature correlation as the evaluation feature, analyzing the evaluation feature, and obtaining the cognitive impairment evaluation line; Step S4 includes the following sub-steps:

[0175] Step S401, comparing the time domain correlation with the frequency domain correlation, if the time domain correlation is greater than or equal to the frequency domain correlation, outputting a time domain correlation signal; if the time domain correlation is less than the frequency domain correlation, outputting a frequency domain correlation signal;

[0176] Step S402: if a time domain correlation signal is output, the time domain feature is used as the evaluation feature; if a frequency domain correlation signal is output, the frequency domain feature is used as the evaluation feature;

[0177] Step S403, analyzing the assessment features to obtain a cognitive impairment assessment line;

[0178] Step S403 includes the following sub-steps:

[0179] Step S4031, marking the time domain analysis graph or the frequency domain analysis graph corresponding to the evaluation feature as an evaluation analysis graph, and naming the trapezoid in the evaluation analysis graph as an evaluation trapezoid;

[0180] Step S4032, obtaining the midpoint of the upper base of the evaluation trapezoid, marking it as the upper midpoint, and obtaining the midpoint of the lower base of the evaluation trapezoid, marking it as the lower midpoint;

[0181] Step S4033, connecting the upper midpoint and the lower midpoint, and marking the resulting straight line as a cognitive impairment assessment line;

[0182] Step S4034, dividing the evaluation analysis chart into two parts through the cognitive impairment evaluation line, naming the part close to the normal straight line as the normal cognitive area, and the part close to the abnormal straight line as the cognitive impairment area, the normal straight line includes the normal time domain straight line and the normal frequency domain straight line, and the abnormal straight line includes the abnormal time domain straight line and the abnormal frequency domain straight line;

[0183] Step S5, collecting the user's evaluation features, marking them as features to be analyzed, detecting and analyzing the features to be analyzed, and determining whether the user has cognitive dysfunction; Step S5 includes the following sub-steps:

[0184] Step S501, collecting the user's evaluation features and calculating the evaluation index, where the evaluation index is a time domain index or a frequency domain index;

[0185] Step S502, based on the user's evaluation characteristics and evaluation indicators, find the corresponding coordinate point in the evaluation analysis diagram and name it as the user evaluation point;

[0186] Step S503, finding out whether the user evaluation point is in the normal cognitive area or the cognitive impairment area, if the user evaluation point is in the normal cognitive area, outputting a normal cognitive signal; if the user evaluation point is in the cognitive impairment area, outputting a cognitive impairment signal;

[0187] Step S504: if a normal cognition signal is output, the user is notified, indicating that the user's cognitive function is normal; if a cognitive impairment signal is output, the user is notified, indicating that the user's cognitive function is impaired.

[0188] Embodiment 3. The present application also provides a computer program product, which includes a computer program stored on a computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute an EEG signal analysis method for cognitive function detection provided by the above methods, the method including: preprocessing the collected EEG signals; extracting signal features from the preprocessed EEG signals; analyzing the feature correlation of time domain features and frequency domain features based on the signal features of normal volunteers and volunteers with cognitive impairment; selecting the signal features with the largest feature correlation as evaluation features, and analyzing the evaluation features; and judging whether the user has cognitive dysfunction.

[0189] Embodiment 4, the present application also provides a computer-readable storage medium, the present application provides a storage medium, on which a computer program is stored, and when the computer program is executed by the processor, the steps in the EEG signal analysis method for cognitive function detection are executed as above. Through the above technical scheme, when the computer program is executed by the processor, the method in any optional implementation of the above embodiment is executed to achieve the following functions: preprocessing the collected EEG signal; extracting signal features from the preprocessed EEG signal; analyzing the feature correlation of time domain features and frequency domain features based on the signal features of normal volunteers and volunteers with cognitive impairment; selecting the signal feature with the largest feature correlation as the evaluation feature, and analyzing the evaluation feature; judging whether the user has cognitive dysfunction.

[0190] Through the description of the above implementation methods, the embodiments of the present invention can be provided as methods, systems or computer program products. Based on such an understanding, the above technical solutions can be essentially or partly contributed to the prior art in the form of software products, which can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and include several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0191] In the embodiments provided in the present application, it should be understood that the disclosed system or method can be implemented in other ways. The embodiments described above are merely schematic. For example, the division of modules or units is only a logical function division. There may be other division methods in actual implementation. For example, multiple modules or units can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of systems, modules and units can be electrical, mechanical or other forms.

[0192] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for analyzing electroencephalogram signals for cognitive function detection, characterized in that: The steps include: Using an electroencephalogram instrument to collect the EEG signal data of the subject, and preprocessing the collected EEG signal; Extracting signal features from the preprocessed EEG signal, wherein the signal features include time domain features and frequency domain features; The subjects were divided into normal volunteers and cognitively impaired volunteers according to their own cognitive functions, and the characteristic correlation of time domain features and frequency domain features was analyzed based on the signal characteristics of the normal volunteers and cognitively impaired volunteers; The signal feature with the largest feature correlation is selected as the evaluation feature, and the evaluation feature is analyzed to obtain the cognitive impairment assessment line; Collect the user's evaluation features, mark them as features to be analyzed, and perform detection and analysis on the features to be analyzed to determine whether the user has cognitive dysfunction; The subjects are divided into normal volunteers and cognitive impairment volunteers according to their own cognitive functions. The feature correlation of time domain features and frequency domain features is analyzed based on the signal features of the normal volunteers and cognitive impairment volunteers, including the following sub-steps: The subjects were divided into normal volunteers and cognitively impaired volunteers according to their own cognitive functions, and their signal characteristics were collected; Based on the time domain characteristics of normal volunteers and volunteers with cognitive impairment, the correlation between time domain characteristics and cognitive function was analyzed; The correlation between frequency domain characteristics and cognitive function was analyzed based on the frequency domain characteristics of normal volunteers and volunteers with cognitive impairment; The subjects are divided into normal volunteers and cognitively impaired volunteers according to their own cognitive functions, and their signal characteristics are collected, including the following sub-steps: The volunteers include the first number of volunteers with cognitive impairment and volunteers with normal cognitive function; Volunteers with normal cognitive function were labeled as normal volunteers, and volunteers with cognitive impairment were labeled as cognitive impairment volunteers; The time domain features and frequency domain features of the EEG signals of normal volunteers are collected and marked as normal time domain and normal frequency domain respectively; The time domain features and frequency domain features of the EEG signals of volunteers with cognitive impairment are collected and marked as abnormal time domain and abnormal frequency domain respectively; The analysis of the correlation between time domain characteristics and cognitive function based on the time domain characteristics of normal volunteers and volunteers with cognitive impairment includes the following sub-steps: Setting a time domain index calculation formula, substituting a normal time domain into the time domain index calculation formula to calculate a normal time domain index, substituting an abnormal time domain into the time domain index calculation formula to calculate an abnormal time domain index, wherein the normal time domain index and the abnormal time domain index are collectively referred to as time domain index; The time domain indicator calculation formula is configured as: , where T is the time domain index, Aa is the average amplitude, is the variance, Amax is the maximum value, and Amin is the minimum value; A plane rectangular coordinate system is established with the average amplitude as the X-axis and the time domain index as the Y-axis, named as the time domain analysis chart, and the time domain normal index and the time domain abnormal index are entered into the time domain analysis chart according to the average amplitude; Mark the coordinate points corresponding to the normal time domain indicators in the time domain analysis graph as normal time domain coordinate points, and mark the coordinate points corresponding to the abnormal time domain indicators as abnormal time domain coordinate points; Linear regression is performed on the time domain normal coordinate points and the time domain abnormal coordinate points in the time domain analysis diagram, and the two straight lines obtained are named the normal time domain straight line and the abnormal time domain straight line respectively; The minimum value of the normal time domain straight line is marked as the normal time domain endpoint 1, and the maximum value of the normal time domain straight line is marked as the normal time domain endpoint 2; the minimum value of the abnormal time domain straight line is marked as the abnormal time domain endpoint 1, and the maximum value of the abnormal time domain straight line is marked as the abnormal time domain endpoint 2; Connect the normal time domain endpoint 1 with the abnormal time domain endpoint 1, and connect the normal time domain endpoint 2 with the abnormal time domain endpoint 2 to obtain a trapezoid, which is named as the time domain correlation reference; Calculate the area of ​​the temporal correlation reference, named temporal correlation; The analysis of the correlation between frequency domain features and cognitive function based on the frequency domain features of normal volunteers and volunteers with cognitive impairment includes the following sub-steps: Setting a frequency domain index calculation formula, substituting the normal frequency domain into the frequency domain index calculation formula to calculate the normal frequency domain index, substituting the abnormal frequency domain into the frequency domain index calculation formula to calculate the abnormal frequency domain index, the normal frequency domain index and the abnormal frequency domain index are collectively referred to as the frequency domain index; The frequency domain index calculation formula is configured as: ; Where A is the frequency domain index, PSD is the power spectrum density, Smax is the maximum value of the spectrum bandwidth, and Smin is the minimum value of the spectrum bandwidth; A plane rectangular coordinate system is established with power spectrum density as the X-axis and frequency domain index as the Y-axis, named as frequency domain analysis chart, and the normal frequency domain index and the abnormal frequency domain index are entered into the frequency domain analysis chart according to the power spectrum density; Mark the coordinate points corresponding to the normal frequency domain indicators in the frequency domain analysis graph as normal frequency domain coordinate points, and mark the coordinate points corresponding to the abnormal frequency domain indicators as abnormal frequency domain coordinate points; Linear regression is performed on the normal frequency domain coordinate points and the abnormal frequency domain coordinate points in the frequency domain analysis diagram, and the two straight lines obtained are named the normal frequency domain straight line and the abnormal frequency domain straight line respectively; The minimum value of the normal frequency domain straight line is marked as the normal frequency domain endpoint 1, and the maximum value of the normal frequency domain straight line is marked as the normal frequency domain endpoint 2; the minimum value of the abnormal frequency domain straight line is marked as the abnormal frequency domain endpoint 1, and the maximum value of the abnormal frequency domain straight line is marked as the abnormal frequency domain endpoint 2; Connect the first end point of the normal frequency domain with the first end point of the abnormal frequency domain, and connect the second end point of the normal frequency domain with the second end point of the abnormal frequency domain to obtain a trapezoid, which is named frequency domain correlation reference; Calculate the area of ​​the frequency domain correlation reference, named frequency domain correlation; The signal feature with the largest feature correlation is selected as the evaluation feature, and the evaluation feature is analyzed to obtain the cognitive impairment assessment line, which includes the following sub-steps: Compare the time domain correlation with the frequency domain correlation. If the time domain correlation is greater than or equal to the frequency domain correlation, then output the time domain correlation signal; if the time domain correlation is less than the frequency domain correlation, then output the frequency domain correlation signal. If the time domain related signal is output, the time domain feature is used as the evaluation feature; if the frequency domain related signal is output, the frequency domain feature is used as the evaluation feature; The assessment characteristics were analyzed to obtain the cognitive impairment assessment line; Analyzing the assessment characteristics and obtaining the cognitive impairment assessment line includes the following sub-steps: The time domain analysis graph or the frequency domain analysis graph corresponding to the evaluation feature is marked as an evaluation analysis graph, and the trapezoid in the evaluation analysis graph is named as an evaluation trapezoid; Get the midpoint of the upper base of the evaluation trapezoid, mark it as the upper midpoint, get the midpoint of the lower base of the evaluation trapezoid, mark it as the lower midpoint; Connect the upper midpoint with the lower midpoint and label the resulting straight line as the cognitive impairment assessment line; The evaluation analysis chart is divided into two parts by the cognitive impairment evaluation line. The part close to the normal straight line is named the normal cognitive area, and the part close to the abnormal straight line is named the cognitive impairment area. The normal straight line includes a normal time domain straight line and a normal frequency domain straight line, and the abnormal straight line includes an abnormal time domain straight line and an abnormal frequency domain straight line.

2. The method for analyzing electroencephalogram signals for cognitive function detection according to claim 1, characterized in that: The EEG signal data of the subject is collected by using an EEG instrument, and the collected EEG signal is preprocessed, including the following sub-steps: Use an electroencephalogram instrument to collect the subject's brain electrical signals; The EEG signal is filtered by a bandpass filter to remove noise and obtain a preprocessed signal.

3. The method for analyzing electroencephalogram signals for cognitive function detection according to claim 2, characterized in that: Extracting signal features from the preprocessed EEG signal, wherein the signal features include time domain features and frequency domain features, comprises the following sub-steps: Extracting time domain features of the preprocessed signal, wherein the time domain features include average amplitude, variance, maximum value and minimum value; The frequency domain features of the preprocessed signal are extracted by Fourier transform, and the frequency domain features include power spectrum density and spectrum bandwidth.

4. The method for analyzing electroencephalogram signals for cognitive function detection according to claim 3, characterized in that: Collecting the user's evaluation features, marking them as features to be analyzed, and detecting and analyzing the features to be analyzed to determine whether the user has cognitive dysfunction includes the following sub-steps: Collecting user evaluation features and calculating evaluation indicators, wherein the evaluation indicators are time domain indicators or frequency domain indicators; Based on the user's evaluation characteristics and evaluation indicators, find the corresponding coordinate point in the evaluation analysis diagram and name it as the user evaluation point; Find out whether the user evaluation point is in the normal cognitive area or the cognitive impairment area. If the user evaluation point is in the normal cognitive area, output a normal cognitive signal. If the user evaluation point is in the cognitive impairment area, a cognitive impairment signal is output; If a normal cognitive signal is output, it indicates that the user's cognitive function is normal; If a cognitive impairment signal is output, it indicates that the user's cognitive function is impaired.

5. An electroencephalogram signal analysis system for cognitive function detection, used to implement an electroencephalogram signal analysis method for cognitive function detection as described in any one of claims 1 to 4, characterized in that: It includes a data preprocessing module, a feature extraction module, a correlation analysis module, a cognitive analysis module and a cognitive evaluation module; the data preprocessing module, the feature extraction module, the correlation analysis module and the cognitive evaluation module are respectively connected with the cognitive analysis module data; The data preprocessing module is used to collect the EEG signal data of the subject using an EEG instrument and preprocess the collected EEG signal; The feature extraction module is used to extract signal features from the preprocessed EEG signal, and the signal features include time domain features and frequency domain features; The association analysis module is used to divide the subjects into normal volunteers and cognitive impairment volunteers according to their own cognitive functions, and analyze the feature correlation of time domain features and frequency domain features based on the signal features of the normal volunteers and cognitive impairment volunteers; The cognitive analysis module is used to select the signal feature with the greatest feature correlation as the evaluation feature, establish a cognitive function detection model, analyze the evaluation feature, and obtain a cognitive impairment evaluation line; The cognitive assessment module is used to collect assessment features of the user, mark them as features to be analyzed, detect and analyze the features to be analyzed, and determine whether the user has cognitive dysfunction.

6. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the electroencephalogram signal analysis method for cognitive function detection as described in any one of claims 1 to 4 are executed.

Images