Psychological counseling earphone based on brain wave detection technology

Abstract

The invention discloses a psychological counseling earphone based on a brain wave detection technology, and relates to the technical field of detection, and the psychological counseling earphone comprises a sensing module, a signal acquisition module, an AI analysis module, an audio counseling module, a sound module and a database. The method comprises the following steps: collecting electroencephalogram signals and heart rate signals of a patient in real time, carrying out noise suppression and signal enhancement processing, carrying out deep fusion on the electroencephalogram signals and the heart rate signals of the patient to analyze a psychological state mode of the patient, judging whether the patient needs to carry out music guidance intervention, and if so, carrying out music guidance intervention. According to the method, the complex psychological state can be accurately recognized, the accuracy of electroencephalogram signal collection and the accuracy of heart rate signals are guaranteed, and the effectiveness of a psychological counseling strategy is also guaranteed.

Description

Technical Field

[0001] The present invention relates to the field of detection technology, and in particular to a psychological counseling headset based on brain wave detection technology. Background Art

[0002] With the trend of deep integration of consumer electronics and biomedical technology, psychological counseling headphones with integrated brainwave and heart rate monitoring are entering a critical development period relying on breakthroughs in non-invasive biosensing technology. Heart rate sensors based on the principle of photoplethysmography have achieved chip-level miniaturization integration with the help of MEMS micromachining technology. By optimizing the sensor layout and anti-motion interference algorithm in areas such as the temple and earlobe, they can accurately capture heart rate variability data and provide real-time physiological indicators for stress assessment. In the field of EEG signal acquisition, single-channel dry electrode sensors represented by NeuroSkyMindWave2, combined with analog-to-digital conversion and Bluetooth transmission technology, break through the volume limitations of traditional brain-computer interface devices, realize dynamic monitoring of EEG characteristics such as α / β waves in the forehead, and promote the lightweight application of EEG data in consumer-grade devices.

[0003] Traditional psychological counseling headphones based on brainwave detection technology collect the patient's EEG signals or heart rate signals based on a single-channel dry electrode sensor or a PPG-based heart rate sensor, and identify the patient's psychological state based on the patient's EEG signals or the patient's heart rate signals. Obviously, this type of psychological counseling headphones based on brainwave detection technology has at least the following shortcomings: 1. Traditional psychological counseling headphones based on brainwave detection technology use a single-channel dry electrode sensor to collect the patient's EEG signals, and the existing single-channel dry electrode sensor is easily affected by skin contact status and movement displacement, has high noise and incomplete feature extraction, is difficult to accurately reflect real EEG activity, and cannot guarantee the accuracy of EEG signal acquisition.

[0004] 2. Traditional psychological counseling headphones based on brainwave detection technology collect the patient's heart rate signal through a PPG-based heart rate sensor. In scenarios such as strong light, exercise, and skin color differences, the light signal of the PPG-based heart rate sensor is easily distorted and seriously interfered by motion artifacts, making it impossible to guarantee the accuracy of the heart rate signal.

[0005] 3. Traditional psychological counseling headphones based on brainwave detection technology can only independently monitor brainwave signals and heart rate signals, lack deep fusion analysis, and cannot accurately identify complex psychological states, and thus cannot guarantee the effectiveness of psychological counseling strategies. Summary of the Invention

[0006] In view of the above-mentioned technical deficiencies, the purpose of the present invention is to provide a psychological counseling headset based on brain wave detection technology.

[0007] In order to solve the above technical problems, the present invention adopts the following technical solution: The present invention provides a psychological counseling headset based on brain wave detection technology, including the following modules: a sensing module, a signal acquisition module, an AI analysis module, an audio counseling module, a sound module and a database.

[0008] The sensing module is used to combine and arrange the electroencephalographic sensing electrodes and the electrocardiographic sensing electrodes.

[0009] The signal acquisition module is used to synchronously acquire the patient's EEG signals and heart rate signals through the EEG induction electrodes and the ECG induction electrodes, and perform noise suppression and signal enhancement processing.

[0010] The AI analysis module is used to analyze the patient's psychological state pattern based on the collected EEG signals and heart rate signals.

[0011] The audio guidance module is used to trigger the sound module to play the corresponding guidance audio after identifying the patient's psychological state pattern.

[0012] The audio module is used to play corresponding guidance audio according to a preset program.

[0013] The database is used to store joint feature vectors under different mental state modes, corresponding mental state labels, and various dangerous mental state modes.

[0014] The beneficial effects of the present invention are: 1. The present invention provides a psychological counseling headset based on brain wave detection technology, which combines brain wave sensing electrodes and heart wave sensing electrodes, collects the patient's brain wave signals and heart rate signals in real time, and performs noise suppression and signal enhancement processing. At the same time, the patient's brain wave signals and heart rate signals are deeply integrated to analyze the patient's psychological state pattern, and determine whether the patient needs counseling music intervention. If counseling music intervention is needed, the corresponding counseling audio is played, which can accurately identify complex psychological states, ensure the accuracy of brain wave signal acquisition and the accuracy of heart rate signals, and also ensure the effectiveness of psychological counseling strategies.

[0015] 2. The psychological counseling earphones of the present invention contain a chip and two electrodes. The chip and the two electrodes are connected by two orange short wires to form a sensing structure. The two orange short wires are respectively connected to the electrodes on the left and right sides. The left electrode is an EEG sensing electrode, which fits the forehead. The right electrode is an EEG sensing electrode. The EEG sensing electrode and the EEG sensing electrode are timestamp-aligned through the chip's built-in synchronous clock module. At the same time, noise suppression and signal enhancement processing are performed on the EEG signal and heart rate signal, ensuring the accuracy of EEG signal acquisition and the accuracy of heart rate signals.

[0016] 3. The present invention performs short-time Fourier transform on the EEG signal and calculates the relative power values of α wave, β wave, θ wave and δ wave, and extracts the approximate entropy, sample entropy, permutation entropy and zero-crossing rate of the EEG signal at the same time, performs fast Fourier transform on the heart rate cycle sequence, extracts low-frequency power, high-frequency power and LF / HF ratio, and extracts the root mean square of the difference between adjacent heart rate cycles, the standard deviation of the heart rate cycle and the average heart rate, and constructs a joint feature vector. The patient's psychological state pattern is identified based on the patient's joint feature vector, which can accurately identify complex psychological states and ensure the effectiveness of psychological counseling strategies. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 This is a schematic diagram of the system structure connection of the present invention.

[0019] Figure 2 This is a schematic diagram of the earphone structure of the present invention. DETAILED DESCRIPTION

[0020] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making creative efforts are within the scope of protection of the present invention.

[0021] See also Figure 1 As shown, the present invention provides a psychological counseling headset based on brain wave detection technology, including the following modules: a sensing module, a signal acquisition module, an AI analysis module, an audio counseling module, a sound module and a database.

[0022] The sensing module is connected to the signal acquisition module, the signal acquisition module is connected to the AI analysis module, the AI analysis module is connected to the audio guidance module, the audio guidance module is connected to the sound module, and the database is connected to the AI analysis module.

[0023] The sensing module is used to combine and arrange the electroencephalographic sensing electrodes and the electrocardiographic sensing electrodes.

[0024] In a specific embodiment, the EEG induction electrodes and the ECG induction electrodes are combined and arranged, and the specific process is as follows: the psychological counseling earphones contain a chip and two electrodes. The chip and the two electrodes are connected by two orange short wires to form an induction structure. The two orange short wires are respectively connected to the electrodes on the left and right sides. The left electrode is the EEG induction electrode, which is attached to the forehead, and the right electrode is the ECG induction electrode.

[0025] The signal acquisition module is used to synchronously acquire the patient's EEG signals and heart rate signals through the EEG induction electrodes and the ECG induction electrodes, and perform noise suppression and signal enhancement processing.

[0026] In a specific embodiment, the synchronous acquisition of EEG signals and heart rate signals is as follows: the EEG sensing electrodes sample real-time EEG signals at a 256Hz sampling rate, and the ECG sensing electrodes synchronously sample ECG signals at a 100Hz sampling rate. At the same time, the EEG sensing electrodes and the ECG sensing electrodes are time-stamp aligned through the chip's built-in synchronous clock module.

[0027] It should be noted that the synchronous clock module generates a stable clock signal based on a high-precision crystal oscillator to ensure that the EEG electrodes and the ECG electrodes can maintain time synchronization when collecting signals.

[0028] In the above, timestamp alignment is achieved through the synchronous clock module. The specific process is as follows: the EEG electrodes and the ECG electrodes are connected to the synchronous clock module and receive the reference clock signal emitted by the synchronous clock module. When the EEG electrodes and the ECG electrodes respectively sample real-time EEG signals and heart rate signals at 256 Hz and 100 Hz sampling rates, the EEG electrodes and the ECG electrodes determine the collection moments of the EEG electrodes and the ECG electrodes based on the received reference clock signal. At the same time, based on the clock signal status of each EEG electrode and the clock signal status of each ECG electrode at each collection moment, the timestamps of each EEG electrode and the ECG electrodes are generated.

[0029] It should be noted that the circuit design inside the EEG electrodes and ECG electrodes is equipped with a dedicated clock signal receiving module. When the EEG electrodes and ECG electrodes receive the reference clock signal, they divide the clock signal to reduce the frequency of the reference clock signal to a frequency corresponding to the sampling rate. When a rising edge or a falling edge appears in the divided clock signal, it is regarded as a sampling trigger signal. After receiving this trigger signal, the acquisition circuit of the EEG electrodes and ECG electrodes samples the EEG signal and heart rate signal, and records the clock signal state at this time as the basic timestamp information of the EEG signal sample and the heart rate signal sample collected this time.

[0030] It should also be noted that the clock signal state includes the number of clock cycles and phase, etc.

[0031] In another specific embodiment, the noise suppression and signal enhancement processing are performed, and the specific process is as follows: for the EEG signal, the power frequency interference is removed by a 50Hz notch filter, and a bandpass filter is used to retain the effective EEG signal frequency band, and at the same time, an adaptive noise cancellation algorithm is combined to reduce skin contact noise and motion artifacts.

[0032] It should be noted that the frequency range of the bandpass filter is 0.5Hz-30Hz, that is, the lower cutoff frequency of the bandpass filter is 0.5Hz and the upper cutoff frequency is 30Hz. For the EEG signal components with frequencies below 0.5Hz and above 30Hz, the filter will attenuate them to reduce their amplitude, while the EEG signals in the frequency band of 0.5Hz-30Hz can pass through the filter relatively smoothly, thereby achieving the purpose of retaining the effective EEG frequency band.

[0033] It should also be noted that the collected EEG signal is input into the adaptive filter. The adaptive filter continuously adjusts its own parameters according to the relationship between the reference signal and the EEG signal, outputs an estimated value of the noise, and then subtracts this estimated noise value from the EEG signal to obtain the EEG signal after noise cancellation.

[0034] For heart rate signals, a dual-wavelength light fusion algorithm is used to correct skin color differences and ambient light interference, and a moving average filter is used to remove high-frequency motion artifacts. At the same time, an R-wave peak detection algorithm is used to extract the heart rate cycle.

[0035] It should be noted that the process of correcting skin color differences and ambient light interference through the dual-wavelength light fusion algorithm is as follows: a light-emitting diode that can emit green light and infrared light is used as a light source, and a photodetector is used to receive the reflected green light signal and infrared light signal. The green light signal and infrared light signal are processed, and then weighted summation is performed to obtain the fused heart rate signal.

[0036] It should also be noted that the process of using moving average filtering to remove high-frequency motion artifacts is: set a window containing N sampling points, starting from the starting point of the heart rate signal, calculate the average value of the N sampling points, and use it as the output value of the position after filtering. Then move the window backward by one sampling point, calculate the average value of the N sampling points in the new window, and so on, until the entire signal is processed.

[0037] The process of extracting the heart rate cycle through the R-wave peak detection algorithm is as follows: after detecting similar R-wave peaks in the heart rate signal through the Pan-Tompkins algorithm, the time interval between adjacent R-wave peaks is calculated and used as the heart rate cycle.

[0038] The AI analysis module is used to analyze the patient's psychological state pattern based on the collected EEG signals and heart rate signals.

[0039] In a specific embodiment, the AI analysis module has the following specific process: deep feature extraction and fusion of the patient's EEG signals and heart rate signals, and construction of the patient's joint feature vector, while obtaining the joint feature vectors under different psychological state modes and the corresponding psychological state labels from the database, and inputting them into the AI model. The AI model learns the mapping relationship between each psychological state mode and the joint feature vector based on the input joint feature vectors under different psychological state modes and the corresponding psychological state labels, and then inputs the patient's joint feature vector into the AI model to identify the patient's psychological state mode and analyze whether the patient needs counseling audio intervention.

[0040] It should be noted that the psychological state pattern includes happiness, happiness, anxiety and high stress. When the psychological state pattern is happy, the psychological state label is happy.

[0041] It should also be noted that the AI model identifies which psychological state pattern the patient's joint feature vector corresponds to based on the learned mapping relationship between each psychological state pattern and the joint feature vector.

[0042] In the above, the specific process of deep feature extraction and fusion is as follows: short-time Fourier transform is performed on the EEG signal, and the relative power values of α wave, β wave, θ wave and δ wave are calculated, and the approximate entropy, sample entropy, permutation entropy and zero-crossing rate of the EEG signal are extracted at the same time; fast Fourier transform is performed on the heart rate cycle sequence, low-frequency power, high-frequency power and LF / HF ratio are extracted, and the root mean square of the difference between adjacent heart rate cycles, the standard deviation of the heart rate cycle and the average heart rate are extracted.

[0043] It should be noted that the frequency range of α waves is 8Hz-13Hz, the frequency range of β waves is 13Hz-30Hz, the frequency range of θ waves is 4Hz-8Hz, and the frequency range of δ waves is 0.5Hz-4Hz.

[0044] It should also be noted that the EEG signal is converted into a frequency domain signal through short-time Fourier transform, and the power spectrum values within the frequency range of α wave, β wave, θ wave and δ wave are added respectively to obtain the power of α wave, β wave, θ wave and δ wave, and the power of α wave, β wave, θ wave and δ wave are added to obtain the total power of the EEG signal, and the power of α wave, β wave, θ wave and δ wave are divided by the total power of the EEG signal to obtain the relative power values of α wave, β wave, θ wave and δ wave.

[0045] Among them, the embedding dimension m and the similarity tolerance r are determined. First, the EEG signal at each acquisition moment is recorded as x(n), where n represents the number of each acquisition moment, n = 1, 2, 3, ..., N, N represents the total number of acquisition moments, and n and N are both positive integers. Then, each acquisition moment is divided into subsequences of length m, recorded as X i , X i =[x(i),x(i+1),...,x(i+m-1)], where i represents the number of each subsequence, i=1,2,3,...,N-m+1, and the distance between each subsequence is calculated, denoted as d(X i ,X j ), i≠j, X j represents the jth subsequence, d(X i ,X j )=max k＝0,1,...,m-1 |x(i+k)-x(j+k)|, d(X i ,X j ) and r, and statistically analyze d(X i ,X j )≤r, denoted as Simultaneous calculation Then increase the subsequence length to m+1 and repeat the above steps to get B m+1 (r), then the approximate entropy is: For sample entropy, Then the sample entropy is:

[0046] For permutation entropy, X i Arrange the elements in ascending order to obtain the arrangement pattern π of the i-th subsequence i , and count the number of times different arrangement patterns appear, recorded as n π , and calculate the probability of different arrangement patterns: Then the permutation entropy is: -∑ π p π lnp π .

[0047] For the zero-crossing rate, the product of the EEG signals at each adjacent acquisition moment is calculated. If the product of the EEG signals at a certain adjacent acquisition moment is less than 0, it means that a zero-crossing occurs at that adjacent acquisition moment. The number of zero-crossings in the EEG signal is counted and recorded as Z. The zero-crossing rate is:

[0048] It should be explained that the frequency range of low-frequency power is 0.04-0.15Hz, and the frequency range of high-frequency power is 0.15-0.4Hz. The heart rate signal is converted into a frequency domain signal by fast Fourier transform, and the power spectrum values in the frequency range of low-frequency power and the frequency range of high-frequency power are added to obtain the low-frequency power and high-frequency power, respectively.

[0049] In the above, the difference between each adjacent heart rate cycle is calculated, and the square value of the difference between each adjacent heart rate cycle is calculated. At the same time, the average value of the square value of the difference between each adjacent heart rate cycle is calculated. Finally, the square root of the average value of the square value of the difference between each adjacent heart rate cycle is calculated to obtain the root mean square of the difference between adjacent heart rate cycles.

[0050] Each heart rate cycle is converted into a heart rate value, and the average of the heart rate values of each heart rate cycle is calculated to obtain the average heart rate. The formula for converting the heart rate cycle into a heart rate value is:

[0051] In the above, the specific process of constructing the joint eigenvector is as follows: the relative power values of the α wave, β wave, θ wave and δ wave of the EEG signal, as well as the approximate entropy, sample entropy, permutation entropy and zero-crossing rate of the EEG signal are spliced with the low-frequency power, high-frequency power and LF / HF ratio of the heart rate signal, as well as the root mean square of the difference between adjacent heart rate cycles, the standard deviation of the heart rate cycle and the average heart rate to obtain the joint eigenvector, and the dimensional difference of the joint eigenvector is eliminated.

[0052] It should be noted that the joint feature vector contains comprehensive information about brain electrical activity and cardiac autonomic nervous function. The various features of the EEG signal and the various features of the heart rate signal are arranged in the joint feature vector according to a preset order. The various features of the EEG signal include the relative power values of α waves, β waves, θ waves and δ waves, as well as the approximate entropy, sample entropy, permutation entropy and zero-crossing rate of the EEG signal. The various features of the heart rate signal include low-frequency power, high-frequency power and LF / HF ratio, as well as the root mean square of the difference between adjacent heart rate cycles, the standard deviation of the heart rate cycle and the average heart rate.

[0053] Among them, the preset arrangement order is used to evaluate whether the arrangement order of each feature of the EEG signal and each feature of the heart rate signal is reasonable, and is set by relevant staff. When the arrangement order of each feature of the EEG signal and each feature of the heart rate signal is different from the preset arrangement order, it represents that the arrangement order of each feature of the EEG signal and each feature of the heart rate signal is unreasonable. When the arrangement order of each feature of the EEG signal and each feature of the heart rate signal is the same as the preset arrangement order, it represents that the arrangement order of each feature of the EEG signal and each feature of the heart rate signal is reasonable.

[0054] It should also be noted that the dimensionality difference of the joint eigenvector is eliminated by normalization.

[0055] In the above, the analysis of whether the patient needs audio counseling intervention is as follows: the patient's psychological state pattern is compared with each dangerous psychological state pattern in the database. If the patient's psychological state pattern is different from each dangerous psychological state pattern in the database, it means that the patient does not need audio counseling intervention. If the patient's psychological state pattern is the same as a dangerous psychological state pattern in the database, it means that the patient needs audio counseling intervention. It should be noted that each dangerous psychological state pattern includes high stress and anxiety. The audio counseling module is used to trigger the sound module to play the corresponding counseling audio after identifying the patient's psychological state pattern.

[0056] In one specific embodiment, the audio counseling module operates as follows: When a patient requires audio counseling intervention, a corresponding signal is generated based on the patient's psychological state pattern and fed back to the audio module. The audio module compares the received signal with the signals of various counseling audio programs in a preset program. If the received signal matches a particular counseling audio program, the preset program is invoked to play the counseling audio program. The audio module is configured to play the corresponding counseling audio program according to the preset program. The database is configured to store joint feature vectors for different psychological state patterns, as well as corresponding psychological state labels and various dangerous psychological state patterns. This embodiment of the present invention utilizes a combined layout of EEG and ECG electrodes to collect the patient's EEG and heart rate signals in real time, perform noise suppression and signal enhancement processing, and simultaneously deeply fuse the patient's EEG and heart rate signals to analyze the patient's psychological state pattern and determine whether the patient requires music counseling intervention. If so, the corresponding counseling audio program is played. This allows for precise identification of complex psychological states, ensuring the accuracy of EEG signal acquisition and heart rate signals, and ensuring the effectiveness of psychological counseling strategies. The above contents are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the described specific embodiments or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined in this specification, they should all fall within the scope of protection of the present invention.

Claims

1. A psychological counseling headset based on brainwave detection technology includes a psychological counseling headset body and a psychological counseling headset control system for controlling the operation of the psychological counseling headset body. The psychological counseling headset control system includes an audio module and a sensing module. The audio module is used to play corresponding counseling audio according to a preset program, and the sensing module is used to combine and arrange EEG induction electrodes and ECG induction electrodes; characterized in that: The psychological counseling earphone control system also includes: The signal acquisition module is used to synchronously collect the patient's EEG signals and heart rate signals through EEG induction electrodes and ECG induction electrodes, and perform noise suppression and signal enhancement processing; AI analysis module, used to analyze the patient's psychological state pattern based on collected EEG signals and heart rate signals; The audio guidance module is used to trigger the audio module to play the corresponding guidance audio after identifying the patient's psychological state pattern; The database is used to store joint feature vectors under different mental state modes, corresponding mental state labels and various dangerous mental state modes.

2. The psychological counseling earphones based on brainwave detection technology according to claim 1, characterized in that: The specific process of combining and laying out the electroencephalographic electrodes and the electrocardiographic electrodes is as follows: The psychological counseling earphones contain a chip and two electrodes. The chip and the two electrodes are connected by two short orange wires to form a sensing structure. The two short orange wires are connected to the electrodes on the left and right sides respectively. The left electrode is the EEG electrode, which is attached to the forehead, and the right electrode is the EKG electrode.

3. The psychological counseling earphone based on brainwave detection technology according to claim 1, characterized in that: The specific process of synchronously collecting EEG signals and heart rate signals is as follows: The EEG electrodes sample EEG signals in real time at a 256Hz sampling rate, while the ECG electrodes synchronously sample ECG signals at a 100Hz sampling rate. At the same time, the EEG electrodes and ECG electrodes are time-stamp aligned through the chip's built-in synchronous clock module.

4. The psychological counseling earphones based on brainwave detection technology according to claim 3, characterized in that: Timestamp alignment is achieved through the synchronous clock module. The specific process is as follows: The EEG electrodes and the ECG electrodes are connected to a synchronous clock module and receive a reference clock signal from the synchronous clock module. When the EEG electrodes and the ECG electrodes collect real-time EEG signals and heart rate signals at a sampling rate of 256 Hz and 100 Hz, respectively, the EEG electrodes and the ECG electrodes determine each collection moment of the EEG electrodes and each collection moment of the ECG electrodes based on the received reference clock signal. Simultaneously, based on the reference clock signal status at each collection moment of the EEG electrodes and the reference clock signal status at each collection moment of the ECG electrodes, timestamps for each collection moment of the ECG electrodes and timestamps for each collection moment of the ECG electrodes are generated.

5. The psychological counseling earphones based on brainwave detection technology according to claim 1, characterized in that: The noise suppression and signal enhancement processing is performed, and the specific process is as follows: For EEG signals, a 50Hz notch filter is used to remove power frequency interference, and a bandpass filter is used to retain the effective EEG signal frequency band. At the same time, an adaptive noise cancellation algorithm is used to reduce skin contact noise and motion artifacts. For heart rate signals, a dual-wavelength light fusion algorithm is used to correct skin color differences and ambient light interference, and a moving average filter is used to remove high-frequency motion artifacts. At the same time, an R-wave peak detection algorithm is used to extract the heart rate cycle.

6. The psychological counseling earphones based on brainwave detection technology according to claim 1, characterized in that: The specific process of the AI analysis module is as follows: Deep feature extraction and fusion are performed on the patient's EEG signals and heart rate signals, and the patient's joint feature vector is constructed. At the same time, the joint feature vectors under different psychological state modes and the corresponding psychological state labels are obtained from the database and input into the AI model. The AI model learns the mapping relationship between each psychological state mode and the joint feature vector based on the input joint feature vectors under different psychological state modes and the corresponding psychological state labels. The patient's joint feature vector is then input into the AI model to identify the patient's psychological state mode and analyze whether the patient needs counseling audio intervention.

7. The psychological counseling earphone based on brainwave detection technology according to claim 6, characterized in that: The specific process of deep feature extraction and fusion is as follows: The EEG signals were subjected to short-time Fourier transform, and the relative power values of α, β, θ, and δ waves were calculated. The approximate entropy, sample entropy, permutation entropy, and zero-crossing rate of the EEG signals were extracted. The heart rate cycle sequence was subjected to fast Fourier transform to extract the low-frequency power, high-frequency power, and LF / HF ratio. The root mean square of the difference between adjacent heart rate cycles, the standard deviation of the heart rate cycle, and the average heart rate were also extracted.

8. The psychological counseling earphone based on brainwave detection technology according to claim 6, characterized in that: The specific process of constructing the joint feature vector is as follows: The relative power values of α, β, θ and δ waves of the EEG signal, as well as the approximate entropy, sample entropy, permutation entropy and zero-crossing rate of the EEG signal are concatenated with the low-frequency power, high-frequency power and LF / HF ratio of the heart rate signal, as well as the root mean square of the difference between adjacent heart rate cycles, the standard deviation of the heart rate cycle and the average heart rate to obtain a joint eigenvector, and the dimensional difference of the joint eigenvector is eliminated.

9. The psychological counseling earphone based on brainwave detection technology according to claim 6, characterized in that: The specific process of analyzing whether the patient needs audio guidance intervention is as follows: The patient's mental state pattern is compared with the dangerous mental state patterns in the database. If the patient's mental state pattern is different from the dangerous mental state patterns in the database, it means that the patient does not need counseling audio intervention. If the patient's mental state pattern is the same as a dangerous mental state pattern in the database, it means that the patient needs counseling audio intervention.

10. The psychological counseling earphone based on brainwave detection technology according to claim 1, characterized in that: The specific process of the audio guidance module is as follows: When a patient needs counseling audio intervention, a corresponding signal is generated according to the patient's psychological state pattern and fed back to the audio module. The audio module compares the received signal with the signals of each counseling audio in the preset program. If the signal received by the audio module is the same as the signal of a counseling audio, the preset program is called to play the counseling audio.

Images