Neurovascular coupling-based prognosis prediction system for spinal cord electrical stimulation in patients with disorders of consciousness

Abstract

This invention discloses a spinal cord electrical stimulation (SCS) prognostic prediction system for patients with consciousness disorders based on neurovascular coupling, belonging to the field of brain-computer interface technology. The system simultaneously acquires the patient's EEG and cerebral blood flow signals using an EEG acquisition device and transcranial Doppler ultrasound. After preprocessing, it calculates brain network connectivity using a weighted pairwise phase consistency index and calculates neurovascular coupling parameters based on a phase-amplitude coupling method. Finally, the results are presented through a display module to predict the prognosis. This system addresses the lack of prognostic assessment indicators for SCS treatment in patients with consciousness disorders. It has advantages such as simple and reusable acquisition equipment, assisting doctors in providing personalized clinical plans and reducing treatment cycles. The system includes data acquisition, processing, and result display modules, with a clear calculation process for key parameters, making it suitable for prognostic assessment of patients with consciousness disorders after SCS implantation.

Description

Technical Field

[0001] This invention relates to the field of brain-computer interface (BCI) technology, and in particular to a prognostic prediction system for spinal cord electrical stimulation in patients with consciousness disorders based on neurovascular coupling. Background Technology

[0002] Disorders of consciousness refer to varying degrees of alteration in an individual's state of awareness, typically manifested as a weakened or lost ability to respond to external stimuli. Disorders of consciousness can be caused by a variety of factors, including neurological diseases, metabolic disorders, drug poisoning, infections, and trauma. Traumatic brain injury (TBI) and non-traumatic brain injury (cerebrovascular disease, hypoxic-ischemic encephalopathy) are the main causes. Globally, approximately 50-60 million new TBI patients are diagnosed each year, and about 0.3% of these patients develop disorders of consciousness. In addition, more than 12.2 million new stroke cases each year also lead to many patients with disorders of consciousness.

[0003] The treatment and prognosis of patients with disorders of consciousness (DOC) place considerable pressure and burden on doctors and patients' families. Firstly, the treatment time varies greatly. Acute DOCs, if diagnosed promptly and treated effectively, may recover within days or weeks; however, chronic or severe DOCs, such as those caused by cerebrovascular events or degenerative diseases, may require months or even longer for treatment and rehabilitation. Neuromodulation therapy, a commonly used clinical method for treating DOCs, offers a personalized treatment plan that can be adjusted according to the patient's condition, with advantages such as fewer side effects and reusability. Currently, neuromodulation methods used clinically to treat DOC patients include transcranial magnetic stimulation (TMS), deep brain stimulation (DBS), transcranial direct current stimulation (DCS), spinal cord stimulation (SBS), vagus nerve stimulation (VAS), and functional electrical stimulation (FESPS). Implantable spinal cord stimulation (SCS) involves implanting stimulating electrodes along the epidural space to the C2-C4 level, using a frequency of 70 Hz to increase neuronal activity. However, clinical observations indicate that doctors lack clear predictive indicators for the prognostic effects of SCS implantation. This not only prevents doctors from making timely personalized adjustments to patients' medical plans, but also means that patients' families cannot keep abreast of the patients' treatment progress and condition, resulting in a significant psychological burden.

[0004] Brain-Cyclical Interference (BCI) refers to an artificially constructed pathway between the brain and computers or external devices, distinct from traditional brain information transmission. It can replace, rebuild, strengthen, supplement, or improve the normal output of the central nervous system. Based on EEG signal characteristics, it can be categorized into active, reactive, and passive BCI. Passive BCI refers to inferring the user's state, emotions, or cognitive load by monitoring brain electrical activity or physiological signals (such as EEG, EOG, etc.) without actively controlling brain waves or generating specific intentions. Brain network connectivity calculated based on EEG data; neurovascular coupling parameters calculated based on EEG and TCD data both belong to passive BCI. It reflects the functional synergy of various brain regions and the dynamic coupling between brain activity and blood supply, helping to infer the brain's health status and cognitive load. This is reflected by analyzing the relationship between EEG signals and cerebral blood flow.

[0005] Resting-state brain network connectivity refers to the network connections formed between brain regions through spontaneous activity in the absence of external tasks or specific behavioral instructions. Resting-state brain networks reflect the coordinated activity of different brain regions by detecting functional connectivity patterns between them. This invention uses the Weighted Pairwise Phase Consistency (WPPC) index to calculate and obtain the patient's resting-state brain network connectivity, which can more accurately and robustly reflect the functional connectivity between different brain regions. Brain network connectivity calculated based on EEG has many advantages, including extremely high temporal resolution, relatively low monitoring cost, portable equipment, and simple data acquisition.

[0006] Neurovascular coupling (NVC) refers to the interaction between neural activity and local blood flow. In the brain, increased neural activity induces regulation of blood flow, thereby providing more oxygen and nutrients to meet the energy demands of activated areas. Phase-amplitude coupling methods based on EEG and TCD blood flow data are used to calculate NVC, primarily by analyzing the phase and amplitude relationship between neural electrical activity and blood flow to explore the dynamic coupling mechanism between them. This method not only boasts high temporal resolution and accuracy but also enables the analysis of nonlinear and complex neurovascular interactions, while also offering advantages such as low cost and reusability.

[0007] Passive brain-computer interface (BCI) offers advantages such as non-invasiveness, ease of testing, reusability, and low cost, leading to its widespread application in various clinical examinations. The brain network connectivity and neurovascular coupling proposed in this invention hold great potential in predicting postoperative prognosis in patients with disorders of consciousness, thereby helping physicians assess the patient's postoperative condition. Summary of the Invention

[0008] The purpose of this invention is to provide a prognostic prediction system for spinal cord stimulation in patients with disorders of consciousness based on neurovascular coupling, which solves the problems in the prior art where there are no clear indicators for observing and evaluating the therapeutic effect of implanted spinal cord nerve stimulators on patients with disorders of consciousness, and the prognostic effect of implanted spinal cord nerve stimulators is often difficult to predict.

[0009] To achieve the above objectives, the present invention provides a spinal cord electrical stimulation prognostic prediction system for patients with impaired consciousness based on neurovascular coupling, comprising:

[0010] The data acquisition module uses EEG acquisition equipment and transcranial Doppler ultrasound to simultaneously acquire the patient's EEG and cerebral blood flow signals;

[0011] The EEG and cerebral blood flow data processing module preprocesses the raw EEG signals acquired by the data acquisition module and calculates brain network connectivity and neurovascular coupling based on the preprocessed data.

[0012] The output display module displays the feature parameters obtained after processing by the EEG and cerebral blood flow data processing modules.

[0013] Preferably, when the data acquisition module is working, the transcranial Doppler ultrasound probe is fixed in the temporal region, that is, about 2-3 cm above the ear, usually at the midpoint of the line connecting the eyebrow and the ear; the sampling rate is set to 125 Hz to acquire information such as cerebral blood flow velocity in the middle cerebral artery (MCA); considering that the TCD probe occupies part of the head, the number of leads in the EEG cap needs to be reduced as much as possible, so the EEG cap adopts a 19-lead EEG acquisition system designed by the international 10-20 system; standard Ag / Agcl electrodes are used, the sampling rate is set to 512 Hz, the forehead is grounded, and the impedance between the scalp and the electrodes is kept below 10KΩ; the acquired EEG signals and cerebral blood flow information are saved as EDF and txt files respectively.

[0014] Preferably, the EEG and cerebral blood flow data processing module includes an EEG data preprocessing unit, a cerebral blood flow data processing unit, a brain network connectivity calculation unit, and a neurovascular coupling calculation unit.

[0015] Preferably, the EEG data preprocessing unit uses EEGLAB, a brainwave processing toolbox developed based on MATLAB, to preprocess the raw EEG signal; including the following steps:

[0016] S11. Bandpass filtering: A third-order Butterworth bandpass filter with a filtering range of 0.5-40Hz is used to filter the data and remove extremely low and extremely high frequency interference in the EEG signal.

[0017] S12, 50Hz notch filter: Removes 50Hz power frequency interference from the acquisition system;

[0018] S13. Interpolation Bad Lead: Use the interpolation bad lead function in the EEGLAB toolkit to eliminate distorted leads;

[0019] S14. Downsampling: When calculating neurovascular coupling, the EEG sampling rate is reduced to 125Hz to maintain the synchronization of EEG and cerebral blood flow data.

[0020] S15, ICA artifact removal: Remove electrooculography and electromyography artifacts that occur during the acquisition process using the ICA artifact removal function in the EEGLAB toolkit;

[0021] S16. Data Segmentation: Divide the EEG data into several 5-second segments, calculate the brain network connectivity for each segment, and take the average value as the average brain network connectivity state for that time segment.

[0022] Preferably, the cerebral blood flow data processing module obtains the frequency band of interest from the cerebral blood flow envelope value acquired by TCD through filtering, and extracts the corresponding phase information.

[0023] Preferably, the connectivity of brain networks is calculated based on the weighted pairwise phase consistency index, and the calculation steps are as follows:

[0024] S21. Phase Extraction: Perform wavelet transform on the time series of each brain region to extract phase information;

[0025] S22. Calculate the phase difference: For each lead, calculate its phase difference using the following formula:

[0026] Δφ ij (t)=φ i (t)-φ j (t) (1)

[0027] Where, Δφ ij (t) represents the phase difference between the i-th lead and the j-th lead at time t; φ i (t) represents the phase value of the EEG signal in the i-th lead; φ j (t) represents the phase value of the EEG signal in the j-th lead;

[0028] S23. Calculate the weighted pairwise phase consistency index: Calculate the magnitude of the phase difference using the following formula:

[0029]

[0030] Among them, WPPC ij represents the weighted pairwise phase coherence index between leads i and j; T represents the total number of time points of the signal; exp() is a complex exponential function;

[0031] S24. Normalization: Perform normalization processing, the formula is:

[0032]

[0033] Among them, Normalized WPPC ij This represents the weighted pairwise phase consistency index after normalization, with an output value in [0, 1].

[0034] Preferably, the neurovascular coupling calculation unit applies phase-amplitude coupling to calculate neurovascular coupling, obtaining the coupling relationship between low-frequency blood flow and high-frequency neuronal activity. The calculation steps are as follows:

[0035] S31. Blood Flow Phase Information Extraction: The phase information of the frequency band of interest in the blood flow signal is extracted using Hilbert transform. The calculation formula is as follows:

[0036]

[0037] in, The instantaneous phase value at time t is represented by arg(); arg() calculates the phase angle of the blood flow signal; Hilbert() performs a Hilbert transform on the input signal to generate an analytic signal; CBFv low (t) Low-frequency cerebral blood flow velocity signal at time t;

[0038] S32. EEG Amplitude Information Extraction: The amplitude information of the frequency band of interest in the EEG signal is extracted using Hilbert transform. The calculation formula is as follows:

[0039] AEEG(t)=|Hilbert(EEG high (t))| (5)

[0040] Where AEEG(t) represents the instantaneous amplitude of high-frequency EEG at time t; EEG hight (t) represents the high-frequency EEG signal at time t;

[0041] S33. Calculate the average amplitude in the phase using the following formula:

[0042]

[0043] Where Z(t) represents the neurovascular coupling strength value at time t; It is a complex exponential function that combines blood flow phase information and EEG amplitude information; This represents the instantaneous blood flow phase at time t;

[0044] S34. Calculate the coupling value: Calculate the average phase and magnitude of the complex exponential function and take the modulus. The calculation formula is as follows:

[0045]

[0046] Wherein, PAC represents the neurovascular coupling strength value; This represents the average of the complex exponential function, where T represents the total number of time points.

[0047] Preferably, the display module includes three parts: patient information, detailed results, and comparative prediction. The patient information part includes information such as the patient's age, gender, medical history, coma duration, coma recovery scale score, and patient's nursing level. The detailed results part allows users to view detailed analysis results of different frequency bands of EEG data from a single patient study. The comparative prediction part provides a prediction of the patient's prognosis by comparing data collected at different time points.

[0048] Therefore, the above-mentioned spinal cord electrical stimulation prognostic prediction system for patients with impaired consciousness based on neurovascular coupling, as described in this invention, has the following beneficial effects:

[0049] (1) By collecting the patient's electroencephalogram (EEG) and cerebral blood flow data, the clinical prognosis of the patient in the wakefulness-promoting treatment was predicted;

[0050] (2) The data acquisition device of the present invention is simple, convenient and reusable;

[0051] (3) By objectively assessing the recovery of patients’ neuronal activity and cerebral blood flow, doctors can be assisted in providing personalized clinical plans for patients, reducing the treatment cycle for patients and avoiding waste of doctors and medical resources.

[0052] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0053] Figure 1 This is a diagram of the brain network and the brain state observation system module of the spinal cord electrical stimulation prognostic prediction system for patients with consciousness disorders based on neurovascular coupling, as described in this invention.

[0054] Figure 2 This is a schematic diagram showing the brain network connectivity results at different frequency bands in an embodiment of the present invention;

[0055] Figure 3 This is a schematic diagram showing the effect of neurovascular coupling results in an embodiment of the present invention;

[0056] Figure 4 This is a schematic diagram illustrating the actual display effect of an embodiment of the present invention;

[0057] Figure 5 This is a schematic diagram showing the detailed results of an embodiment of the present invention.

[0058] Figure 6This is a schematic diagram showing the interface of the comparison and prediction section in an embodiment of the present invention. Detailed Implementation

[0059] The following detailed description of embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0060] Please see Figure 1 The spinal cord electrical stimulation prognostic prediction system for patients with consciousness disorders based on neurovascular coupling mainly includes a data acquisition module, an EEG and cerebral blood flow data processing module, and an output display module.

[0061] The data acquisition module utilizes EEG acquisition equipment and transcranial Doppler ultrasound to simultaneously acquire the patient's EEG and cerebral blood flow signals. Basic patient information (age, gender, medical history, duration of coma, CRS-R score, nursing level, etc.) is recorded to confirm the implantation of a spinal cord stimulator (SCS) and that the patient is in the postoperative monitoring period. The patient's scalp is cleaned and conductive ointment is applied. A 19-lead EEG cap is fitted according to the international 10-20 system, ensuring electrode impedance <10kΩ (measured using an impedance meter). The forehead electrode is grounded, and the reference electrode is placed on the earlobe to avoid motion artifacts. The TCD probe (2MHz) is fixed at the patient's temporal window position (2-3cm above the midpoint of the line connecting the eyebrow and ear), and after applying coupling gel, it is secured with an elastic bandage to ensure close contact between the probe and the skin. The EEG acquisition equipment (512Hz sampling rate, EDF format) and the TCD equipment (125Hz sampling rate, TXT format) are connected, and synchronous acquisition is initiated via the parallel communication module, confirming that the timestamp alignment error is <1ms. The collected EEG signals and cerebral blood flow information were saved as EDF and txt files respectively for subsequent analysis.

[0062] The main functions of the EEG and cerebral blood flow data processing module include preprocessing the raw EEG signals (acquired by the data acquisition module); and calculating brain network connectivity and neurovascular coupling through the preprocessed data. The module includes: an EEG data preprocessing unit, a cerebral blood flow data processing unit, a brain network connectivity calculation unit, and a neurovascular coupling calculation unit.

[0063] 1. EEG Data Preprocessing Unit

[0064] EEG data preprocessing in this system involves using the EEGLAB EEG processing toolbox, developed based on MATLAB, to preprocess the raw EEG signals. This includes preprocessing operations such as bandpass filtering, 50Hz notch filtering, interpolation bad derivatives, downsampling, ICA artifact removal, and data segmentation. The steps are as follows:

[0065] Step 1, Bandpass Filtering: A third-order Butterworth bandpass filter with a filtering range of 0.5-40Hz is used to filter the data to remove extremely low and extremely high frequency interference in the EEG signal;

[0066] Step 2, 50Hz notch filtering: mainly to remove 50Hz power frequency interference from the acquisition system;

[0067] Step 3, Interpolation Bad Leads: When the acquired signal of individual leads is severely distorted, the interpolation bad lead function in the EEGLAB toolkit can be used to remove distorted leads.

[0068] Step 4, Downsampling: Brain network connectivity calculations can retain the sampling rate. When calculating neurovascular coupling, the EEG sampling rate needs to be reduced to 125Hz to maintain the synchronization of EEG and cerebral blood flow data.

[0069] Step 5, ICA artifact removal: Use the ICA artifact removal function in the EEGLAB toolkit to remove electrooculography (EOG) and electromyography (EMG) artifacts that may occur during the acquisition process.

[0070] Step 6: Data Segmentation: Brain network calculation requires segmenting the EEG data. The brain network connectivity is calculated by dividing the EEG data into several 5-second segments and averaging them to obtain the average brain network connectivity status for that time segment.

[0071] 2. Cerebral blood flow data processing unit

[0072] Cerebral blood flow data processing involves filtering to obtain the frequency bands of interest from the cerebral blood flow envelope values ​​acquired by TCD and extracting the corresponding phase information.

[0073] 3. Brain network connectivity computing unit

[0074] The weighted pairwise phase coherence index (WPI) is a metric used to measure the phase synchronization between different brain regions in a brain network. Based on EEG time-series data, the degree of phase coupling between different brain regions is calculated to assess the connectivity of the brain network. The calculation steps are as follows:

[0075] Step 1, Phase Extraction: Perform wavelet transform on the time series of each brain region to extract phase information.

[0076] Step 2: Calculate the phase difference: For each lead, calculate their phase difference using the following formula:

[0077] Δφ ij (t)=φ i (t)-φ j (t) (1)

[0078] Where, Δφ ij (t) represents the phase difference between the i-th lead and the j-th lead at time t;

[0079] φ i (t) represents the phase value of the EEG signal in the i-th lead; φ j (t) represents the phase value of the EEG signal in the j-th lead;

[0080] Step 3: Calculate the weighted pairwise phase consistency index: This measures the degree of coupling between different leads by calculating the magnitude of the phase difference. The formula for the weighted pairwise phase consistency index is:

[0081]

[0082] Among them, WPPC ij represents the weighted pairwise phase coherence index between leads i and j; T represents the total number of time points of the signal; exp() is a complex exponential function;

[0083] Step 4: Normalization: Perform normalization to ensure the value falls between 0 and 1, where 0 indicates no phase synchronization and 1 indicates complete synchronization. Normalization can be performed using the following formula:

[0084]

[0085] 4. Neurovascular Coupling Calculation Unit

[0086] Neurovascular coupling calculation, specifically phase-amplitude coupling (PAC), is a technique used to analyze the interactions between different frequency bands in neural signals, particularly exploring the coupling relationship between low-frequency signals (such as theta waves and alpha waves) and high-frequency signals (such as gamma waves and beta waves). This invention applies PAC to calculate neurovascular coupling, thereby obtaining the coupling relationship between low-frequency blood flow and high-frequency neuronal activity. The calculation steps are as follows:

[0087] Step 1: Extraction of blood flow phase information: Extract the phase information of the frequency band of interest in the blood flow signal using Hilbert transform.

[0088]

[0089] in, The instantaneous phase value at time t is represented by arg(); arg() calculates the phase angle of the blood flow signal; Hilbert() performs a Hilbert transform on the input signal to generate an analytic signal; CBFv low (t) Low-frequency cerebral blood flow velocity signal at time t;

[0090] Step 2, EEG Amplitude Information Extraction: Amplitude information of the frequency band of interest in the EEG signal is extracted using Hilbert transform.

[0091] AEEG(t)=|Hilbert(EEGhigh (t))| (5)

[0092] Where AEEG(t) represents the instantaneous amplitude of high-frequency EEG at time t; EEG high (t) represents the high-frequency EEG signal at time t;

[0093] Step 3: Calculate the average amplitude in the phase:

[0094]

[0095] Where Z(t) represents the neurovascular coupling strength value at time t; It is a complex exponential function that combines blood flow phase information and EEG amplitude information; This represents the instantaneous blood flow phase at time t;

[0096] S34. Calculate the coupling value: Calculate the average phase and magnitude of the complex exponential function and take the modulus. The calculation formula is as follows:

[0097]

[0098] Wherein, PAC represents the neurovascular coupling strength value; This represents the average of the complex exponential function, where T represents the total number of time points.

[0099] The display module presents the processed feature parameters in an intuitive and easy-to-understand way, making it convenient for doctors to receive guidance on medical plans.

[0100] Brain network connectivity is a numerical value ranging from 0 to 1. Calculating the brain network connectivity values ​​across 19 leads in five frequency bands of EEG (Delta, Theta, Alpha, Beta, Gamma) will yield five 19×19 matrices. Figure 2 The connection status is displayed in a concise and clear manner.

[0101] The connecting lines between the electrodes in the diagram represent stronger connections; the greener and thicker the connecting line, the stronger the connection. Doctors can predict a patient's prognosis by comparing the connectivity of brain networks at different frequency bands before and after surgery. Existing data shows that when the connectivity of the frontal lobe brain networks in the Theta, Alpha, Beta, and Gamma frequency bands is enhanced, the patient's prognosis improves, with a higher probability of awakening.

[0102] Neurovascular coupling is a numerical value between 0 and 5. A 7×5 array can be obtained by dividing the brain into regions and different frequency bands of brain electrical activity. Adjusting the brain differentiation can output seven parts: left frontal lobe, right frontal lobe, left temporal lobe, right temporal lobe, central region, parietal lobe, and occipital lobe; or it can output the frontal lobe, temporal lobe, central region, parietal lobe, and occipital lobe. The output effect of a single frequency band for different regions in a single dataset is shown below. Figure 3 As shown.

[0103] Doctors can predict a patient's prognosis by comparing changes in neurovascular coupling values ​​before and after surgery. Existing data suggests that increased neurovascular coupling in the frontal lobe correlates with a better prognosis and a higher probability of recovery.

[0104] The final display module effect is as follows: Figure 4 As shown, it is mainly divided into three parts: patient information, detailed results, and comparison and prediction (see detailed function demonstration). Figure 5 , Figure 6 Clicking on any patient's data in the patient database allows you to view their postoperative prediction information. The patient information section includes the patient's age, gender, medical history, coma duration, Coma Recovery Scale-Revised (CRS-R) score, and nursing care level. The detailed results section displays detailed analysis results of different frequency bands of a single EEG data point. Clicking the toggle button at the top allows you to view data from different time points, such as preoperative and postoperative information. Clicking the toggle button at the bottom allows you to view information from different frequency bands of a single data point, such as Delta, Theta, Alpha, Beta, and Gamma. The comparative prediction section compares data collected at different time points to provide predictions of the patient's prognosis. To more intuitively illustrate changes in neurovascular coupling, preoperative and postoperative results are normalized to highlight improvements in different brain regions.

[0105] Therefore, this invention employs the aforementioned spinal cord electrical stimulation prognostic prediction system for patients with DOC based on neurovascular coupling, and innovatively proposes a method for predicting the prognosis of DOC patients after SCS implantation. By collecting patients' EEG and cerebral blood flow data, the clinical prognosis can be accurately predicted. This solves the problem of prognostic uncertainty for DOC patients after neuromodulation therapy, helps doctors provide timely and personalized treatment plans, shortens the patient's treatment cycle, and is expected to achieve considerable social and economic benefits.

[0106] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A spinal cord electrical stimulation prognostic prediction system for patients with consciousness disorders based on neurovascular coupling, characterized in that, include: The data acquisition module uses EEG acquisition equipment and transcranial Doppler ultrasound to simultaneously acquire the patient's EEG and cerebral blood flow signals; The EEG and cerebral blood flow data processing module preprocesses the raw EEG signals acquired by the data acquisition module and calculates brain network connectivity and neurovascular coupling based on the preprocessed data; specifically, it includes an EEG data preprocessing unit, a cerebral blood flow data processing unit, a brain network connectivity calculation unit, and a neurovascular coupling calculation unit. The neurovascular coupling calculation unit applies phase-amplitude coupling to calculate neurovascular coupling, obtaining the coupling relationship between low-frequency blood flow and high-frequency neuronal activity. The calculation steps are as follows: S31. Blood Flow Phase Information Extraction: The phase information of the frequency band of interest in the blood flow signal is extracted using Hilbert transform. The calculation formula is as follows: （4） in, Indicates time The instantaneous phase value; This indicates the calculation of the blood flow signal phase angle; Perform a Hilbert transform on the input signal to generate an analytic signal; time Low-frequency cerebral blood flow velocity signals; S32. EEG Amplitude Information Extraction: The amplitude information of the frequency band of interest in the EEG signal is extracted using Hilbert transform. The calculation formula is as follows: （5） in, Indicates time The instantaneous amplitude of high-frequency brainwaves; Indicates time High-frequency brainwave signals; S33. Calculate the average amplitude in the phase using the following formula: （6） in, Indicates time The neurovascular coupling strength value; It is a complex exponential function that combines blood flow phase information and EEG amplitude information; Indicates time The instantaneous blood flow phase; S34. Calculate the coupling value: Calculate the average phase and magnitude of the complex exponential function and take the modulus. The calculation formula is as follows: （7） in, Indicates the neurovascular coupling strength value; This indicates taking the average of the complex exponential function. T Indicates the total number of time points; The output display module displays the feature parameters obtained after processing by the EEG and cerebral blood flow data processing modules, including three parts: patient information, detailed results, and comparative prediction. The patient information section includes the patient's age, gender, medical history, coma duration, coma recovery scale score, and nursing care level. The detailed results section allows users to view detailed analysis results of different frequency bands of EEG data from a single patient sample. The comparative prediction section provides a prediction of the patient's prognosis by comparing data collected at different time points.

2. The spinal cord electrical stimulation prognostic prediction system for patients with impaired consciousness based on neurovascular coupling according to claim 1, characterized in that, When the data acquisition module is working, the transcranial Doppler ultrasound probe is fixed in the temporal region, usually at the midpoint of the line connecting the eyebrow and the ear; the sampling rate is set to 125Hz; the EEG cap uses a 19-lead EEG acquisition system designed by the international 10-20 system; standard Ag / Agcl electrodes are used, the sampling rate is set to 512Hz, the forehead is grounded, and the impedance between the scalp and the electrodes is kept below 10KΩ; the acquired EEG signals and cerebral blood flow information are saved as EDF and txt files respectively.

3. The spinal cord electrical stimulation prognostic prediction system for patients with impaired consciousness based on neurovascular coupling according to claim 1, characterized in that, The EEG data preprocessing unit uses the EEGLAB EEG processing toolbox developed based on MATLAB to preprocess the raw EEG signals; Includes the following steps: S11. Bandpass filtering: A third-order Butterworth bandpass filter with a filtering range of 0.5-40Hz is used to filter the data and remove extremely low and extremely high frequency interference in the EEG signal. S12, 50Hz notch filter: Removes 50Hz power frequency interference from the acquisition system; S13. Interpolation Bad Lead: Use the interpolation bad lead function in the EEGLAB toolkit to eliminate distorted leads; S14. Downsampling: When calculating neurovascular coupling, the EEG sampling rate is reduced to 125Hz to maintain the synchronization of EEG and cerebral blood flow data. S15, ICA artifact removal: Remove electrooculography and electromyography artifacts that occur during the acquisition process using the ICA artifact removal function in the EEGLAB toolkit; S16. Data Segmentation: Divide the EEG data into several 5-second segments, calculate the brain network connectivity for each segment, and take the average value as the average brain network connectivity state for that time segment.

4. The spinal cord electrical stimulation prognostic prediction system for patients with impaired consciousness based on neurovascular coupling according to claim 1, characterized in that, The cerebral blood flow data processing module obtains the frequency band of interest from the cerebral blood flow envelope values ​​acquired by the TCD through filtering, and extracts the corresponding phase information.

5. The spinal cord electrical stimulation prognostic prediction system for patients with impaired consciousness based on neurovascular coupling according to claim 1, characterized in that, The connectivity of brain networks is calculated based on the weighted pairwise phase consistency index. The calculation steps are as follows: S21. Phase Extraction: Perform wavelet transform on the time series of each brain region to extract phase information; S22. Calculate the phase difference: For each lead, calculate its phase difference using the following formula: （1） in, Indicates the first The first lead and the first Each lead at time The phase difference; Indicates the first Phase values ​​of the EEG signals in each lead; Indicates the first Phase values ​​of the EEG signals in each lead; S23. Calculate the weighted pairwise phase consistency index: Calculate the magnitude of the phase difference using the following formula: （2） in, Indicates the first With the Weighted pairwise phase coherence index of leads; Indicates the total number of time points of the signal; It belongs to the category of complex exponential functions; S24. Normalization: Perform normalization processing, the formula is: （3） in, This represents the weighted pairwise phase consistency index after normalization, with an output value in [0, 1].

Images