Chronic lumbago analgesic function evaluation platform and method based on biological signals

Abstract

The invention relates to the technical field of medical instruments, and discloses a chronic lumbago analgesia function evaluation platform and method based on biological signals, and the platform comprises a multi-mode biological signal synchronous collection system which is used for synchronously collecting electroencephalogram, galvanic skin and blood oxygen signals, an electroencephalogram cap, a galvanic skin sensor, an oximeter and a data acquisition card for coordinating data acquisition and transmission are arranged in the multi-mode biological signal synchronous acquisition system, and a hardware triggering or software synchronization algorithm is provided to ensure signal synchronization. According to the biological signal-based chronic lumbago analgesic function evaluation platform and method, through a multi-modal biological signal synchronous acquisition system, an electroencephalogram cap, a galvanic skin sensor and an oximeter are used for respectively acquiring electroencephalogram, galvanic skin and blood oxygen signals, and a multi-modal acquisition mode is adopted; the biological signals related to chronic lumbago are comprehensively acquired, and compared with traditional single signal acquisition, comprehensive data are provided, and a foundation is laid for accurate evaluation.

Description

Technical Field

[0001] The present invention relates to the technical field of medical devices, and specifically provides a chronic low back pain analgesic function evaluation platform and method based on biological signals. Background Art

[0002] Chronic low back pain is an extremely common and intractable health problem, showing a high incidence globally. According to relevant data of the World Health Organization, about 15% of the global population is suffering from chronic low back pain. The low back pain of patients not only seriously affects the quality of personal daily life, causing patients to suffer a lot in daily activities such as walking, sitting, and sleeping, but also imposes a heavy burden on social medical resources. From the perspective of the pathogenesis, chronic low back pain involves complex interactions among multiple systems such as nerves, muscles, and bones. The complex pathogenesis makes traditional evaluation methods have some deficiencies in analyzing the condition. For example: Common visual analogue scale and Oswestry disability index mainly rely on the subjective description of patients, are easily interfered by factors such as psychology and emotion, resulting in large deviations in evaluation results. Moreover, existing evaluation methods only rely on a single biological signal for evaluation, unable to comprehensively reflect the complex activity changes of pain modulation neural pathways, difficult to provide systematic treatment guidance, and traditional methods are mostly static evaluations, unable to track the dynamic changes of the pain modulation mechanism in patients during the treatment process in real time.

[0003] In view of the above problems, there is an urgent need for innovative design based on the original low back pain evaluation methods. Summary of the Invention

[0004] The purpose of the present invention is to provide a chronic low back pain analgesic function evaluation platform and method based on biological signals, so as to solve the problems raised in the above background art that the existing evaluations mainly rely on the subjective description of patients, are easily interfered by factors such as psychology and emotion, resulting in large deviations in evaluation results, and existing evaluation methods only rely on a single biological signal for evaluation, unable to comprehensively reflect the complex activity changes of pain modulation neural pathways, difficult to provide systematic treatment guidance, and traditional methods are mostly static evaluations, unable to track the dynamic changes of the pain modulation mechanism in patients during the treatment process in real time.

[0005] To achieve the above purpose, the present invention provides the following technical solution: A chronic low back pain analgesic function evaluation platform based on biological signals, comprising: A multimodal biological signal synchronous acquisition system for synchronously acquiring electroencephalogram, electrodermal activity, and blood oxygen signals. The electroencephalogram cap, electrodermal sensors, blood oxygen meter, and data acquisition card for coordinating data acquisition and transmission are arranged in the multimodal biological signal synchronous acquisition system, and a hardware trigger or software synchronization algorithm is provided to ensure signal synchronization; The signal processing and feature extraction module preprocesses and extracts features from the collected signals, including calculating the alpha wave power spectral density and beta wave synchrony index for the electroencephalogram (EEG) signals collected by the EEG cap, analyzing the amplitude and frequency of the skin conductance response for the skin conductance signals collected by the skin conductance signal acquisition device, and extracting the concentration change characteristics of oxyhemoglobin and reduced hemoglobin for the blood oxygen signals collected by the oximeter. The pain regulation neural pathway activity quantification model constructs a pain regulation pathway activity index model based on a Bayesian network, and quantifies the activities of the descending inhibitory pathway, autonomic nerve pathway, and neurovascular pathway by integrating the EEG, skin conductance, and blood oxygen signal features collected in the signal processing and feature extraction module. The analgesic function evaluation platform and intervention system real-time displays the signal waveforms and the changing trend of the PAI value, generates a three-dimensional heat map, and automatically triggers and adjusts the transcutaneous electrical stimulation analgesic intervention measures according to the PAI value, and conducts neurofeedback training in combination with VR technology.

[0006] Adopting the above technical solution, through the multimodal bio-signal synchronous acquisition system, integrating the EEG, skin conductance, and blood oxygen signal acquisition devices, using hardware or software synchronous algorithms to ensure signal synchronization, and providing comprehensive bio-signal data.

[0007] Preferably, the acquisition devices in the multimodal bio-signal synchronous acquisition system include an EEG signal acquisition device, a skin conductance signal acquisition device, and a blood oxygen signal acquisition device, and the data collected by the multimodal bio-signal synchronous acquisition system is centrally collected by a data acquisition card.

[0008] Adopting the above technical solution, ensuring that various bio-signals can be effectively converged, providing a basis for subsequent data processing and analysis, and improving the systematicness and efficiency of data collection.

[0009] Preferably, the EEG signal acquisition device includes an EEG cap covering specific brain regions and EEG electrode patches, and the EEG electrode patches of the EEG cap are made of silver chloride, and the electrodes cover the prefrontal lobe, central region, parietal lobe and other brain regions related to pain perception and regulation.

[0010] Adopting the above technical solution, the EEG cap in the EEG signal acquisition device uses silver chloride electrode patches to cover the key brain regions, can accurately collect the EEG signals related to pain perception and regulation, improve the quality and pertinence of EEG signal acquisition, and enhance the accuracy of evaluation.

[0011] Preferably, the skin conductance signal acquisition device includes: a skin conductance sensor covering the epidermis and skin conductance electrode patches, and the skin conductance electrode patches of the skin conductance sensor are disposable adhesive type, and are placed on the palm or finger to measure the change of skin conductivity.

[0012] With the above technical solution, the disposable adhesive electrode of the skin electrical signal acquisition device is placed on the palm or finger part, which can accurately measure the change of skin conductivity, ensure the stability and reliability of skin electrical signal acquisition, and provide effective data for the evaluation of pain-related autonomic nerve responses.

[0013] Preferably, the data collected by the data acquisition card is uploaded to the data preprocessing module, and the data preprocessing module processes the data, and the data processed by the data preprocessing module is uploaded to the feature extraction unit to extract electroencephalogram and skin electrical features.

[0014] With the above technical solution, the data collected by the data acquisition card is processed by the data preprocessing module and then uploaded to the feature extraction unit to extract electroencephalogram and skin electrical features, forming an orderly data processing flow, ensuring data quality, improving the accuracy and efficiency of feature extraction, and providing reliable data for subsequent analysis.

[0015] Preferably, the data acquisition card, the data preprocessing module and the feature extraction unit constitute a signal processing and feature extraction module, and the signal processing and feature extraction module uploads data to the pain modulation neural pathway activity quantification model, the analgesic function evaluation platform and the intervention system.

[0016] With the above technical solution, the signal processing and feature extraction module is composed of multiple parts, and it uploads the processed data to the key model and system, ensuring the smoothness of data circulation, providing support for the quantification of pain modulation neural pathway activity, the evaluation and intervention of analgesic function, and promoting the efficient operation of the entire evaluation platform.

[0017] A method for evaluating the analgesic function of chronic low back pain based on bio-signals includes: Step 1: The electroencephalogram cap in the multi-modal bio-signal synchronous acquisition system synchronously acquires the electroencephalogram signals of chronic low back pain patients, the skin electrical sensor is used to acquire the skin electrical signals of the patients, and the finger clip type blood oxygen meter is used to acquire the blood oxygen signals of the patients; Step 2: Connect the electroencephalogram cap, the skin electrical sensor and the blood oxygen meter through the data acquisition card, and use the hardware trigger or software synchronization algorithm to ensure that the starting time of each signal acquisition is the same, and compensate for the time difference caused by transmission delay, etc., to achieve precise synchronization; Step 3: Use independent component analysis to remove artifacts such as electrooculogram and electromyogram from the collected electroencephalogram signals, filter the skin electrical signals to eliminate high-frequency noise interference, remove abnormal fluctuations from the blood oxygen signals through a smoothing algorithm, and then extract features for the electroencephalogram signals, skin electrical signals and blood oxygen signals; Step 4: Build a pain modulation pathway activity index model based on the Bayesian network, and comprehensively quantify the activities of the descending inhibition pathway, the autonomic nerve pathway and the neurovascular pathway according to the electroencephalogram, skin electrical and blood oxygen signal features; Step 5: Generate a pain regulation pathway activity assessment report based on the PAI value calculated by the PAI model. Develop a personalized analgesic treatment plan according to the assessment report. During the treatment process, regularly repeat the above steps of biological signal acquisition, processing, quantification, and evaluation, and monitor and adjust the treatment measures in real time according to the change of the PAI value.

[0018] By adopting the above technical solution, dynamic monitoring and adjustment are carried out during the treatment, providing a complete and dynamic assessment and treatment process for patients with chronic low back pain, and improving the treatment effect.

[0019] Preferably, in the multi-modal biological signal acquisition step, the sampling frequency of the electroencephalogram signal is set to 500Hz - 1000Hz, and the sampling frequency of the galvanic skin response signal is set to 100Hz - 200Hz.

[0020] By adopting the above technical solution, setting an appropriate sampling frequency can accurately capture signal characteristics, improve the accuracy of signal acquisition, provide high-quality data for subsequent analysis, and enhance the scientific nature of the assessment.

[0021] Preferably, in step 4, the Bayesian network structure is learned from the collected data through data mining algorithms such as the K2 algorithm and the expectation-maximization algorithm, and the parameters in the network are estimated by methods such as maximum likelihood estimation.

[0022] By adopting the above technical solution, the Bayesian network structure is constructed and the parameters are estimated through specific algorithms, enabling the pain regulation pathway activity index model to more accurately synthesize biological signal characteristics, improving the accuracy and reliability of the assessment, and providing a more accurate basis for clinical diagnosis and treatment.

[0023] Preferably, in step 5, formulating a personalized analgesic treatment plan includes automatically triggering transcutaneous electrical stimulation analgesic intervention measures when the PAI value is lower than the set threshold, dynamically adjusting the stimulation intensity according to the amplitude of the galvanic skin response signal SCR, and at the same time combining virtual reality technology to enhance the patient's prefrontal inhibitory ability to pain through neurofeedback training.

[0024] By adopting the above technical solution, automatically triggering transcutaneous electrical stimulation based on the PAI value and dynamically adjusting the intensity, combined with VR neurofeedback training, realizes precise and personalized intervention, effectively improves the analgesic treatment effect, reduces the patient's pain, and improves the quality of medical services.

[0025] Compared with the prior art, the beneficial effects of the present invention are as follows: The chronic low back pain analgesic function assessment platform and method based on biological signals: 1. The platform and method in the present invention use a multi-modal biological signal synchronous acquisition system to collect electroencephalogram (EEG), skin conductance (SC), and blood oxygen saturation (SpO₂) signals respectively through an EEG cap, an SC sensor, and a pulse oximeter. The electrodes of the EEG cap are made of silver chloride, covering the brain regions related to pain perception and regulation and having adjustable tightness to accurately collect EEG signals. The disposable adhesive electrode pads of the SC sensor are placed on the palm or finger parts to accurately measure the change in skin conductivity. The finger clip pulse oximeter monitors the SpO₂ in real-time and continuously, and the sampling frequency can be adjusted as needed. By adopting a multi-modal acquisition method, biological signals related to chronic low back pain are comprehensively obtained, providing more comprehensive data compared with traditional single-signal acquisition, laying a foundation for accurate evaluation; 2. Further, by using hardware triggering or software synchronization algorithms, a data acquisition card is used to ensure that the start times of each signal acquisition are consistent and compensate for time differences, achieving precise synchronization and ensuring the accurate correspondence of multi-source signals in the time dimension, making subsequent analysis more reliable. In the signal processing stage, independent component analysis is used to remove artifacts from EEG signals, filtering and denoising are performed on SC signals, and smoothing processing is carried out on SpO₂ signals, effectively improving the signal quality and reducing the influence of interference factors on the analysis results. Accurate and reliable signal data provides a solid guarantee for subsequent feature extraction and quantification of neural pathway activity. Along with the signal processing and feature extraction module composed of a data acquisition card, a data preprocessing module, and a feature extraction unit, the processed data is uploaded to the pain regulation neural pathway activity quantification model, the analgesic function evaluation platform, and the intervention system. The orderly data management and analysis process improve the data processing efficiency and ensure the efficient operation of the evaluation platform and method, providing strong support for clinical applications; 3. The analgesic function evaluation platform and intervention system display the signal waveforms and the change trends of PAI values in real-time and generate a three-dimensional heat map, intuitively showing the coordinated changes in brain region activation, autonomic nerve responses, and tissue oxygenation. Doctors can comprehensively and intuitively understand the patient's condition based on this. According to the PAI value, when it is lower than the set threshold, analgesic intervention measures such as transcutaneous electrical stimulation are automatically triggered, and the stimulation intensity is dynamically adjusted in combination with the amplitude of the skin conductance response of the SC signal. At the same time, VR technology is combined for neurofeedback training. Through intelligent and personalized intervention methods, the treatment plan is accurately adjusted according to the patient's real-time pain regulation state, improving the analgesic treatment effect, reducing the patient's pain, and enhancing the quality and efficiency of medical services. Description of the Drawings

[0026] Figure 1 It is a schematic diagram of the module process steps of the chronic low back pain analgesic function evaluation platform of the present invention; Figure 2 It is a schematic diagram of the biological information acquisition process of the present invention; Figure 3 It is a schematic diagram of the biological signal data processing process of the present invention; Figure 4This is a schematic diagram of the steps for evaluating the analgesic function of chronic low back pain in the present invention.

[0027] In the figure: 1. Electroencephalogram (EEG) signal acquisition device; 101. EEG cap; 102. EEG electrode patch; 2. Electrodermal (ED) signal acquisition device; 201. ED sensor; 202. ED electrode patch; 3. Blood oxygen signal acquisition device; 301. Pulse oximeter; 4. Data acquisition card; 5. Data preprocessing module; 6. Feature extraction unit; 7. Quantification model of pain-modulating neural pathway activity; 8. Analgesic function evaluation platform and intervention system. Specific implementation mode

[0028] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0029] Please refer to Figures 1 - 4 , the present invention provides a technical solution: an analgesic function evaluation platform and method for chronic low back pain based on biological signals, including an electroencephalogram (EEG) signal acquisition device 1, an EEG cap 101, an EEG electrode patch 102, an electrodermal (ED) signal acquisition device 2, an ED sensor 201, an ED electrode patch 202, a blood oxygen signal acquisition device 3, a pulse oximeter 301, a data acquisition card 4, a data preprocessing module 5, a feature extraction unit 6, a quantification model of pain-modulating neural pathway activity 7, and an analgesic function evaluation platform and intervention system 8; Among them, the construction of the analgesic function evaluation platform for chronic low back pain based on biological signals is sequentially constructed by four units: a multimodal biological signal synchronous acquisition system, a signal processing and feature extraction module, a quantification model of pain-modulating neural pathway activity 7, and an analgesic function evaluation platform and intervention system 8; The multimodal biological signal synchronous acquisition system is used to synchronously acquire EEG, ED, and blood oxygen signals. The EEG cap 101, ED sensor 201, pulse oximeter 301, and a data acquisition card 4 for coordinating data acquisition and transmission are set in the multimodal biological signal synchronous acquisition system, and it has a hardware trigger or software synchronization algorithm to ensure signal synchronization; The acquisition devices in the multimodal biological signal synchronous acquisition system include an electroencephalogram (EEG) signal acquisition device 1, an electrodermal (ED) signal acquisition device 2, and a blood oxygen signal acquisition device 3, and the data acquired by the multimodal biological signal synchronous acquisition system are centrally collected by the data acquisition card 4 The electroencephalogram (EEG) signal acquisition device 1 includes an EEG cap 101 covering specific brain regions and EEG electrode patches 102, and the EEG electrode patches 102 of the EEG cap 101 are made of silver chloride, and the electrodes cover brain regions related to pain perception and regulation such as the prefrontal lobe, central region, and parietal lobe; The galvanic skin response (GSR) signal acquisition device 2 includes: a GSR sensor 201 covering the epidermis and a GSR electrode patch 202. The GSR electrode patch 202 of the GSR sensor 201 is a disposable adhesive type, which is placed on the palm or finger to measure the change in skin conductivity; During actual installation and use, an EGI 128-channel EEG cap 101 is selected as the EEG signal acquisition device 1. The EEG electrode patches 102 are accurately covered on the prefrontal lobe, central region, parietal lobe and brain regions related to pain perception and regulation. The tightness of the EEG cap 101 is adjusted to ensure that the silver chloride electrode patches are closely attached to the scalp to reduce contact impedance, ensuring that the collected EEG signals are stable and accurate. The sampling frequency is set at 500 Hz - 1000 Hz to capture the details of rapidly changing EEG activities, such as alpha waves, beta waves and other EEG rhythms related to pain; The GSR100C GSR module is used as the GSR signal acquisition device 2. After cleaning the skin, the disposable adhesive GSR electrode patch 202 is pasted on the palm or finger of the patient. The sampling frequency is set at 100 Hz - 200 Hz, which can effectively record the amplitude and frequency changes of the skin conductance response, providing data support for evaluating the activity of the autonomic nerve pathway. The blood oxygen signal acquisition device 3, that is, the pulse oximeter 301 of the CMS50D+ finger clip type, is used to collect blood oxygen signals. The finger clip is correctly clipped on the patient's finger. The sampling frequency can be adjusted between 1 - 10 Hz according to actual needs. During a pain attack, the sampling frequency is increased to 5 - 10 Hz to timely capture the instantaneous changes in blood oxygen saturation, reflecting the local tissue oxygenation status and the function of the neurovascular pathway; Combined with the Figures 1 - 3 shown in the accompanying drawings of the specification, the EEG cap 101, the GSR sensor 201 and the pulse oximeter 301 are connected through a data acquisition card 4. During actual use, according to actual installation needs, hardware trigger synchronization is adopted, and a synchronous trigger line is connected between the acquisition devices to ensure that the acquisition start times of the devices are the same; When it is difficult for hardware to support time synchronization, a software synchronization algorithm is adopted. In the data acquisition software LabVIEW, the collected data is time-calibrated based on the timestamp information of each device to compensate for the time differences caused by transmission delays, etc. During actual use, the EEG signal sampling frequency is set at 500 Hz - 1000 Hz, the GSR signal sampling frequency is set at 100 Hz - 200 Hz, and the blood oxygen signal sampling frequency is adjusted between 1 - 10 Hz according to requirements; The signal processing and feature extraction module preprocesses and extracts features from the collected signals, including calculating the alpha wave power spectral density and beta wave synchrony index of the EEG signals collected by the EEG cap 101, analyzing the amplitude and frequency of the skin conductance response of the GSR signals collected by the GSR signal acquisition device 2, and extracting the characteristics of the changes in the concentrations of oxyhemoglobin and reduced hemoglobin of the blood oxygen signals collected by the pulse oximeter 301; Combined with the accompanying drawings of the specification Figures 1 - 3 As shown, in the data acquisition software, the collected electroencephalogram (EEG) signals are processed using the independent component analysis algorithm to remove the artifacts of electrooculogram (EOG) and electromyogram (EMG), and the galvanic skin response (GSR) signals are filtered. For example, a Butterworth low-pass filter is used to eliminate high-frequency noise interference. For the blood oxygenation level-dependent (BOLD) signals, smoothing algorithms such as moving average are used to remove abnormal fluctuations; Then, for the EEG signals, algorithms are written to calculate characteristic parameters such as the power spectral density of the alpha wave, i.e., 8 - 13 Hz, and the beta wave synchrony index, i.e., 13 - 30 Hz. For the GSR signals, the amplitude and frequency of the skin conductance response are extracted. From the BOLD signals, the changes in the concentrations of oxyhemoglobin and deoxyhemoglobin are calculated, and their ratio is obtained. Through the feature extraction algorithm, it is integrated into the data acquisition software to automatically perform feature extraction after data acquisition; The pain regulation neural pathway activity quantification model 7 constructs a pain regulation pathway activity index model based on the Bayesian network, and comprehensively quantifies the activities of the descending inhibitory pathway, autonomic nerve pathway, and neurovascular pathway using the EEG, GSR, and BOLD signal features collected in the signal processing and feature extraction module; When the pain regulation neural pathway activity quantification model 7 is used, a large amount of EEG, GSR, and BOLD signal data of chronic low back pain patients in different pain states are collected, and the clinical information of the patients, such as VAS scores, pain attack frequencies, etc., is recorded at the same time. According to the clinical evaluation, the data is labeled to clarify the pain regulation state corresponding to each data sample. Further, tools such as the pgmpy library in Python are used, and through data mining algorithms such as the K2 algorithm and the expectation-maximization algorithm, the structure of the Bayesian network is learned from the collected data, and methods such as maximum likelihood estimation are used to estimate the parameters in the network to construct a pain regulation pathway activity index model. This model comprehensively uses the EEG, GSR, and BOLD signal features to quantify the activities of the descending inhibitory pathway, autonomic nerve pathway, and neurovascular pathway; The analgesic function evaluation platform and intervention system 8 can display the signal waveforms and the changing trend of the PAI value in real time, generate a three-dimensional heat map, and automatically trigger and adjust the transcutaneous electrical stimulation analgesic intervention measures according to the PAI value, and conduct neurofeedback training in combination with VR technology; When the analgesic function evaluation platform and intervention system 8 is used, a software platform integrating data acquisition, processing, analysis, and result display is developed using a graphical interface development framework such as Qt. In the software interface, a real-time signal waveform display area is set to display the EEG, GSR, and BOLD signal waveforms in real time. Further, a PAI value and related parameter display area is set to present the calculated pain regulation pathway activity index and the key characteristic parameters of each signal. A data storage and management area is set to facilitate the classified storage, query, and export of the collected data; an operation control area is set, which includes function buttons such as start acquisition, stop acquisition, and parameter adjustment; By setting the threshold value of the PAI value, when the PAI value is lower than the set threshold, the galvanic skin stimulation device connected to the patient is automatically triggered for analgesic intervention, and the stimulation intensity is dynamically adjusted according to the amplitude of the skin conductance signal SCR. At the same time, the development interface of the VR device is integrated, and neurofeedback training is carried out in combination with VR technology. By monitoring the activity changes in the prefrontal cortex of the electroencephalogram signal, the VR training content is adjusted in real time to enhance the patient's ability to inhibit pain in the prefrontal lobe; The data collected by the data acquisition card 4 is uploaded to the data preprocessing module 5, and the data preprocessing module 5 processes the data. And the data processed by the data preprocessing module 5 is uploaded to the feature extraction unit 6 to extract electroencephalogram and galvanic skin features. The data acquisition card 4, the data preprocessing module 5 and the feature extraction unit 6 constitute the signal processing and feature extraction module, and the signal processing and feature extraction module uploads data to the pain regulation neural pathway activity quantification model 7 and the analgesic function evaluation platform and intervention system 8; Combined with the attached drawings of the specification Figures 1 - 4 As shown, according to the multimodal biosignal synchronous acquisition method of the evaluation platform, the electroencephalogram cap 101 synchronously acquires the electroencephalogram signal of chronic low back pain patients, the galvanic skin sensor 201 acquires the galvanic skin signal, and the pulse oximeter 301 acquires the blood oxygen signal to provide raw data for subsequent analysis. And the sampling frequency of the electroencephalogram signal is usually set to 500Hz - 1000Hz, the sampling frequency of the galvanic skin signal is set to 100Hz - 200Hz. Each acquisition device is connected through the data acquisition card 4, and the hardware trigger or software synchronization algorithm is used to ensure that the acquisition start times of the electroencephalogram, galvanic skin, and blood oxygen signals are the same, and compensate for the time difference caused by transmission delay, etc., so that the three signals are accurately corresponding in time, ensuring the effectiveness and analyzability of the data. The three types of collected signals are respectively preprocessed to remove interference and abnormal fluctuations, and then their respective key features are extracted. For the PAI model constructed based on the Bayesian network, the network structure is learned through data mining algorithms such as the K2 algorithm and the maximum expectation algorithm, and the network parameters are estimated by methods such as maximum likelihood estimation. After that, a pain regulation pathway activity evaluation report is generated according to the PAI value calculated by the PAI model, and the doctor formulates a personalized analgesic treatment plan based on the report; Combined with the attached drawings of the specification Figures 1 - 4 As shown, in the actual usage method, the patient should maintain a quiet rest state 15 - 20 minutes before the detection, avoid strenuous exercise and drinking stimulating beverages. After the medical staff cleans the patient's skin, the electroencephalogram electrode patches 102 and galvanic skin electrode patches 202 are pasted on the corresponding parts, and the finger clip type pulse oximeter 301 is worn for device calibration, including the electrode impedance test of the electroencephalogram cap 101 to ensure that the impedance is lower than 5kΩ - 10kΩ, the zero calibration of the galvanic skin sensor 201, and the calibration of the pulse oximeter 301 with a standard blood oxygen concentration simulator; After that, start the data acquisition software to begin synchronous acquisition of EEG, GSR, and blood oxygen signals. After the acquisition is completed, the software automatically processes and analyzes the signals, calculates the PAI value, and generates a pain regulation pathway activity assessment report based on the PAI value calculated according to the PAI model. The report content includes the patient's basic information, the collected biological signal characteristic data, the PAI value, and the analysis of the corresponding pain regulation neural pathway activity status. The PAI value range is set from 0 to 100. The lower the value, the lower the activity of the pain regulation neural pathway and the worse the pain control ability. The higher the value, the more normal the neural pathway activity and the better the pain regulation function. Generate a three-dimensional heat map to visually display the coordinated changes in brain region activation, autonomic nerve response, and tissue oxygenation, providing comprehensive and intuitive condition assessment information for doctors. The model constructed based on the Bayesian network constitutes the PAI model. Doctors formulate personalized analgesic treatment plans based on the assessment report. When the PAI value is lower than the set threshold, analgesic intervention measures such as transcutaneous electrical stimulation are automatically triggered. In the actual treatment process, the NeuroCareTENS therapeutic instrument is selected, and the stimulation intensity is dynamically adjusted according to the GSR signal SCR amplitude. For every 1 μS increase in the SCR amplitude, the stimulation intensity is reduced by 5% to achieve precise analgesia. At the same time, combined with virtual reality technology, such as using the VR device of HTC Vive, the inhibitory ability of the patient's prefrontal lobe to pain is enhanced through neurofeedback training. During the VR training process, the activity changes of the prefrontal cortex in the EEG signal are monitored in real time, and the VR training content is adjusted in real time according to the patient's neural response, including but not limited to changing the difficulty of the virtual scene, task requirements, etc., to guide the patient to enhance the self-regulation ability to pain. During the treatment process, the above-mentioned biological signal acquisition, processing, quantification, and assessment steps are repeated regularly, and the changes in the activity of the patient's pain regulation neural pathway are monitored in real time according to the changes in the PAI value, and the treatment plan is adjusted in time, such as adjusting the TENS stimulation parameters, VR training plan, drug dosage, etc., to improve the treatment effect and promote the patient's recovery.

[0030] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention.

Claims

1. A chronic low back pain analgesia function assessment platform based on biological signals, characterized in that: include: A multimodal biological signal synchronous acquisition system, used for synchronously acquiring EEG, electrodermal and blood oxygen signals, wherein the multimodal biological signal synchronous acquisition system includes an EEG cap (101), an electrodermal sensor (201), a blood oxygen meter (301) and a data acquisition card (4) for coordinating data acquisition and transmission, and is provided with a hardware trigger or software synchronization algorithm to ensure signal synchronization; The signal processing and feature extraction module performs preprocessing and feature extraction on the collected signals, including calculating the alpha wave power spectrum density and beta wave synchronization index of the electroencephalogram signal collected by the electroencephalogram cap (101), analyzing the skin conductance response amplitude and frequency of the electrodermal signal collected by the electrodermal signal collection device (2), and extracting the concentration change characteristics of oxygenated hemoglobin and reduced hemoglobin from the blood oxygen signal collected by the blood oximeter (301); A pain regulation neural pathway activity quantification model (7) is constructed based on a Bayesian network to construct a pain regulation pathway activity index model, which integrates the EEG, electrodermal, and blood oxygen signal features collected in the signal processing and feature extraction module to quantify the activities of the descending inhibitory pathway, autonomic nerve pathway, and neurovascular pathway; The analgesic function evaluation platform and intervention system (8) can display the signal waveform and PAI value change trend in real time, generate a three-dimensional heat map, and automatically trigger and adjust the transcutaneous electrical stimulation analgesic intervention measures based on the PAI value, and combine VR technology for neurofeedback training.

2. The biosignal-based chronic low back pain analgesia function assessment platform according to claim 1, characterized in that: The acquisition devices in the multimodal biological signal synchronous acquisition system include an electroencephalogram signal acquisition device (1), a skin electrodermal signal acquisition device (2) and a blood oxygen signal acquisition device (3), and the data acquired by the multimodal biological signal synchronous acquisition system are centrally collected by a data acquisition card (4).

3. The biosignal-based chronic low back pain analgesia function assessment platform according to claim 2, characterized in that: The electroencephalogram signal acquisition device (1) comprises an electroencephalogram cap (101) and an electroencephalogram electrode sheet (102) covering a specific brain region, wherein the electroencephalogram electrode sheet (102) of the electroencephalogram cap (101) is made of silver chloride, and the electrodes cover the frontal lobe, central region, parietal lobe and brain regions related to pain perception and regulation.

4. The biosignal-based chronic low back pain analgesia function assessment platform according to claim 2, characterized in that: The galvanic skin signal acquisition device (2) comprises: a galvanic skin sensor (201) and a galvanic skin electrode sheet (202) covering the epidermis, wherein the galvanic skin electrode sheet (202) of the galvanic skin sensor (201) is of a disposable adhesive type and is placed on the palm or finger to measure changes in skin conductivity.

5. The biosignal-based chronic low back pain analgesia function assessment platform according to claim 2, characterized in that: The data collected by the data collection card (4) is uploaded to the data preprocessing module (5), and the data preprocessing module (5) processes the data, and the data processed by the data preprocessing module (5) is uploaded to the feature extraction unit (6) to extract EEG and electrocutaneous features.

6. The biosignal-based chronic low back pain analgesia function assessment platform according to claim 1, characterized in that: The data acquisition card (4), data preprocessing module (5) and feature extraction unit (6) constitute a signal processing and feature extraction module, and the data of the signal processing and feature extraction module are uploaded to the pain regulation neural pathway activity quantification model (7) and the analgesic function evaluation platform and intervention system (8).

7. A method for evaluating the analgesic function of chronic low back pain based on biological signals, characterized in that: include: Step 1: The EEG cap (101) in the multimodal biosignal synchronous acquisition system synchronously acquires EEG signals of the patient with chronic low back pain, the galvanic skin sensor (201) acquires the galvanic skin signals of the patient, and the finger-clip oximeter (301) acquires the blood oxygen signal of the patient; Step 2: Connect the EEG cap (101), the galvanic skin sensor (201) and the oximeter (301) via a data acquisition card (4), and use a hardware trigger or software synchronization algorithm to ensure that the start time of each signal acquisition is consistent, and compensate for the time difference caused by transmission delay to achieve accurate synchronization; Step 3: Use independent component analysis to remove artifacts of electrooculography and electromyography from the collected EEG signals, filter the electrocutaneous signals to eliminate high-frequency noise interference, and use a smoothing algorithm to remove abnormal fluctuations from the blood oxygen signal. Then, extract features from the EEG signals, electrocutaneous signals, and blood oxygen signals. Step 4: Construct a pain regulation pathway activity index model based on the Bayesian network, integrate the characteristics of EEG, electrodermal, and blood oxygen signals, and quantify the activity of descending inhibitory pathways, autonomic nerve pathways, and neurovascular pathways; Step 5: Generate a pain regulation pathway activity assessment report based on the PAI value calculated by the PAI model, formulate a personalized analgesic treatment plan based on the assessment report, and regularly repeat the above biological signal collection, processing, quantification and evaluation steps during the treatment process, and monitor and adjust treatment measures in real time according to changes in the PAI value.

8. The method for evaluating the analgesic function of chronic low back pain based on biological signals according to claim 7, characterized in that: In the multimodal bio-signal acquisition step, the EEG signal sampling frequency is set to 500 Hz-1000 Hz, and the electrodermal signal sampling frequency is set to 100 Hz-200 Hz.

9. The method for evaluating the analgesic function of chronic low back pain based on biological signals according to claim 7, characterized in that: In step 4, the Bayesian network structure is learned from the collected data through the data mining algorithm of the K2 algorithm and the maximum expectation algorithm, and the parameters in the network are estimated using the maximum likelihood estimation method.

10. The method for evaluating the analgesic function of chronic low back pain based on biological signals according to claim 7, characterized in that: In step 5, the formulation of a personalized analgesic treatment plan includes automatically triggering transcutaneous electrical stimulation analgesic intervention measures when the PAI value is lower than the set threshold, and dynamically adjusting the stimulation intensity according to the SCR amplitude of the skin electrode signal. At the same time, combined with virtual reality technology, the patient's prefrontal lobe's ability to inhibit pain is enhanced through neurofeedback training.

Images