Device and method for detecting frequency characteristics of EEG (electroencephalogram) acquisition equipment

Abstract

The invention relates to a frequency characteristic detection device and method for EEG acquisition equipment, and belongs to the field of medical instrument quality control. The device comprises a programmable signal generator, a white noise generation module, a signal superposition module, a signal transmission module, a positioning plate and a fixing piece. The method for detecting the frequency characteristic by using the device comprises the following steps of: preprocessing an electroencephalogram signal acquired by the EEG acquisition equipment to obtain an effective electroencephalogram signal; the effective electroencephalogram signals are divided into a plurality of frequency bands, and each frequency band generates a plurality of discrete frequency points; for each discrete frequency point, calculating a relative error between the power spectral density of the effective electroencephalogram signal and the power spectral density of a theoretical signal; assigning different preset weights to each frequency band, and weighting and summing relative errors of all discrete frequency points based on the preset weights to obtain a final error; and carrying out frequency characteristic detection on the EEG acquisition equipment based on the final error. The technical problem that frequency characteristic detection accuracy is difficult to guarantee in the prior art is solved.

Description

Technical Field

[0001] This invention relates to a frequency characteristic detection device and method for EEG acquisition equipment, belonging to the field of medical device quality control. Background Technology

[0002] With the development of biomedical and information technologies, research on human brain activity is becoming increasingly in-depth. Electroencephalography (EEG), as a non-invasive method for monitoring brain function, plays a crucial role in clinical diagnosis, sleep research, and cognitive science research. By recording the weak electrical signals spontaneously generated by the brain through electrodes placed on the scalp, EEG can provide real-time information on the state of brain activity, which is of great value for understanding the brain's working mechanisms and diagnosing diseases. However, because EEG signals are very weak and easily interfered with by environmental noise, and although EEG acquisition equipment is relatively mature, some problems still exist in practical applications, such as data distortion caused by electrode oxidation after prolonged use and inconsistencies between different batches of products. These problems not only affect the quality of EEG data but also pose challenges to EEG-based applications.

[0003] To improve the stability and data accuracy of EEG acquisition equipment, it is necessary to develop an effective detection device and method. Among these, the frequency characteristics of the EEG acquisition equipment are a key performance indicator. Its bandwidth, noise suppression capability, and dynamic response range directly determine the fidelity and anti-interference capability of the EEG signal (usually 1Hz~100Hz). For example, insufficient drift suppression in the low-frequency band (<1Hz) can lead to distortion of slow waves (such as delta waves), while aliasing effects in the high-frequency band (>30Hz) may mask fine signal features such as gamma waves (γ waves).

[0004] Currently, there are few tools available on the market specifically for testing the frequency characteristics of EEG acquisition equipment, which cannot meet the frequency characteristic control requirements of EEG acquisition equipment. Especially with the rapid development of new portable or wearable EEG acquisition devices, how to efficiently and accurately perform frequency characteristic testing has become one of the urgent problems to be solved. Summary of the Invention

[0005] The purpose of this invention is to provide a frequency characteristic detection device and method for EEG acquisition equipment, aiming to solve the technical problem that the existing technology cannot guarantee the accuracy of frequency characteristic detection.

[0006] To achieve the above objectives, the technical solution of the present invention is: a frequency characteristic detection device and method for EEG acquisition equipment, the device comprising:

[0007] Programmable signal generator: used to generate EEG signals of different frequencies;

[0008] White noise generation module: used to generate Gaussian white noise signals;

[0009] Signal superposition module: used to linearly superimpose Gaussian white noise signals with EEG signals;

[0010] Signal transmission module: used to transmit the linearly superimposed signal;

[0011] Positioning plate: includes positioning holes and signal coupling holes. The positioning holes are used to place the signal transmission module, and the signal transmission module transmits the linearly superimposed signal to the electrode through the signal coupling holes.

[0012] Fixing component: Used to fix the signal transmission module to the positioning plate.

[0013] Furthermore, to achieve the above objectives, the present invention also provides a method for frequency characteristic detection using the frequency characteristic detection device of the EEG acquisition equipment, comprising the following steps:

[0014] Step 1: Preprocess the collected EEG data to obtain valid EEG data;

[0015] Step 2: Divide the effective EEG data into several frequency bands, and generate several discrete frequency points for each frequency band;

[0016] Step 3: For each discrete frequency point, calculate the relative error between the power spectral density of the effective EEG data and the power spectral density of the theoretical signal;

[0017] Step 4: Assign different preset weights to each frequency band, and sum the relative errors of all discrete frequency points based on the preset weights to obtain the final error;

[0018] Step 5: Detect the frequency characteristics of the EEG acquisition device based on the final error.

[0019] Optionally, the EEG data is generated by the frequency characteristic detection device of the EEG acquisition equipment. Specifically, a programmable signal generator generates EEG signals of different frequencies, a white noise generation module generates Gaussian white noise signals, a signal superposition module linearly superimposes the Gaussian white noise signals with the EEG signals to obtain EEG data, and a signal transmission module transmits the EEG data.

[0020] Optionally, the preprocessing includes:

[0021] The EEG data was subjected to trap filtering to remove mains power interference;

[0022] The EEG data is bandpass filtered to remove invalid frequency components.

[0023] Optionally, the discrete frequency points are generated using a preset fixed step size.

[0024] Optionally, the power spectral density is obtained based on the Fourier transform, and its expression is:

[0025]

[0026] In the formula, For power spectral density, The Fourier transform of the effective EEG data is given, where N is the number of signal points, determined by the sampling rate of the EEG acquisition device and the duration of the signal output by the programmable signal generator.

[0027] Optionally, Step 5 includes:

[0028] If the final error This indicates that the EEG acquisition device has good frequency characteristics;

[0029] like This indicates that the frequency characteristics of the EEG acquisition device are generally normal.

[0030] like This indicates that the frequency characteristics of the EEG acquisition device are poor.

[0031] The beneficial effects of this invention are:

[0032] 1. This invention generates Gaussian white noise to simulate the noise generated during actual detection, making the signal more consistent with the characteristics of EEG signals. It also uses an operational amplifier to achieve signal isolation, making the output signals independent and non-interfering with each other, and allowing the frequency characteristics of each electrode to be detected separately.

[0033] 2. This invention uses frequency band differentiation for weighted detection and achieves quantitative evaluation of the accuracy of multi-frequency band signal acquisition through weighted error. It effectively solves the problem of accurate detection of EEG signal acquisition equipment, establishes a standardized testing process for quality control of medical electronic equipment, and provides technical support for the scientific evaluation of the frequency characteristics of EEG equipment. Attached Figure Description

[0034] Figure 1 This is a circuit diagram of the white noise generation module in this embodiment;

[0035] Figure 2 This is a circuit diagram of the signal superposition module in this embodiment;

[0036] Figure 3 This is the signal transmission module in this embodiment;

[0037] Figure 4 This is a front view of the fastener in this embodiment;

[0038] Figure 5 This is a left view of the fastener in this embodiment;

[0039] Figure 6 This is a top view of the fastener in this embodiment;

[0040] Figure 7 This is an external view of the positioning plate in this embodiment, wherein component 1 is the positioning plate, component 2 is the positioning hole, and component 3 is the signal coupling hole;

[0041] Figure 8 This is a simplified schematic diagram of the EEG acquisition device, signal transmission device, fixture, and positioning plate assembled in this embodiment.

[0042] Figure 9 This is a system block diagram of the frequency characteristic detection device of the EEG acquisition equipment in this embodiment. Detailed Implementation

[0043] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0044] Example 1

[0045] In this embodiment, the frequency characteristic detection device of the EEG acquisition equipment includes:

[0046] Programmable signal generator: used to generate EEG signals of different frequencies;

[0047] White noise generation module: used to generate Gaussian white noise signals;

[0048] Signal superposition module: used to linearly superimpose Gaussian white noise signals with EEG signals;

[0049] Signal transmission module: used to transmit the linearly superimposed signal;

[0050] Positioning plate: includes positioning holes and signal coupling holes. The positioning holes are used to place the signal transmission module, and the signal transmission module transmits the linearly superimposed signal to the electrode through the signal coupling holes.

[0051] Fixing component: Used to fix the signal transmission module to the positioning plate.

[0052] Optionally, in this embodiment, the programmable signal generator is used to simulate EEG signals. The amplitude of the generated simulated EEG signals is between 5-100uV, which conforms to the amplitude range of EEG signals. It can simultaneously generate signals in five different frequency bands. In some specific implementations, the specific frequency range can be: , , , , .

[0053] Optionally, in this embodiment, the signal parameters of each frequency band are dynamically adjusted according to the following rules:

[0054] Frequency control strategy: Each cycle is 30 seconds, and the frequency of each frequency band increases by a preset step size;

[0055] Fixed amplitude strategy: The amplitude values ​​of the five frequency bands are locked at constant outputs of 80μV, 50μV, 75μV, 20μV, and 5μV respectively;

[0056] The specific parameter configurations are shown in Table 1:

[0057] Table 1. Programmable Signal Generator Output Signal Frequency Setting Values

[0058]

[0059] In Table 1, the starting point (0s) is defined as the moment when initialization is completed and data acquisition begins.

[0060] Optionally, the initial phase of all signals input to the programmable signal generator is set to zero.

[0061] Optionally, Figure 1 The Gaussian white noise signal generated by the white noise generation module has energy that is evenly distributed across the entire EEG spectrum (1-100Hz), and can simulate various environmental noises. In this embodiment, the Gaussian white noise is resistive thermal noise, which is generated by the thermal disturbance of charge carriers inside the resistor as a noise source.

[0062] Optionally, Figure 1 , Figure 2 , Figure 3 The operational amplifiers used can all be precision operational amplifiers. Figure 1 The operational amplifier model can be selected as LTC2063. The capacitance value of capacitor C1 is 22pF, the capacitance values ​​of capacitors C2 and C3 are 100nF, the capacitance value of capacitor CX1 is 47nF, the resistance value of resistor R1 is 10MΩ, the resistance value of resistor R2 is 4.99KΩ, the resistance value of resistor R3 is 1MΩ, and the resistance value of resistor RS1 is 10KΩ. The amplification factor of the operational amplifier is 1+R3 / R2=21. The capacitors C2 and C3 can filter out high-frequency disturbances and enhance anti-interference ability.

[0063] Optionally, Figure 1 The white noise output in the middle is through Figure 2 The operational amplifier in the signal source is added to the signal output from the signal source, where, Figure 2The operational amplifier model used can be LM324J. The resistors R_1, R_2, R_3, R_4, R_5, R4, Ri1, and Rf1 are all 10K ohms, and the capacitor C4 is 100nF.

[0064] Optionally, Figure 3 Lieutenant General Figure 2 The output signal is converted into four signals after passing through an operational amplifier array, among which, Figure 3 The operational amplifier model can be selected as LM324ADR. The resistance of resistors R5, R7, R8, R9, R10, R12, and R13 is 10KΩ, the resistance of resistor R11 is 20KΩ, and the capacitance of capacitors C5, C6, C7, C8, and C9 is 100nF.

[0065] Understandably, in Figure 3 The use of operational amplifiers can isolate the preceding and following circuits, and each output is isolated from the others so that they will not affect each other. Here, ELECTRODE_PAD refers to the electrode pad that is in contact with the electrodes of the EEG acquisition device.

[0066] Optionally, the fastener structure is as follows: Figure 4 As shown, it can be made of ABS material. The fastener has a disc at the top, a circular protrusion in the middle, and a 3mm thread at the bottom.

[0067] The front view of the fastener structure is as follows Figure 4 As shown, the diameters of the cylindrical structures are 10mm and 4mm respectively, and the threads are M3 machine threads.

[0068] Left view of the fastener structure as shown Figure 5 As shown, the width of each cylindrical structure is 2.5mm and 3mm, and the thread length is 6.5mm.

[0069] Top view of the fastener structure as follows Figure 6 As shown, supplement the width direction and the projection relationship of each cylinder.

[0070] Optionally, Figure 7 This is an external view of the positioning plate in this embodiment. It is made of ABS material. In the figure, 2 represents the positioning hole, which is used to install the fixing component. In the figure, 3 represents the signal coupling hole, which is used to transmit signals to the EEG electrode.

[0071] Optionally, Figure 8 This is a simplified schematic diagram of the EEG acquisition device, signal transmission device, fixing component, and positioning plate after assembly in this embodiment. The signal transmission device is connected to the positioning plate through the fixing component.

[0072] Optionally, Figure 9This is a system block diagram of the frequency response detection device for the EEG acquisition equipment in this embodiment. The signal generated by the programmable signal generator and the signal from the white noise generation module are linearly superimposed by the signal superposition module, and then transmitted to the EEG electrodes of the EEG acquisition equipment via the signal transmission module to test the frequency response of the EEG acquisition equipment.

[0073] It is understood that the fixing device is used to maintain a fixed distance between the EEG acquisition device and the measuring device, ensuring that there is an appropriate distance between them. Therefore, it can not only reduce mutual interference, but also optimize the accuracy and stability of signal acquisition.

[0074] Optionally, in this embodiment, since the signal value recorded by the EEG acquisition device is the difference between the signal of a certain electrode and the signal of the reference electrode, the operational amplifier of the signal transmission module corresponding to the reference electrode has a gain of 2. Except for the reference electrode, the operational amplifier of the signal transmission module corresponding to other electrodes has a gain of 1. The signal value of the reference electrode is twice the signal value of other electrodes. The white noise generation module and the signal superposition module are collectively referred to as the circuit module.

[0075] Optionally, in this embodiment, the positioning plate may be made of ABS material, and signal coupling holes may be pre-set on the positioning plate in accordance with the International Leading Association (ITA) standard.

[0076] Optionally, in this embodiment, the signal transmission module circuit board can be made of a flexible substrate. The flexible substrate can be made of a material with a high degree of flexibility. Multiple signal transmission modules are mounted on the positioning plate and spliced ​​into a style that conforms to the International Leading Association (ITA) standard.

[0077] Optionally, in this embodiment, the EEG acquisition device used employs a 32-channel system based on the INTERNATIONAL 10-20 standard.

[0078] It should be noted that, in order to ensure the best test results, in this embodiment, the modules of the frequency characteristic detection device of the EEG acquisition equipment must be installed in a specific order, and before installation, a program should be written on the programmable signal generator to output signals as shown in Table 1. The specific installation order is as follows:

[0079] First, place the signal transmission module in the pre-set position on the positioning plate and secure it firmly through the positioning holes.

[0080] Then, a white noise generation module and a signal superposition module are installed below the positioning plate. The signal generator is connected to the signal superposition module via wires, and the white noise generation module is connected to the signal superposition module via wires.

[0081] Finally, check that all connections are correct and ensure that signal transmission between modules is unimpeded.

[0082] In the technical solution provided in this embodiment, all modules work together to effectively evaluate and ensure the frequency characteristics of the EEG acquisition device under various usage conditions, enabling it to meet the stringent requirements of clinical diagnosis, scientific research and other fields.

[0083] Example 2

[0084] Based on Embodiment 1, a method for detecting frequency characteristics using the frequency characteristic detection device of the EEG acquisition equipment includes the following steps:

[0085] Step 1: Preprocess the collected EEG data to obtain valid EEG data.

[0086] Optionally, the EEG data is generated by the frequency characteristic detection device of the EEG acquisition equipment. Specifically, a programmable signal generator generates EEG signals of different frequencies, a white noise generation module generates Gaussian white noise signals, a signal superposition module linearly superimposes the Gaussian white noise signals with the EEG signals to obtain EEG data, and a signal transmission module transmits the EEG data.

[0087] Optionally, before using the frequency characteristic detection device of the EEG acquisition equipment to generate EEG data, the frequency characteristic detection device of the EEG acquisition equipment and the EEG acquisition equipment need to be installed together using fasteners. Before installation, it is necessary to ensure that the electrode positions of the EEG acquisition equipment are consistent with the lead standards of the frequency characteristic detection device of the EEG acquisition equipment, and to observe whether the positions of the frequency characteristic detection device of the EEG acquisition equipment and the EEG acquisition equipment match. If they do not match, they need to be reinstalled.

[0088] Furthermore, turn on the frequency characteristic detection device of the EEG acquisition equipment and the power supply of the EEG acquisition equipment, and run the written program to record the timestamp through the programmable signal generator. After waiting for seven minutes, the system initialization and signal transmission are completed.

[0089] Optionally, the preprocessing includes:

[0090] The EEG data was subjected to trap filtering to remove mains power interference;

[0091] The EEG data was bandpass filtered to retain the effective frequency components.

[0092] Optionally, in some specific implementations, a 50Hz trap filter and a 1-100Hz bandpass filter are performed.

[0093] Step 2: Divide the effective EEG data into several frequency bands, and generate several discrete frequency points for each frequency band;

[0094] Optionally, in this embodiment, within the frequency range [1Hz, 100Hz], 100 discrete frequency points are generated with a fixed step size of 1Hz, denoted as . , .

[0095] Step 3: For each discrete frequency point, calculate the relative error between the power spectral density of the effective EEG data and the power spectral density of the theoretical signal;

[0096] Specifically, the power spectral density measured by the EEG acquisition device is first calculated. Compared with the theoretical value of the output signal absolute error :

[0097]

[0098] After that Divide by the corresponding discrete frequency point The relative error is obtained. :

[0099]

[0100] The power spectral density is obtained based on the Fourier transform, and its expression is:

[0101]

[0102]

[0103] In the formula, For power spectral density, The valid EEG data Fourier transform, where N is the signal The number of points is determined by the sampling rate of the EEG acquisition device and the duration of the signal output by the programmable signal generator.

[0104] The theoretical signal is the signal that is theoretically output after the program is run. The error between this signal and the actual output signal of the device is negligible, so it can be used as a theoretical signal.

[0105] Step 4: Assign different preset weights to each frequency band, and sum the relative errors of all discrete frequency points based on the preset weights to obtain the final error;

[0106] Optionally, since different frequency bands have varying importance in EEG data analysis, this embodiment assigns different weights to the discrete frequency points of the power spectral density of different frequency bands, thus assigning different weights to the frequency bands. , , , , Each discrete frequency point of the power spectral density is assigned a weight of 0.01, 0.02, 0.03, 0.03, and 0.003, respectively, and then the final error is obtained by weighted summation. The frequency bands are divided based on the frequency bands of brain waves, which are divided into alpha waves, beta waves, gamma waves, etc. The weights are assigned according to the importance of the signal in the actual analysis. and It plays a secondary role in EEG signal analysis, therefore it was given a smaller weight. In practical analysis, this signal plays a dominant role, so it is assigned a larger weight. The specific formula for the final error is as follows:

[0107]

[0108] Step 5: Detect the frequency characteristics of the EEG acquisition device based on the final error.

[0109] Optionally, based on the requirements for frequency response characteristics in GB 9706.226-2021 standard, the final error judgment criteria in this embodiment are as follows:

[0110] If the final error This indicates that the EEG acquisition device has good frequency characteristics, meaning it has extremely high accuracy and stability during signal acquisition, processing, and transmission.

[0111] like This indicates that the frequency characteristics of the EEG acquisition device are generally poor, meaning that if it is to be used in a precision measurement scenario, the device needs to be calibrated.

[0112] like This indicates that the frequency characteristics of the EEG acquisition device are poor, meaning that the validity and reliability of the acquired data are low. In this case, detailed calibration and maintenance are required, or the hardware and software systems need to be updated to improve its frequency characteristics.

[0113] Furthermore, in this embodiment, after testing the EEG acquisition device, the power spectral density of the signal output by the device of the present invention and the signal acquired by the EEG acquisition device were calculated respectively, and the final error was found to be 5%. It can be considered that the frequency characteristics of the EEG acquisition device in this embodiment are good.

[0114] The specific embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of the present invention.

Claims

1. A frequency characteristic detection device for an EEG acquisition device, characterized in that, The device includes: Programmable signal generator: used to generate EEG signals of different frequencies; White noise generation module: used to generate Gaussian white noise signals; Signal superposition module: used to linearly superimpose Gaussian white noise signals with EEG signals; Signal transmission module: used to transmit the linearly superimposed signal; Positioning plate: includes positioning holes and signal coupling holes. The positioning holes are used to place the signal transmission module, and the signal transmission module transmits the linearly superimposed signal to the electrode through the signal coupling holes. Fixing component: Used to fix the signal transmission module to the positioning plate.

2. A method for frequency characteristic detection using the frequency characteristic detection device of the EEG acquisition equipment as described in claim 1, characterized in that, Includes the following steps: Step 1: Preprocess the collected EEG data to obtain valid EEG data; Step 2: Divide the effective EEG data into several frequency bands, and generate several discrete frequency points for each frequency band; Step 3: For each discrete frequency point, calculate the relative error between the power spectral density of the effective EEG data and the power spectral density of the theoretical signal; Step 4: Assign different preset weights to each frequency band, and sum the relative errors of all discrete frequency points based on the preset weights to obtain the final error; Step 5: Detect the frequency characteristics of the EEG acquisition device based on the final error.

3. The method according to claim 2, characterized in that, The EEG data is generated by the frequency characteristic detection device of the EEG acquisition equipment. Specifically, a programmable signal generator generates EEG signals of different frequencies, a white noise generation module generates Gaussian white noise signals, a signal superposition module linearly superimposes the Gaussian white noise signals with the EEG signals to obtain EEG data, and a signal transmission module transmits the EEG data.

4. The method according to claim 2, characterized in that, The preprocessing includes: The EEG data was subjected to trap filtering to remove mains power interference; The EEG data is bandpass filtered to remove invalid frequency components.

5. The method according to claim 2, characterized in that, The discrete frequency points are generated using a preset fixed step size.

6. The method according to claim 2, characterized in that, The power spectral density is obtained based on the Fourier transform, and its expression is: ； In the formula, For power spectral density, The Fourier transform of the effective EEG data is given, where N is the number of signal points, determined by the sampling rate of the EEG acquisition device and the duration of the signal output by the programmable signal generator.

7. The method according to claim 2, characterized in that, Step 5 includes: If the final error This indicates that the EEG acquisition device has good frequency characteristics; like This indicates that the frequency characteristics of the EEG acquisition device are generally normal. like This indicates that the frequency characteristics of the EEG acquisition device are poor.

Images