An intraoperative electrophysiological signal monitoring device and method based on SSEP signal analysis

Abstract

This invention discloses an intraoperative electrophysiological signal monitoring device and method based on SSEP signal analysis, belonging to the field of medical device technology. The intraoperative electrophysiological signal monitoring device of this invention includes: an electrical stimulation signal output unit for outputting a constant current pulse stimulation current to the target stimulation point; an electrophysiological signal acquisition unit for acquiring electroencephalogram (EEG) signals and / or electromyogram (EMG) signals generated after stimulation by the constant current pulse stimulation current; and a signal processing unit for processing and analyzing the acquired electrophysiological signals. This invention designs an intraoperative electrophysiological signal monitoring device with high safety performance, high real-time monitoring accuracy, and stable operation, providing a new method and approach for improving the accuracy of intraoperative monitoring.

Description

Technical Field

[0001] This invention belongs to the field of medical device technology, specifically relating to an intraoperative electrophysiological signal monitoring device and method based on SSEP signal analysis. Background Technology

[0002] With the continuous development of neuroscience, event-related potentials (EPs) are increasingly widely used in medicine. Evoked potentials (EPs) are bioelectrical activities generated by the central nervous system in response to external or internal stimuli. Human sensory organs, such as the eyes, ears, and skin, after receiving specific evoked stimuli such as light, sound, or weak electrical currents, transmit the perceived information to the central nervous system via their unique neural pathways. This information is continuously combined at various levels along the pathways, eventually reaching the cortex and triggering a series of activities.

[0003] Somatosensory evoked potentials (SEPs) are evoked potentials (EPs) produced by stimulating a subject's limbs with weak electrical currents, mechanical stimulation, or other forms of stimulation. Based on the latency of the evoked potential after stimulation, they can be classified into short-latency, medium-latency, and long-latency evoked potentials. Short-latency somatosensory evoked potentials (SSEPs) are less affected by various factors, have more stable waveforms, and can be repeatedly recorded, thus making them more widely used in clinical practice.

[0004] Somatosensory evoked potential (SEP) monitoring technology boasts advantages such as non-invasiveness, real-time performance, accuracy, and versatility, making it one of the most common electrophysiological signal monitoring techniques in clinical surgery. It plays a crucial role in helping surgeons ensure surgical safety and improve outcomes. Furthermore, due to the time-locked relationship between SEP and the stimulation signal, its waveform repeatability is excellent, and it exhibits a specific projection relationship with nerve function. Therefore, it plays a positive role in rapidly identifying nerve injury sites, improving surgical safety, and reducing patient suffering.

[0005] However, in existing technologies, the superposition averaging method is usually used to monitor and process the collected electrophysiological signals, which results in relatively poor accuracy and timeliness of signal analysis. Summary of the Invention

[0006] 1. The problem to be solved

[0007] The purpose of this invention is to provide an intraoperative electrophysiological signal monitoring device and method based on SSEP signal analysis, thereby solving the problems of relatively poor accuracy and real-time performance in existing electrophysiological signal analysis and processing technologies. This invention improves the accuracy and real-time performance of signal analysis by packaging preprocessed signals into real-time sequences and importing them into the CSEEA model for real-time extraction of SSEP signals.

[0008] 2. Technical Solution

[0009] To solve the above problems, the technical solution adopted by the present invention is as follows:

[0010] This invention provides an intraoperative electrophysiological signal monitoring device based on SSEP signal analysis, comprising:

[0011] An electrical stimulation signal output unit is used to output a constant current pulse stimulation current to the target stimulation point;

[0012] The electrophysiological signal acquisition unit is used to acquire electroencephalogram (EEG) signals and / or electromyogram (EMG) signals generated after stimulation by a constant current pulsed stimulation current.

[0013] The signal processing unit is used to process and analyze the acquired electrophysiological signals;

[0014] The signal processing unit includes a signal preprocessing module and an SSEP signal separation module. The signal preprocessing module is used to preprocess the raw signal acquired by the electrophysiological signal acquisition unit to obtain a first signal sequence X(t). The SSEP signal separation module is used to import X(t) into the CSEEA method model to extract a second signal sequence, namely the SSEP signal. Based on the extracted SSEP signal, the electrophysiological signal is monitored and analyzed.

[0015] Furthermore, the signal preprocessing module is used to sequentially filter, correct the baseline, and perform rereference processing on the acquired raw signal.

[0016] Furthermore, the extraction process of the SSEP signal separation module includes:

[0017] First, the signal sequence is decomposed into IMFs. Then, effective IMFs are selected based on the normalized optimization denoising index r and superimposed to obtain a denoised sequence containing SSEP signals, low-frequency noise, and spontaneous EEG signals.

[0018] For denoised sequences Windowing is performed, and all windows are z-scored to obtain the windows after score elimination, denoted as the sequence.

[0019] Will Projected onto principal component space And calculate each child window The root mean square mean μ and standard deviation σ are used to define the rejection threshold. Where k is a user-defined cutoff parameter;

[0020] Will Windowing is performed, but scoring is not eliminated, resulting in a complete... The sequence is projected onto the principal component space. Continue to The eigenvalues ​​of the square root are decomposed to obtain the eigenvector matrix D. t and eigenvector E t ;

[0021]

[0022] For each window, determine its D t Each eigenvalue vector is checked against a rejection threshold. If it is less, the eigenvalue vector is replaced with a zero vector K. Finally, spontaneous EEG signals less than the rejection threshold are removed as background noise, thus completing the SSEP signal extraction. Let the extracted effective component sequence, i.e., the SSEP signal sequence, be S. sig Then S sig Represented as:

[0023]

[0024] In the above formula, W sig Represented as The square root of.

[0025] Furthermore, the signal processing unit also includes a baseline coupling module, which is used to segment the extracted SSEP signal and introduce a coupling degree about the signal matrix for calculation, and determines the output of the alarm signal based on the coupling degree.

[0026] Furthermore, after the extracted SSEP signal is imported into the baseline coupling module, it is output at an electrical stimulation signal frequency f. z As the segmentation point, extract the signal matrix N from the interval of -20ms to 80ms for each segmentation point. i Acquire the zero-interference induced signal matrix S ref Induced signal S after interference sam traverse S ref Extract the maximum and minimum values ​​and their place values ​​for each row, and set them as F. ref Similarly, extract S sam Maximum and minimum values ​​and their place numbers f samThe coupling degree SSC between the induced signal after interference and the induced signal with zero interference is calculated using the following formula:

[0027]

[0028] Where p is a user-defined parameter, and the value of p is generally between 1 and 6; if the calculated coupling degree exceeds the set threshold, an alarm signal will be output.

[0029] Furthermore, the electrical stimulation signal output unit is used to output a square wave current with a frequency of 2 to 9 Hz, a pulse width of 0.2 to 0.6 ms, and a current amplitude of 0 to 33 mA.

[0030] Furthermore, the pulse signal output by the electrical stimulation signal output unit contains three square wave stimulation current pulses within a standard stimulation cycle. Each square wave stimulation current pulse consists of a positive square wave stimulation current pulse and a negative square wave stimulation current pulse.

[0031] Furthermore, the amplitude and pulse width of the first square wave stimulation current pulse and the second square wave stimulation current pulse are the same, while the amplitude and pulse width of the third square wave stimulation current pulse are 15% to 25% higher than the corresponding amplitude and pulse width values ​​of the first two pulses.

[0032] Furthermore, the stimulation parameters of the first square wave stimulation current pulse and the second square wave stimulation current pulse are adjusted and determined according to the dual-channel signal feedback. One channel is used to feed back the patient's fingertip vibration signal to the electrical stimulation signal output unit, and the other channel is used to feed back the amplitude of the SSEP signal to the electrical stimulation signal output unit.

[0033] The present invention provides a method for monitoring intraoperative electrophysiological signals based on SSEP signal analysis, employing the detection device of the present invention, comprising:

[0034] The electrical stimulation signal output unit outputs a constant current pulse stimulation current to the target stimulation point.

[0035] Electroencephalogram (EEG) and / or electromyogram (EMG) signals generated after constant current pulse stimulation are acquired by an electrophysiological signal acquisition unit.

[0036] The electrophysiological signal acquisition unit transmits the acquired signals to the signal processing unit. After signal preprocessing, the SSEP signal is extracted, and the monitoring and analysis results of intraoperative electrophysiological signals are obtained based on the extracted SSEP signal.

[0037] 3. Beneficial effects

[0038] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0039] (1) The present invention outputs constant current pulse stimulation current to the target stimulation point through an electrostimulation signal output unit, and then collects the electroencephalogram (EEG) and / or electromyogram (EMG) signals generated after stimulation by constant current pulse stimulation current through an electrophysiological signal acquisition unit; the electrophysiological signal acquisition unit transmits the acquired signals to the signal processing unit, and after signal preprocessing, the SSEP signal is extracted based on the CSEEA method model. Finally, the intraoperative electrophysiological signal monitoring and analysis is realized based on the extracted SSEP signal. Compared with the traditional superposition averaging method, the real-time performance of SSEP signal acquisition and analysis is improved, and the accuracy of intraoperative electrophysiological signal monitoring can be effectively guaranteed.

[0040] (2) The signal processing unit of the present invention further includes a baseline coupling module, which segments the extracted SSEP signal and introduces a coupling degree about the signal matrix for calculation, thereby determining whether to output an alarm signal based on the calculated value of the coupling degree. The introduction of the coupling degree SSC makes the alarm signal more intuitive and timely.

[0041] (3) The present invention further optimizes the pulse signal output by the electrical stimulation signal output unit, which contains three square wave stimulation current pulses in a standard stimulation cycle. Each square wave stimulation current pulse consists of a positive square wave stimulation current pulse and a negative square wave stimulation current pulse, thus better inducing the required electrophysiological signal. At the same time, by using the flip square wave signal, the maximum voltage required to maintain the constant current circuit is reduced, thereby improving the safety of the device.

[0042] (4) This invention designs a method for adjusting stimulation parameters using dual-channel signal feedback, which can determine the optimal stimulation parameters to the greatest extent possible in order to obtain the best SSEP signal extraction. Specifically, it adopts a method of introducing the amplitude of the signal returned by the vibration sensor and the SSEP signal into the analysis, which quantifies the effective response brought by the stimulation signal and prevents the occurrence of unsatisfactory SSEP signal extraction due to stimulation parameters determined by different analysts and differences in patients' physical sensations. Attached Figure Description

[0043] Figure 1 This is a schematic diagram illustrating the working principle of an intraoperative electrophysiological signal monitoring device based on SSEP signal analysis in one embodiment of the present invention.

[0044] Figure 2 This is a schematic diagram showing the placement of the stimulation electrode in one embodiment of the present invention;

[0045] Figure 3 This is a schematic diagram of a square wave pulse current in one embodiment of the present invention;

[0046] Figure 4 This is a schematic diagram of a method for adjusting stimulation parameters using dual-channel signal feedback in one embodiment of the present invention;

[0047] Figure 5 This is a diagram illustrating the effect of SSEP signal separation in one embodiment of the present invention. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application. Furthermore, unless otherwise specified, the embodiments and features described herein can be combined with each other.

[0049] This invention provides an intraoperative electrophysiological signal monitoring device based on SSEP signal analysis, combined with... Figure 1 It includes:

[0050] An electrical stimulation signal output unit is used to output a constant current pulse stimulation current to the target stimulation point;

[0051] An electrophysiological signal acquisition unit is used to acquire electrophysiological signals generated after stimulation with a constant current pulse stimulation current of a specific waveform; and

[0052] The signal processing unit is used to process and analyze the acquired electrophysiological signals.

[0053] A constant current pulse stimulation signal is applied to the human body through the electrical stimulation signal output unit. After the somatosensory evoked potential is generated, the signal is acquired by the electrophysiological signal acquisition unit and sent to the signal processing unit for processing.

[0054] As a further improvement to this embodiment of the invention, the signal processing unit includes a signal preprocessing module and an SSEP signal separation module, wherein:

[0055] The signal preprocessing module is used to preprocess the original signal to obtain the first signal sequence X(t), and export the first signal sequence X(y) to the SSEP signal separation module;

[0056] The SSEP signal separation module is used to import X(t) into the CSEEA method model to extract the second signal sequence, namely the SSEP signal. Then, based on the extracted SSEP signal, the monitoring and analysis of intraoperative electrophysiological signals can be realized, which improves the real-time performance of SSEP signal acquisition and analysis compared with the traditional superposition averaging method.

[0057] More preferably, the signal processing unit further includes a baseline coupling module, which segments the second signal sequence, calculates the coupling degree, and determines the output of the alarm signal based on the coupling degree. The introduction of the coupling degree (SSC) makes the alarm signal more intuitive and timely.

[0058] In a preferred embodiment of the present invention, the electrical stimulation signal output unit is used to output a square wave current with a frequency of 2 to 9 Hz, a pulse width of 0.2 to 0.6 ms, and a current amplitude of 0 to 33 mA. For example, an STM32F103ZET6 can be used as the main control chip. However, it should be noted that the type of main control chip is not limited to the specific type mentioned here.

[0059] Depending on the surgical location, the stimulation applied to the human body can be divided into upper limb SEP stimulation and lower limb SEP stimulation. The stimulation points, stimulation parameters, and signal acquisition points of the two types of stimulation are different.

[0060] like Figure 2 The diagram shows the placement of surface electrodes for upper and lower limb SEP stimulation, for example, using 44×22mm double button electrodes, connected to positive and negative electrodes respectively. For upper limb SEP stimulation, the electrode pads are placed on the skin along the median nerve at the wrist, with the anode placed at the wrist crease. For lower limb SEP stimulation, the electrodes are placed on the posterior aspect of the medial malleolus, midway between the posterior border of the Achilles tendon and the medial malleolus, with the cathode placed at the midpoint of the Achilles tendon.

[0061] As a further preferred option, such as Figure 3 As shown, the electrical stimulation signal output unit outputs a square wave pulse as a pulse signal. The pulse signal within a standard stimulation cycle contains three square wave stimulation current pulses. Each square wave stimulation current pulse consists of a forward square wave stimulation current pulse and a reverse square wave stimulation current pulse, with the same amplitude and pulse width but opposite current directions. The square wave stimulation current pulse is obtained by reversing the waveform direction of a square wave at half its pulse width, thereby allowing a smaller somatosensory voltage amplitude to achieve the required current amplitude, improving safety. More preferably, the amplitude and pulse width of the first and second square wave stimulation current pulses are the same, and the amplitude and pulse width of the third square wave stimulation current pulse are 15% to 25% higher than the corresponding amplitude and pulse width values ​​of the first two pulses, more preferably 20% higher. By adding a rhythmic stimulation signal with a brief increase in stimulation amplitude during a standard stimulation process, significant somatosensory evoked potential signals can be induced.

[0062] To better determine the optimal stimulation parameters, this invention further provides a method for adjusting stimulation parameters using dual-channel signal feedback in some embodiments. This method optimizes and determines the parameters of the first square wave stimulation current pulse and the second square wave stimulation current pulse, combined with… Figure 4 Specifically, the stimulation parameters are adjusted through dual-channel signal feedback. Two feedback channels are constructed to determine the optimal stimulation parameters. One channel uses vibration sensors placed at the patient's limb extremities (fingers, toes) to provide feedback on the vibration signal from the patient's fingertips during stimulation. The other channel provides feedback on the amplitude of the SSEP signal separated by the SSEP signal separation module. The parameters from both channels are fed back to the electrical stimulation signal output unit to determine the optimal stimulation parameters, thus completely avoiding inaccuracies caused by human interference and individual differences in sensory perception.

[0063] Specifically, the fingertip vibration signal is first extracted by upper and lower envelope extraction, resulting in two envelope signal sequences G1(x) and G2(x). Then, G1(x) and G2(x) are normalized by mean and peak value extracted to obtain two oscillating non-convergent signals g1(x) and g2(x). The SSEP signal extracted by the SSEP signal separation module is sensitive to stimulation parameters. If it does not produce a significant evoked potential in response to the stimulation signal, this can be reflected in the amplitude. Based on previous verification, thresholds are set for the two feedback channels. The preset amplitude range of the SSEP signal is P~N∈(8,20), which is considered a significant waveform, and flag S1 is set to 1. The preset difference between the two processed envelope signals |g1(x)-g2(x)|∈(0.1,0.8) is the fingertip stimulation receiving range; within this range, flag S2 is set to 1.

[0064] For the output frequency, pulse width, and current amplitude of the electrical stimulation signal, the output frequency and pulse width should first be controlled to their minimum values ​​(e.g., 2Hz and 0.2ms respectively), and the output current should be gradually increased from small to large. Observe the flag S1; after it is set to 1, maintain the current magnitude, and then control the output frequency and pulse width so that S2 is also set to 1. This can minimize the use of self-determined parameters and ensure the output of optimal stimulation parameters.

[0065] After somatosensory evoked potentials are generated, they need to be acquired using an electrophysiological signal acquisition unit. In some embodiments of the present invention, the electrophysiological signal acquisition unit preferably uses an ADS1299 as the front-end analog signal acquisition chip and a PIC32 as the acquisition main control chip, with a sampling frequency of 1000Hz and a signal-to-noise ratio of 121dB. Sixteen signal acquisition channels are set, including 8 channels for EEG signal acquisition and 8 channels for EMG signal acquisition.

[0066] During signal acquisition, the placement of the acquisition electrodes differs depending on whether upper limb SEP stimulation or lower limb SEP stimulation is required: For upper limb SEP stimulation, the electromyography (EMG) signal acquisition electrodes can be placed on muscle groups such as the extensor digitorum minora, extensor digitorum commonis, or brachioradialis; while the electroencephalogram (EEG) signal acquisition electrodes should be placed at the C3', C4', and C2 spinous processes according to the universal 10-20 system. For lower limb SEP stimulation, the EMG signal acquisition electrodes are placed on the posterior group of calf muscles, and the EEG signal acquisition electrodes are placed at the C2 and C2 spinous processes.

[0067] After acquiring electrophysiological signals, the electrophysiological signal acquisition unit transmits them to the signal processing unit for signal processing. The signal sequence sequentially passes through the signal preprocessing module, the SSEP signal separation module, and the baseline coupling module. In one embodiment of the invention, the signal processing unit uses the signal preprocessing module to preprocess the original signal, specifically by sequentially filtering, baseline correction, and rereference, thereby obtaining the first signal sequence X(t). Here, filtering, baseline correction, and rereference are all existing signal preprocessing methods, therefore their specific operations will not be described in detail.

[0068] In some embodiments of the present invention, the preprocessed first signal sequence is transmitted to the SSEP signal separation module, and the CSEEA method model is used to remove the pure SSEP signal to obtain the second signal sequence. The working principle and process of the SSEP signal separation module in processing the first signal sequence are as follows:

[0069] First, the first signal sequence is decomposed into IMFs. Then, effective IMFs are selected and superimposed based on a normalized optimization denoising index r. This effectively removes artifacts such as ECG signals and external noise, resulting in a denoised sequence containing SSEP signals, low-frequency noise, and spontaneous EEG signals.

[0070] Then With a window width of 100ms and Windowing is performed using a step size, all windows are z-scored, and windows with scores between -3 and 6 are retained and concatenated, denoted as a sequence. Will The formula for projecting onto the principal component space is as follows:

[0071]

[0072] Principal component space S is represented as The decomposed potential components, W sig Represented as The square root of W sig Perform eigenvalue decomposition to obtain the eigenvector matrix D.r and eigenvector E r .

[0073] Calculate each child window The root mean square mean μ and standard deviation σ are used to define the rejection threshold. Where k is a user-defined cutoff parameter.

[0074] Similarly, Windowing is performed, but scoring is not eliminated, resulting in a complete... Sequence, projected onto principal component space Continue to The eigenvalues ​​of the square root are decomposed to obtain the eigenvector matrix D. t and eigenvector E t .

[0075]

[0076] For each window, determine its D t Each eigenvalue vector is checked against a rejection threshold; if it is, the eigenvalue vector is replaced with a zero vector K. Finally, low-frequency, low-peak-varying spontaneous EEG signals and noise (spontaneous EEG signals below the rejection threshold are considered background noise) are removed, thus completing the SSEP signal extraction. Let the extracted effective component sequence be S. sig Then it can be expressed as:

[0077]

[0078] The final SSEP separation effect is as follows Figure 5 As shown.

[0079] Furthermore, the principle of the baseline coupling module is as follows: the SSEP signal extracted by the SSEP signal separation module is output at a frequency f according to the electrical stimulation signal in the determined stimulation parameters. z As the dividing point, a signal matrix N is created with the time interval from -20ms to 80ms for each dividing point. i Set the induced signal matrix S to collect interference-free signals. ref With the induced signal matrix S after interference san ; Traverse S ref Extract the maximum and minimum values ​​and their place values ​​for each row, and set them as F. ref Similarly, extract S san Maximum, minimum and their place numbers f sam The coupling degree between the induced signal after interference and the induced signal with zero interference is denoted as SSC, and its calculation formula is as follows:

[0080]

[0081] Where p is a user-defined parameter, and the SSC value range is... The closer the SSC value is to its maximum value, the worse the coupling. According to relevant standards, the constant C is calculated to be 0.67. If:

[0082]

[0083] If the threshold is exceeded, an alarm signal is output. The alarm signal output is determined by the calculated coupling degree, thus ensuring the integrity of the patient's neural signal pathways during surgery and guaranteeing patient safety.

[0084] This invention also provides a method for monitoring intraoperative electrophysiological signals based on SSEP signal analysis, the process of which is as follows: A stimulation electrode is attached to the site to be stimulated and connected to an electrical stimulation signal output unit. The electrical stimulation signal output unit outputs a constant current pulse stimulation current with a specific waveform. The optimal stimulation parameters are determined and kept constant by two feedback channels: the parameters returned by the vibration sensor and the amplitude of the SSEP signal separated by the SSEP signal separation module. A signal acquisition electrode is connected to collect the generated electrophysiological signals and transmit them to a signal processing unit. After passing through a signal preprocessing module, an SSEP signal separation module, and a baseline coupling module, the output coupling degree is compared to determine whether an alarm signal should be output.

[0085] In summary, the intraoperative electrophysiological signal monitoring device of the present invention is used to monitor the patient's postural neural integrity in real time during surgery. By stimulating specific points, it collects the specific electrophysiological signals generated: short latency somatosensory evoked potentials (SSEPs). The neural integrity is determined by preprocessing the electrophysiological signals, separating the SSEP signals, and coupling the baseline.

Claims

1. An intraoperative electrophysiological signal monitoring device based on SSEP signal analysis, characterized by, include: An electrical stimulation signal output unit is used to output a constant current pulse stimulation current to the target stimulation point; An electrophysiological signal acquisition unit is used to acquire electroencephalogram (EEG) signals and / or electromyogram (EMG) signals generated after stimulation by a constant current pulse stimulation current. The signal processing unit is used to process and analyze the acquired electrophysiological signals; The signal processing unit includes a signal preprocessing module, an SSEP signal separation module, and a baseline coupling module. The signal preprocessing module is used to preprocess the raw signals acquired by the electrophysiological signal acquisition unit to obtain a first signal sequence. The SSEP signal separation module is used to separate the signal from the signal. Import it into the CSEEA method model to extract the second signal sequence, namely the SSEP signal; The baseline coupling module is used to segment the extracted SSEP signal and introduce a coupling degree about the signal matrix for calculation. The output of the alarm signal is determined by the coupling degree. The extraction process of the SSEP signal separation module includes: First, process the signal sequence Decompose and optimize noise reduction metrics based on normalization. Filter out the effective The sequences are superimposed to obtain a denoised sequence containing SSEP signals, low-frequency noise, and spontaneous EEG signals. ; For denoised sequences Windowing is performed, and all windows are z-scored to obtain the windows after score elimination, denoted as the sequence. ; Will Projected onto principal component space And calculate each child window The root mean square of and standard deviation Define the rejection threshold , where k is a user-defined cutoff parameter; Will Windowing is performed, but scoring is not eliminated, resulting in a complete... The sequence is projected onto the principal component space. Continue to The eigenvalues ​​of the square root are decomposed to obtain the eigenvector matrix. and eigenvalue vectors ; For each window, determine its... If each eigenvector is less than the rejection threshold, then the eigenvector is replaced with a zero vector. Finally, low-frequency, low-peak-value spontaneous EEG signals and noise are removed, thus completing the SSEP signal extraction process. Let the extracted effective component sequence, i.e., the SSEP signal sequence, be... ,but express: In the above formula, is expressed as the square root of.

2. The intraoperative electrophysiological signal monitoring device based on SSEP signal analysis of claim 1, wherein, The signal preprocessing module is used to sequentially filter, correct the baseline, and perform rereference processing on the acquired raw signal.

3. The intraoperative electrophysiological signal monitoring device based on SSEP signal analysis according to claim 1 or 2, characterized in that, The electrical stimulation signal output unit is used to output a square wave current with a frequency of 2~9Hz, a pulse width of 0.2~0.6ms, and a current amplitude of 0~33mA.

4. The intraoperative electrophysiological signal monitoring device based on SSEP signal analysis of claim 3, wherein, The pulse signal output by the electrical stimulation signal output unit contains three square wave stimulation current pulses within a standard stimulation cycle. Each square wave stimulation current pulse consists of a positive square wave stimulation current pulse and a negative square wave stimulation current pulse.

5. The intraoperative electrophysiological signal monitoring device based on SSEP signal analysis of claim 4, wherein, The amplitude and pulse width of the first square wave stimulation current pulse and the second square wave stimulation current pulse are the same. The amplitude and pulse width of the third square wave stimulation current pulse are 15-25% higher than the corresponding amplitude and pulse width values ​​of the first two pulses.

6. The intraoperative electrophysiological signal monitoring device based on SSEP signal analysis according to claim 5, characterized in that, The stimulation parameters of the first square wave stimulation current pulse and the second square wave stimulation current pulse are adjusted and determined according to the dual-channel signal feedback. One channel is used to feed back the patient's fingertip vibration signal to the electrical stimulation signal output unit, and the other channel is used to feed back the amplitude of the SSEP signal to the electrical stimulation signal output unit.

7. A method for monitoring intraoperative electrophysiological signals based on SSEP signal analysis, characterized in that, The monitoring device according to any one of claims 1-6 comprises: The electrical stimulation signal output unit outputs a constant current pulse stimulation current to the target stimulation point. The electroencephalogram (EEG) and / or electromyogram (EMG) signals generated after constant current pulse stimulation are acquired by the electrophysiological signal acquisition unit. The electrophysiological signal acquisition unit transmits the acquired signals to the signal processing unit. After signal preprocessing, the SSEP signal is extracted. Based on the extracted SSEP signal, the monitoring and analysis results of intraoperative electrophysiological signals are obtained. The signal processing unit includes a signal preprocessing module, an SSEP signal separation module, and a baseline coupling module. The signal preprocessing module is used to preprocess the raw signals acquired by the electrophysiological signal acquisition unit to obtain a first signal sequence. The SSEP signal separation module is used to separate the signal from the signal. The CSEEA method model is imported to extract the second signal sequence, namely the SSEP signal. The baseline coupling module is used to segment the extracted SSEP signal and introduces a coupling degree about the signal matrix for calculation. The output of the alarm signal is determined by the coupling degree. The extraction process of the SSEP signal separation module includes: First, process the signal sequence Decompose and optimize noise reduction metrics based on normalization. Filter out the effective The sequences are superimposed to obtain a denoised sequence containing SSEP signals, low-frequency noise, and spontaneous EEG signals. ; For denoised sequences Windowing is performed, and all windows are z-scored to obtain the windows after score elimination, denoted as the sequence. ; Will Projected onto principal component space And calculate each child window The root mean square of and standard deviation Define the rejection threshold , where k is a user-defined cutoff parameter; Will Windowing is performed, but scoring is not eliminated, resulting in a complete... The sequence is projected onto the principal component space. Continue to The eigenvalues ​​of the square root are decomposed to obtain the eigenvector matrix. and eigenvalue vectors ; For each window, determine its... If each eigenvector is less than the rejection threshold, then the eigenvector is replaced with a zero vector. Finally, low-frequency, low-peak-value spontaneous EEG signals and noise are removed, thus completing the SSEP signal extraction process. Let the extracted effective component sequence, i.e., the SSEP signal sequence, be... ,but express: In the above formula, is expressed as the square root of.

Images