A pain monitoring system and monitoring method

Abstract

The present application relates to the technical field of pain monitoring, and discloses a pain monitoring system and a monitoring method, which comprise the following steps: synchronously inputting skin conductivity, skin temperature, three-axis acceleration components and electromyography digital signals of a user, calculating a relative change rate of the skin conductivity and a relative change rate of the skin temperature to construct an impedance drift penalty factor, and correcting an original electromyography signal amplitude by using the impedance drift penalty factor; calculating a ratio of a time-domain instantaneous change rate of the corrected electromyography signal amplitude to a resting initial electromyography change rate to generate an electromyography spasm index; extracting a kinematic distortion degree by using a resultant vector module length of the three-axis acceleration; calculating a decoupling weight according to the electromyography spasm index and the kinematic distortion degree, and compensating and calculating a time-domain first-order difference of the skin conductivity by using the decoupling weight to output a breakthrough pain characteristic index.

Description

Technical Field

[0001] This invention relates to the field of pain monitoring technology, specifically to a pain monitoring system and monitoring method. Background Technology

[0002] Severe and chronic pain significantly impacts users' quality of life. With the shift in healthcare models, many users with pain monitoring needs are choosing to manage their pain long-term at home. Currently, the main clinical pain assessment methods still rely on subjective scales such as the Visual Analogue Scale (VAS). This approach is highly dependent on the user's subjective description, exhibiting significant lag and difficulty in real-time monitoring of sudden events such as explosive pain. To achieve objective pain quantification, existing technologies attempt to collect signals such as heart rate, blood oxygen, and surface electromyography (EMG) using wearable physiological sensors. However, the home environment differs greatly from the controlled environment of a hospital, and complex daily activities pose a significant challenge to the accurate extraction of physiological signals. In practical applications, prolonged sensor wear leads to sweat accumulation at the electrodes, causing skin impedance drift and resulting in severe nonlinear distortion of the collected EMG signals. Furthermore, users' physical movements at home, such as turning over and walking, produce significant mechanical artifacts. These artifacts often overlap with the physiological responses caused by pain in the frequency domain, making the system prone to false alarms. A deeper problem lies in the coupling between the physiological characteristics induced by pain and the characteristics of daily activities. Severe pain attacks are often accompanied by painful muscle spasms, which are highly similar to normal bodily movements in terms of kinematics. Existing denoising algorithms typically filter out motion signals as pure noise, but this leads to the omission of truly diagnostic spasm features. Simultaneously, fluctuations in skin conductivity caused by sympathetic nerve stress (i.e., cold sweats) are easily masked by fluctuations in sweat gland secretion caused by heat production from daily activities. Existing dynamic baseline extraction methods struggle to pinpoint subtle pain shifts in the context of rapidly changing home environments.

[0003] In summary, existing pain monitoring methods are ineffective in addressing impedance drift caused by prolonged wear in complex home environments, struggle to accurately decouple painful spasms from everyday motion artifacts, and fail to extract stable burst pain indicators when physiological characteristics are masked by noise. Therefore, a pain monitoring technology solution is needed that can deeply decouple impedance drift, kinematic distortions, and painful spasms in the home environment, and output stable pain characteristic indices in real time. Summary of the Invention

[0004] This invention provides a pain monitoring system and method, which helps to solve the problems mentioned in the background art.

[0005] This invention provides the following technical solution: a pain monitoring method, comprising:

[0006] Input the user's multimodal physiological signals and convert the digital electromyography signals into raw electromyography signal amplitudes;

[0007] The relative rates of change of current skin conductivity and skin temperature relative to the initial resting state are calculated separately, and an impedance drift penalty factor is constructed based on the absolute value of their product.

[0008] The original electromyographic signal amplitude is multiplied using the impedance drift penalty factor to obtain the corrected electromyographic signal amplitude.

[0009] The instantaneous rate of change of the amplitude of the modified electromyographic signal in the time domain is calculated, and the rate of change is compared with the initial rate of change of the resting electromyographic signal to obtain the electromyographic spasm index;

[0010] Calculate the resultant vector magnitude of the current triaxial acceleration, and extract the relative change of this magnitude from the magnitude of the initial gravitational acceleration in the resting environment to obtain the kinematic distortion.

[0011] The electromyographic spasm index is introduced into the denominator as the damping coefficient, and the kinematic distortion is used as the numerator. The quotient of the damping denominator is then used to obtain the decoupling weight.

[0012] Calculate the first-order time-domain difference of the current skin conductivity and extract the dynamic baseline offset;

[0013] The dynamic baseline offset is used as the numerator, and the sum of the decoupling weight and the value 1 is used as the denominator to calculate the quotient, obtain the compensation offset, and combine it with the resting initial skin conductivity mapping to output the burst pain characteristic index.

[0014] Optionally, the input of the user's multimodal physiological signals and the conversion of the electromyographic digital signals therein into raw electromyographic signal amplitudes includes:

[0015] The multimodal physiological signals include: initial resting skin conductivity, initial resting skin temperature, and initial resting electromyographic rate of change; the initial resting environmental gravitational acceleration modulus is obtained; the system subsequently performs real-time data sampling at a fixed sampling period;

[0016] Obtain the sensor reference voltage and sensor analog-to-digital conversion resolution;

[0017] Calculate the square of the value corresponding to the analog-to-digital conversion resolution and subtract 1 to obtain the conversion denominator;

[0018] The voltage reference voltage is divided by the denominator of the conversion to obtain the voltage reference ratio.

[0019] Multiply the voltage reference ratio by the current value of the digital electromyography signal to obtain the original electromyography signal amplitude at the current time.

[0020] Synchronously input the skin conductivity, skin temperature, and triaxial acceleration components at the current moment.

[0021] Optionally, the step of calculating the relative rates of change of current skin conductivity and skin temperature relative to the initial resting state, and constructing an impedance drift penalty factor based on the absolute value of their product, includes:

[0022] Calculate the difference between the current skin conductivity and the initial resting skin conductivity, and divide this difference by the initial resting skin conductivity to obtain the relative rate of change of skin conductivity;

[0023] Calculate the difference between the current skin temperature and the initial resting skin temperature, and divide this difference by the initial resting skin temperature to obtain the relative rate of change of skin temperature;

[0024] Multiply the relative rate of change of skin conductivity by the relative rate of change of skin temperature, and calculate the absolute value of the product;

[0025] Add the absolute value to the numerical value 1 as the denominator, and use the numerical value 1 as the numerator. The quotient of the two is the impedance drift penalty factor.

[0026] Optionally, the step of multiplying the original electromyographic signal amplitude using the impedance drift penalty factor to obtain the corrected electromyographic signal amplitude includes:

[0027] Multiply the original electromyographic signal amplitude at the current moment by the impedance drift penalty factor, and use the product as the corrected electromyographic signal amplitude.

[0028] Optionally, the step of calculating the instantaneous rate of change in the time domain of the modified electromyographic signal amplitude and comparing this rate of change with the initial rate of change in resting electromyographic signal to obtain the electromyographic spasm index includes:

[0029] Calculate the difference between the current corrected electromyographic signal amplitude and the corrected electromyographic signal amplitude of the previous sampling period;

[0030] The difference is divided by the sampling period, and the absolute value of the quotient is taken as the instantaneous rate of change of the electromyographic signal.

[0031] The electromyographic spasm index is obtained by dividing the instantaneous rate of change of the electromyographic signal by the initial rate of change of the resting electromyographic signal.

[0032] Optionally, the step of calculating the resultant vector magnitude of the current triaxial acceleration and extracting the relative change value of this magnitude from the resting initial environment gravitational acceleration magnitude to obtain the kinematic distortion degree includes:

[0033] Calculate the squares of the three-axis acceleration components at the current moment, and then sum the three squares.

[0034] Calculate the square root of the summation result to obtain the current acceleration modulus;

[0035] Calculate the absolute difference between the current acceleration modulus and the initial gravitational acceleration modulus of the resting environment;

[0036] The kinematic distortion is obtained by quotienting the absolute difference with the magnitude of the initial gravitational acceleration at rest.

[0037] Optionally, the step of introducing the electromyographic spasm index into the denominator as a damping coefficient, and using the kinematic distortion degree as the numerator, quotienting it with the damping denominator to obtain the decoupling weight includes:

[0038] Add 1 to the electromyographic spasm index to obtain the damping denominator;

[0039] The kinematic distortion degree is used as the numerator, and the quotient is calculated with the damping denominator to obtain the decoupling weight.

[0040] Optionally, the step of calculating the time-domain first-order difference of the current skin conductivity and extracting the dynamic baseline offset includes:

[0041] Calculate the difference between the current skin conductivity and the skin conductivity of the previous sampling period;

[0042] The dynamic baseline offset is obtained by dividing the difference by the sampling period.

[0043] Optionally, the step of using the dynamic baseline offset as the numerator and the sum of the decoupling weight and the value 1 as the denominator to perform a quotient calculation to obtain the compensation offset, and combining it with the resting initial skin conductivity mapping to output the burst pain characteristic index, includes:

[0044] Add the value 1 to the decoupling weight to obtain the compensation denominator;

[0045] The compensation offset is obtained by taking the dynamic baseline offset as the numerator and dividing it by the compensation denominator.

[0046] The compensation offset is multiplied by the sampling period to obtain the product; the product is then divided by the initial resting skin conductivity to obtain the burst pain characteristic index of the system's final output.

[0047] Optionally, a system for implementing a pain monitoring method includes:

[0048] Signal processing module: used to synchronously input digital signals of skin conductivity, temperature, triaxial acceleration and electromyography and convert them into physical amplitudes;

[0049] Impedance penalty module: used to calculate skin conductivity and relative temperature change rate to construct impedance drift penalty factor;

[0050] Electromyography correction module: used to multiply the amplitude of the original electromyography signal using the penalty factor;

[0051] Spasm Calculation Module: Used to calculate the instantaneous rate of change of the amplitude of the corrected electromyographic signal and compare it with the benchmark to generate an electromyographic spasm index;

[0052] Motion quantization module: used to calculate the deviation of the magnitude of the resultant acceleration vector to obtain the kinematic distortion.

[0053] Decoupling modulation module: used to modulate the distortion degree with the spasticity index as damping to generate decoupling weights;

[0054] Baseline extraction module: used to calculate the dynamic baseline offset by first-order difference in the time domain of skin conductivity;

[0055] Feature output module: Used to compensate for baseline offset using decoupled weights and map the output burst pain feature index.

[0056] The present invention has the following beneficial effects:

[0057] 1. By inputting multimodal signals such as skin conductivity, skin temperature, electromyography (EMG), and triaxial acceleration from the user, this technical solution addresses the signal distortion problem encountered in long-term continuous monitoring in home settings. During daily life at home, prolonged use of the monitoring device can lead to sweat accumulation at the skin electrodes, causing skin impedance drift and distortion of underlying electrical signals. Simultaneously, normal bodily activities such as turning over and walking can produce mechanical artifacts, interfering with or masking subtle pain-related physiological characteristics. When a user experiences involuntary painful spasms due to severe pain, these spasms physically resemble normal bodily movements and are easily misidentified as ordinary bodily activities by conventional algorithms. Therefore, this solution extracts the relative change rates of skin conductivity and skin temperature to construct an impedance drift penalty factor, which is used to correct EMG signal distortion. This quantifies the EMG spasm index and kinematic distortion, and uses the EMG spasm index as damping to calculate the decoupling weights for the motion features, ultimately compensating for the dynamic baseline offset of skin conductivity. Through the above processing, this solution helps reduce sweat short-circuit interference caused by prolonged wear by users, reduces the impact of body motion artifacts caused by daily activities, and retains pain spasm-related features that are easily misfiltered. This enables the system to extract relatively stable objective quantitative indicators of burst pain from complex electrophysiological signals with a lot of noise in daily life, thereby reducing the risk of false alarms caused by daily activities and reducing the risk of missed alarms caused by misjudgment of pain spasms.

[0058] 2. By inputting initial baselines of various physiological parameters of the user in a pain-free and resting state, as well as environmental gravitational acceleration parameters, an individualized physical reference system can be established for the entire monitoring system. This processing method, which combines initial baseline acquisition with underlying digital-to-physical reconstruction, helps reduce the absolute value deviation caused by individual physiological differences among users in subsequent signal feature extraction. Simultaneously, converting digital electromyography (EMG) signals into raw EMG signal amplitudes with actual physical meaning improves the consistency of physical dimensions when performing joint calculations between various modal signals. By performing physical quantity conversion and baseline alignment at the source, the system can more stably sense changes in underlying physiological potentials. This reduces the static background error introduced by individual differences in hardware devices and different initial wearing states in home settings outside of controlled medical environments, providing a data foundation for subsequent extraction of pain-related physiological features.

[0059] 3. By calculating the relative rates of change of current skin conductivity and current skin temperature relative to the initial resting state, and constructing an impedance drift penalty factor based on the absolute value of the product of the two relative rates of change, the skin conductivity characteristic representing sweat secretion and the skin temperature characteristic representing changes in body surface heat can be jointly verified. Using the product of the two rates of change and taking the absolute value as an additional parameter in the denominator helps capture the thermodynamic and electrochemical correlation changes caused by local sweat accumulation due to prolonged sensor wear. When both skin conductivity and skin temperature deviate significantly, the constructed impedance drift penalty factor value will adaptively decrease, thereby quantifying the changes in skin interface impedance caused by changes in the microenvironment. This multimodal physiological indicator joint processing logic can reduce the risk of environmental misjudgment caused by relying solely on skin conductivity or skin temperature changes, reduce the interference of non-sweat factors such as external environmental temperature changes on impedance assessment, and enable the system to perceive the degree of physical state change of the contact interface in real time, providing a dynamic suppression coefficient for subsequently reducing nonlinear distortion of electromyographic signals.

[0060] 4. By multiplying the amplitude of the original EMG signal obtained through conversion using a pre-constructed impedance drift penalty factor, distortion of the underlying muscle EMG signal can be suppressed immediately. When the system detects sweat accumulation and impedance decay at the electrodes, the reduced impedance drift penalty factor can simultaneously suppress the abnormally amplified amplitude of the original EMG signal, thereby reducing signal artifacts caused by the short-circuiting effect of skin fluid. This multiplicative correction mechanism helps to improve the gradual baseline drift caused by prolonged wear while preserving the high-frequency fluctuation characteristics of the EMG signal. This processing method does not require complex frequency domain filtering and transformation of the EMG signal, reducing the risk of loss of real transient high-frequency spasm signals due to forced filtering. This allows the final obtained corrected EMG signal amplitude to more stably reflect the actual discharge intensity of the user's underlying muscle motor fibers, thereby improving the consistency of the surface EMG signal evaluation scale in long-term home monitoring environments.

[0061] 5. By calculating the instantaneous rate of change of the corrected electromyographic (EMG) signal amplitude within adjacent sampling periods and comparing this instantaneous rate of change with the initial resting EMG rate of change, an EMG spasm index characterizing the intensity of muscle tetanic contraction can be extracted. Calculating the time-domain difference between adjacent moments helps reduce the influence of residual low-frequency mechanical noise in the EMG signal and highlights the instantaneous discharge changes of muscle fibers within a short period. Furthermore, normalizing the ratio of the acquired instantaneous rate of change to the initial resting baseline can accommodate differences in muscle volume, subcutaneous fat thickness, and baseline muscle tone levels among different users. Through this relative change feature extraction method, the final generated EMG spasm index has good cross-individual comparability and dynamic sensitivity, and can help identify unconscious, high-frequency painful tetanic activity induced by explosive pain, providing a physiological feature dimension for subsequent separation of pain-related physiological responses from complex bodily behaviors.

[0062] 6. By calculating the sum of squares of the three-axis acceleration components at the current moment and taking the square root, the resultant vector magnitude of the current acceleration is obtained. Then, the absolute difference between this magnitude and the initial resting environment gravity reference is calculated, and this difference is compared with the initial resting environment gravity acceleration magnitude to assess the user's overall body kinematic distortion. Obtaining the resultant vector magnitude in three-dimensional space reduces the impact of changes in user orientation and sensor attachment angle on motion assessment results and captures the intensity of the user's mechanical displacement in the three-dimensional coordinate system. Difference and ratio calculations between the calculated magnitude and the static gravity field reference help reduce the static bias effect of the Earth's gravity field on motion monitoring, enabling the extracted kinematic distortion to characterize the magnitude of dynamic additional acceleration generated by the user's active or passive actions such as turning over or getting up. This physical dynamics feature extraction scheme helps quantify the level of mechanical artifact noise in the home environment that may affect the physiological monitoring baseline, providing a dynamic reference for subsequent signal decoupling and interference suppression.

[0063] 7. By incorporating the electromyographic spasm index, which characterizes abnormal muscle twitching, into the denominator as a damping coefficient, and calculating the decoupling weight by quotienting the extracted kinematic distortion, the confusion between daily physical activities and pain-induced spasm responses can be improved. When users perform normal physical activities, the electromyographic spasm index is at a low level because there is no obvious accompanying muscle rigidity, allowing more kinematic distortion to be transmitted as decoupling weight, thus marking a high level of motor interference at this stage. Conversely, when users experience severe pain that triggers twitching or struggling, the elevated electromyographic spasm index increases the damping denominator value, thereby reducing the decoupling weight of the feedback. This negative feedback regulation helps reduce the risk of misjudging pain-induced violent movements as normal daily activities and reduces the possibility of incorrectly filtering out true pain-related physiological signals, thereby improving the monitoring system's ability to retain and identify breakthrough pain-related features.

[0064] 8. By calculating the first-order difference in the time domain between the current skin conductivity and the skin conductivity of the previous sampling period, the dynamic baseline shift characterizing the transient stress response of the autonomic nervous system can be extracted. In long-term continuous physiological monitoring at home, skin conductivity is affected by a combination of slowly changing factors such as gradual changes in room temperature, fluctuations in the user's natural biological rhythms, and daily metabolic activities, resulting in a low-frequency drift background. Using the first-order difference calculation between adjacent periods is equivalent to constructing a transient change feature extraction method in the time domain, which can suppress slowly changing irrelevant physiological background noise and capture high-frequency abrupt signals caused by sympathetic nerve excitation due to pain stimulation, leading to changes in sweat gland secretion. This dynamic incremental extraction method helps to improve the problem that the traditional fixed absolute threshold judgment method is prone to failure when facing baseline drift, enabling the device to capture the neurophysiological stress characteristics of transient pain attacks more stably in the context of home physiological changes.

[0065] 9. By utilizing pre-generated decoupling weights, the extracted dynamic baseline offset is compensated by quotient calculation. Further, by combining the sampling period with the initial resting skin conductivity for final-state mapping, a compensated burst pain characteristic index with a unified evaluation scale can be output. Using the decoupling weights as the compensation denominator, when high-weighted conventional somatic motion interference is identified, the spurious skin conductivity offset caused by normal movement-induced heat is proportionally reduced, thereby improving the influence of motion artifacts in pain-related characteristic indicators. Subsequently, the compensated offset is multiplied by the sampling period and divided by the initial resting skin conductivity, completing the conversion from an absolute physical quantity to a dimensionless relative characteristic index. This mapping mechanism enables good horizontal comparability of monitoring and evaluation data from different time spans, different hardware sampling frequencies, and users with different physical conditions, and forms a processing link from low-level signal interference suppression to high-level feature adaptive purification, ultimately providing remote management with highly confident and interference-resistant quantitative reference data for pain characteristics. Attached Figure Description

[0066] Figure 1 This is a schematic diagram of the basic process of the present invention.

[0067] Figure 2 This is a schematic diagram of the generation of impedance drift penalty factor and electromyographic signal correction processing in this invention.

[0068] Figure 3 This is a schematic diagram illustrating the generation of the electromyographic spasm index, kinematic distortion degree, and decoupling weights in this invention.

[0069] Figure 4 This is a schematic diagram illustrating the dynamic baseline offset extraction, reverse compensation, and burst pain characteristic index output of the present invention. Detailed Implementation

[0070] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0071] Example 1, refer to Figure 1 A pain monitoring method, comprising:

[0072] Input the user's multimodal physiological signals and convert the digital electromyography signals into raw electromyography signal amplitudes;

[0073] The relative rates of change of current skin conductivity and skin temperature relative to the initial resting state are calculated separately, and an impedance drift penalty factor is constructed based on the absolute value of their product.

[0074] The original electromyographic signal amplitude is multiplied using the impedance drift penalty factor to obtain the corrected electromyographic signal amplitude.

[0075] The instantaneous rate of change of the amplitude of the modified electromyographic signal in the time domain is calculated, and the rate of change is compared with the initial rate of change of the resting electromyographic signal to obtain the electromyographic spasm index;

[0076] Calculate the resultant vector magnitude of the current triaxial acceleration, and extract the relative change of this magnitude from the magnitude of the initial gravitational acceleration in the resting environment to obtain the kinematic distortion.

[0077] The electromyographic spasm index is introduced into the denominator as the damping coefficient, and the kinematic distortion is used as the numerator. The quotient of the damping denominator is then used to obtain the decoupling weight.

[0078] Calculate the first-order time-domain difference of the current skin conductivity and extract the dynamic baseline offset;

[0079] The dynamic baseline offset is used as the numerator, and the sum of the decoupling weight and the value 1 is used as the denominator to calculate the quotient, obtain the compensation offset, and combine it with the resting initial skin conductivity mapping to output the burst pain characteristic index.

[0080] The input user's multimodal physiological signals, and the conversion of the electromyographic digital signals therein into raw electromyographic signal amplitudes, include:

[0081] The multimodal physiological signals include: initial resting skin conductivity, initial resting skin temperature, and initial resting electromyographic rate of change; the initial resting environmental gravitational acceleration modulus is obtained; the system subsequently performs real-time data sampling at a fixed sampling period;

[0082] Obtain the sensor reference voltage and sensor analog-to-digital conversion resolution;

[0083] Calculate the square of the value corresponding to the analog-to-digital conversion resolution and subtract 1 to obtain the conversion denominator;

[0084] The voltage reference voltage is divided by the denominator of the conversion to obtain the voltage reference ratio.

[0085] Multiply the voltage reference ratio by the current value of the digital electromyography signal to obtain the original electromyography signal amplitude at the current time.

[0086] Synchronously input the skin conductivity, skin temperature, and triaxial acceleration components at the current moment.

[0087] The process involves calculating the relative rates of change of current skin conductivity and skin temperature relative to the initial resting state, and constructing an impedance drift penalty factor based on the absolute value of their product, including:

[0088] Calculate the difference between the current skin conductivity and the initial resting skin conductivity, and divide this difference by the initial resting skin conductivity to obtain the relative rate of change of skin conductivity;

[0089] Calculate the difference between the current skin temperature and the initial resting skin temperature, and divide this difference by the initial resting skin temperature to obtain the relative rate of change of skin temperature;

[0090] Multiply the relative rate of change of skin conductivity by the relative rate of change of skin temperature, and calculate the absolute value of the product;

[0091] Add the absolute value to the numerical value 1 as the denominator, and use the numerical value 1 as the numerator. The quotient of the two is the impedance drift penalty factor.

[0092] The step of multiplying the original electromyographic signal amplitude using the impedance drift penalty factor to obtain the corrected electromyographic signal amplitude includes:

[0093] Multiply the original electromyographic signal amplitude at the current moment by the impedance drift penalty factor, and use the product as the corrected electromyographic signal amplitude.

[0094] The calculation of the instantaneous rate of change of the modified electromyographic signal amplitude in the time domain, and the comparison of this rate of change with the initial rate of change of the resting electromyographic signal to obtain the electromyographic spasm index, includes:

[0095] Calculate the difference between the current corrected electromyographic signal amplitude and the corrected electromyographic signal amplitude of the previous sampling period;

[0096] The difference is divided by the sampling period, and the absolute value of the quotient is taken as the instantaneous rate of change of the electromyographic signal.

[0097] The electromyographic spasm index is obtained by dividing the instantaneous rate of change of the electromyographic signal by the initial rate of change of the resting electromyographic signal.

[0098] The calculation of the resultant vector magnitude of the current triaxial acceleration, and the extraction of the relative change of this magnitude from the resting initial environmental gravitational acceleration magnitude to obtain the kinematic distortion degree, includes:

[0099] Calculate the squares of the three-axis acceleration components at the current moment, and then sum the three squares.

[0100] Calculate the square root of the summation result to obtain the current acceleration modulus;

[0101] Calculate the absolute difference between the current acceleration modulus and the initial gravitational acceleration modulus of the resting environment;

[0102] The kinematic distortion is obtained by quotienting the absolute difference with the magnitude of the initial gravitational acceleration at rest.

[0103] The step of incorporating the electromyographic spasm index into the denominator as a damping coefficient, and using the kinematic distortion degree as the numerator, quotienting it with the damping denominator to obtain the decoupling weight includes:

[0104] Add 1 to the electromyographic spasm index to obtain the damping denominator;

[0105] The kinematic distortion degree is used as the numerator, and the quotient is calculated with the damping denominator to obtain the decoupling weight.

[0106] The calculation of the current skin conductivity using the first-order time-domain difference, and the extraction of the dynamic baseline offset, includes:

[0107] Calculate the difference between the current skin conductivity and the skin conductivity of the previous sampling period;

[0108] The dynamic baseline offset is obtained by dividing the difference by the sampling period.

[0109] The process involves using the dynamic baseline offset as the numerator and the sum of the decoupling weight and the value 1 as the denominator to calculate the quotient, obtaining the compensation offset, and combining this with the resting initial skin conductivity mapping to output the burst pain characteristic index, including:

[0110] Add the value 1 to the decoupling weight to obtain the compensation denominator;

[0111] The compensation offset is obtained by taking the dynamic baseline offset as the numerator and dividing it by the compensation denominator.

[0112] The compensation offset is multiplied by the sampling period to obtain the product; the product is then divided by the initial resting skin conductivity to obtain the burst pain characteristic index of the system's final output.

[0113] Example 2: A pain monitoring method, comprising:

[0114] The input user's multimodal physiological signals, and the conversion of the electromyographic digital signals therein into raw electromyographic signal amplitudes, include:

[0115] Deploy sensors to collect and input the following data when the user is in a pain-free and resting state:

[0116] initial skin conductivity at rest Initial resting skin temperature Initial rate of change of electromyography at rest ; Resting initial environment gravitational acceleration modulus ;

[0117] The system then uses a fixed sampling period. Perform real-time data input;

[0118] In this step, at the current time node It synchronously receives various real-time physiological and motor digital signals from the user and performs physical quantity reconstruction to provide source data for subsequent steps. First, for the input electromyographic digital signal, it calculates the corresponding physical dimension amplitude to obtain the raw electromyographic signal amplitude that can be directly used for feature calculation.

[0119] The amplitude of the original electromyographic signal is calculated using the following formula:

[0120]

[0121] in, This represents the amplitude of the original electromyographic signal at the current moment; This is the directly set sensor reference voltage; The sensor's analog-to-digital conversion resolution is a known quantity that is directly set. The decimal value of the current moment's electromyographic digital signal obtained through direct measurement;

[0122] Simultaneously input the current skin conductivity. Skin temperature ; and triaxial acceleration components , , .

[0123] By inputting initial baselines of various physiological parameters of the user in a pain-free and resting state, along with environmental gravitational acceleration parameters, an individualized physical reference system can be established for the entire monitoring system. This combination of initial baseline acquisition and underlying digital-to-physical reconstruction helps reduce the absolute value deviation caused by individual physiological differences among users in subsequent signal feature extraction. Simultaneously, converting digital electromyography (EMG) signals into raw EMG signal amplitudes with actual physical meaning improves the consistency of physical dimensions when performing joint calculations between various modal signals. By performing physical quantity conversion and baseline alignment at the source, the system can more stably sense changes in underlying physiological potentials. This reduces the static background error introduced by individual differences in hardware devices and varying initial wearing states in home settings outside of controlled medical environments, providing a data foundation for subsequent extraction of pain-related physiological features.

[0124] Reference Figure 2 , Figure 2 This paper illustrates the processing relationship between skin conductivity, temperature, and the generation of a penalty factor to correct electromyographic signals. The calculation of the relative rates of change of current skin conductivity and skin temperature relative to the initial resting state, and the construction of an impedance drift penalty factor based on the absolute value of their product, includes:

[0125] This step aims to address the issue that prolonged wear can lead to sweat buildup, causing abnormal changes in the correlation between skin conductivity and local temperature. This results in decreased impedance between the electrode and the skin, distorting the electromyographic signal. For quantitative correction, the dimensionless relative rate of change of the current skin conductivity relative to the initial state must first be calculated to characterize the change in ion concentration.

[0126] The relative rate of change in skin conductivity is calculated using the following formula:

[0127]

[0128] in, This represents the relative rate of change in skin conductivity at the current moment. Current skin conductivity; Initial skin conductivity at rest;

[0129] Next, in order to eliminate interference from non-sweat factors (such as ambient humidity), it is necessary to simultaneously calculate the dimensionless relative rate of change of the current skin temperature relative to the initial state, which is used to characterize the slight thermodynamic changes caused by the heat absorption of sweat evaporation.

[0130] The relative rate of change of skin temperature is calculated using the following formula:

[0131]

[0132] in, This represents the relative rate of change of skin temperature at the current moment. Current skin temperature; Initial resting skin temperature;

[0133] Subsequently, based on the coupling effect of skin conductivity and temperature changes, a penalty factor was calculated to suppress electromyographic signal drift. When both showed a significant synergistic change, it was determined that sweat accumulation caused impedance attenuation, and the penalty factor would tend to decrease to suppress attenuation.

[0134] The impedance drift penalty factor is calculated using the following formula:

[0135]

[0136] in, This is the impedance drift penalty factor; This is the relative rate of change of skin conductivity calculated beforehand; This is the relative rate of change of temperature calculated beforehand.

[0137] By calculating the relative rates of change of current skin conductivity and current skin temperature relative to the initial resting state, and constructing an impedance drift penalty factor based on the absolute value of the product of the two relative rates of change, the skin conductivity characteristic representing sweat secretion and the skin temperature characteristic representing changes in body surface heat can be jointly verified. Using the product of the two rates of change and taking the absolute value as an additional parameter in the denominator helps capture the thermodynamic and electrochemical correlation changes caused by local sweat accumulation due to prolonged sensor wear. When both skin conductivity and skin temperature deviate significantly, the constructed impedance drift penalty factor value adaptively decreases, thereby quantifying the changes in skin interface impedance caused by changes in the microenvironment. This multimodal physiological indicator joint processing logic can reduce the risk of environmental misjudgment caused by relying solely on skin conductivity or skin temperature changes, reduce the interference of non-sweat factors such as external environmental temperature changes on impedance assessment, and enable the system to perceive the degree of physical state change of the contact interface in real time, providing a dynamic suppression coefficient for subsequently reducing nonlinear distortion of electromyographic signals.

[0138] The step of multiplying the original electromyographic signal amplitude using the impedance drift penalty factor to obtain the corrected electromyographic signal amplitude includes:

[0139] This step utilizes the obtained impedance drift penalty factor to directly multiply the original electromyographic signal, eliminating the artificially high signal caused by reduced sweat impedance, and obtaining the corrected electromyographic signal amplitude that truly reflects the muscle activity potential.

[0140] The corrected electromyographic signal amplitude is calculated using the following formula:

[0141]

[0142] in, To correct the amplitude of electromyographic signals; The amplitude of the original electromyographic signal obtained from the pre-processing; This is the impedance drift penalty factor calculated beforehand.

[0143] By multiplying and correcting the amplitude of the original electromyographic (EMG) signal obtained through conversion using a pre-constructed impedance drift penalty factor, distortion of the underlying muscle EMG signal can be suppressed immediately. When the system detects sweat accumulation and impedance decay at the electrodes, the reduced impedance drift penalty factor simultaneously lowers the abnormally amplified amplitude of the original EMG signal, thereby reducing signal artifacts caused by the short-circuiting effect of skin fluid. This multiplicative correction mechanism helps improve the gradual baseline drift caused by prolonged wear while preserving the high-frequency fluctuation characteristics of the EMG signal. This processing method does not require complex frequency domain filtering and transformation of the EMG signal, reducing the risk of losing true transient high-frequency spasm signals due to forced filtering. This allows the final corrected EMG signal amplitude to more stably reflect the actual discharge intensity of the user's underlying muscle motor fibers, thereby improving the consistency of the surface EMG signal evaluation scale in long-term home monitoring environments.

[0144] Reference Figure 3 , Figure 3 The relationship between the generation of the electromyographic spasm index, kinematic distortion, and decoupling weights is shown. The calculation of the temporal instantaneous rate of change of the modified electromyographic signal amplitude, and the comparison of this rate of change with the initial resting electromyographic rate of change, yields the electromyographic spasm index, including:

[0145] The purpose of this step is to identify whether somatic spasms caused by pain have occurred. This requires performing time-domain first-order difference calculations on the corrected electromyographic signal to extract its instantaneous rate of change.

[0146] The instantaneous rate of change of electromyographic signals is calculated using the following formula:

[0147]

[0148] in, The instantaneous rate of change of the electromyographic signal; The amplitude of the current corrected electromyographic signal is obtained from the previous step; This is the corrected electromyographic signal amplitude from the previous sampling period; The sampling period is known.

[0149] Subsequently, the instantaneous rate of change is compared with the baseline rate of change in the initial resting state to generate a characteristic index for characterizing the severity of spasm. The larger the index, the more intense the muscle tetanic contraction.

[0150] The electromyography spasm index is calculated using the following formula:

[0151]

[0152] in, Electromyography (EMG) index; The instantaneous rate of change of the electromyographic signal obtained beforehand; The initial rate of change in electromyography at rest.

[0153] By calculating the instantaneous rate of change of the corrected electromyographic (EMG) signal amplitude within adjacent sampling periods and comparing this instantaneous rate of change with the initial resting EMG rate of change, an EMG spasm index characterizing the intensity of muscle tetany can be extracted. Calculating the temporal difference between adjacent moments helps reduce the influence of residual low-frequency mechanical noise in the EMG signal and highlights the instantaneous discharge changes of muscle fibers within a short period. Furthermore, normalizing the ratio of the acquired instantaneous rate of change to the initial resting baseline can accommodate differences in muscle volume, subcutaneous fat thickness, and baseline muscle tone levels among different users. Through this relative change feature extraction method, the final generated EMG spasm index has good cross-individual comparability and dynamic sensitivity, and can help identify unconscious, high-frequency painful tetany induced by explosive pain, providing a physiological feature dimension for subsequent separation of pain-related physiological responses from complex bodily behaviors.

[0154] The calculation of the resultant vector magnitude of the current triaxial acceleration, and the extraction of the relative change of this magnitude from the resting initial environmental gravitational acceleration magnitude to obtain the kinematic distortion degree, includes:

[0155] Using data from the accelerometer, the resultant vector magnitude of the triaxial acceleration in Euclidean space at the current moment is first calculated to assess the user's current overall displacement acceleration.

[0156] The current acceleration modulus is calculated using the following formula:

[0157]

[0158] in, This is the current acceleration modulus; , , These are the three-axis acceleration components;

[0159] Then, the degree to which the current acceleration deviates from the resting environment's gravity reference is assessed, and this deviation is defined as the kinematic distortion. This distortion reflects the magnitude of mechanical noise introduced to the physiological baseline by the user's body movements.

[0160] The kinematic distortion is calculated using the following formula:

[0161]

[0162] in, The degree of kinematic distortion; The current acceleration modulus is obtained from the previous calculation; The initial gravitational acceleration modulus of the resting environment.

[0163] By calculating the sum of squares of the three-axis acceleration components at the current moment and taking the square root, the resultant vector magnitude of the current acceleration is obtained. Then, the absolute difference between this magnitude and the initial resting environmental gravity reference is calculated, and this difference is compared with the initial resting environmental gravity acceleration magnitude to assess the user's overall body kinematic distortion. Obtaining the resultant vector magnitude in three-dimensional space reduces the impact of changes in user orientation and sensor attachment angle on motion assessment results and captures the intensity of the user's mechanical displacement in the three-dimensional coordinate system. Difference and ratio calculations between the calculated magnitude and the static gravity field reference help reduce the static bias effect of the Earth's gravity field on motion monitoring, enabling the extracted kinematic distortion to characterize the magnitude of dynamic additional acceleration generated by the user's active or passive actions such as turning over or getting up. This physical dynamics feature extraction scheme helps quantify the level of mechanical artifact noise in the home environment that may affect the physiological monitoring baseline, providing a dynamic reference for subsequent signal decoupling and interference suppression.

[0164] The electromyographic spasm index is introduced into the denominator as the damping coefficient, and the kinematic distortion is used as the numerator. The decoupling weight is obtained by quotienting the damping denominator, including:

[0165] To address the issue that simple stripping movements can lead to the misfiltering of painful spasms, the proposed solution uses the spasticity index as a damping coefficient to decouple and modulate kinematic distortion. During purely daily activities (with an extremely low spasticity index), distortion is fully transmitted; however, during painful spasms (with a high spasticity index), the system weakens the transmission of distortion, preserving the physiological characteristics of pain.

[0166] The decoupling weight is calculated using the following formula:

[0167]

[0168] in, For decoupling weights; The kinematic distortion is obtained from the previous step; The electromyography spasm index is calculated beforehand.

[0169] By incorporating the electromyographic spasticity index (EMG) representing abnormal muscle twitching into the denominator as a damping coefficient, and calculating the decoupling weight by quotienting the extracted kinematic distortion, the confusion between daily physical activities and pain-induced spastic responses can be mitigated. When users engage in normal physical activity, the EMG spasticity index is low due to the absence of significant accompanying muscle rigidity, allowing more kinematic distortion to be transmitted as decoupling weight, thus indicating a higher level of motor interference at this stage. Conversely, when users experience severe pain leading to twitching or struggling, the elevated EMG spasticity index increases the damping denominator, thereby reducing the decoupling weight of the feedback. This negative feedback regulation helps reduce the risk of misinterpreting pain-induced violent movements as normal daily activities and decreases the possibility of incorrectly filtering out genuine pain-related physiological signals, thereby improving the monitoring system's ability to retain and identify breakthrough pain-related features.

[0170] Reference Figure 4 , Figure 4 The diagram illustrates the relationship between dynamic baseline offset extraction, inverse compensation, and the output of the breakthrough pain characteristic index. The calculation of the current skin conductivity using the time-domain first-order difference to extract the dynamic baseline offset includes:

[0171] After obtaining the underlying decoupling weights, we began to extract core indicators representing the stress state of the autonomic nervous system. By calculating the first-order temporal difference of skin conductivity, we characterized the transient micro-fluctuations in sweat gland secretion caused by pain stimuli in response to the sympathetic nervous system.

[0172] The dynamic baseline offset is calculated using the following formula:

[0173]

[0174] in, This is the dynamic baseline offset; The current skin conductivity is calculated beforehand; The skin conductivity of the previous sampling period; The sampling period.

[0175] By calculating the first-order difference in the time domain between the current skin conductivity and the skin conductivity of the previous sampling period, the dynamic baseline shift characterizing the transient stress response of the autonomic nervous system can be extracted. In long-term, continuous physiological monitoring at home, skin conductivity is affected by a combination of slowly changing factors, such as gradual changes in room temperature, fluctuations in the user's natural circadian rhythm, and daily metabolic activities, resulting in a low-frequency drift background. Using the first-order difference calculation between adjacent periods is equivalent to constructing a transient feature extraction method in the time domain. This method can suppress slowly changing irrelevant physiological background noise and capture high-frequency abrupt signals caused by sympathetic nerve excitation triggered by pain stimulation, leading to changes in sweat gland secretion. This dynamic incremental extraction method helps to improve the problem of traditional fixed absolute threshold determination methods easily failing when faced with baseline drift, enabling the device to more stably capture the neurophysiological stress characteristics during transient pain attacks against the background of changing home physiological conditions.

[0176] The process involves using the dynamic baseline offset as the numerator and the sum of the decoupling weight and the value 1 as the denominator to calculate the quotient, obtaining the compensation offset, and combining this with the resting initial skin conductivity mapping to output the burst pain characteristic index, including:

[0177] First, decoupling weights are used to perform inverse compensation suppression on the dynamic baseline offset, eliminating artifact offsets caused by non-painful motion, and obtaining the purified pure baseline offset.

[0178] The compensation offset is calculated using the following formula:

[0179]

[0180] in, To compensate for the offset; The dynamic baseline offset calculated beforehand; The decoupling weights are calculated in advance.

[0181] Finally, to ensure that the judgment data output by the system is comparable across users, a reference benchmark system composed of the initial resting skin conductivity and the sampling period is used to transform the compensated offset with physical dimensions into a dimensionless characteristic index, which serves as the final pain quantification data sent by the system to the remote medical end.

[0182] The final burst pain characteristic index is calculated using the following formula:

[0183]

[0184] in, The burst pain characteristic index, which is the final output of the system, is used as the target quantity for evaluation. The compensation offset is calculated beforehand; The sampling period; The initial skin conductivity at rest.

[0185] By utilizing pre-generated decoupling weights, the extracted dynamic baseline offset is compensated by quotient calculation. Further, by combining the sampling period with the initial resting skin conductivity for final-state mapping, a compensated burst pain characteristic index with a unified evaluation scale can be output. Using the decoupling weights as the compensation denominator, when high-weighted conventional somatic motion interference is identified, the spurious skin conductivity offset caused by normal movement-induced heat is proportionally reduced, thereby improving the influence of motion artifacts in pain-related characteristic indicators. Subsequently, the compensated offset is multiplied by the sampling period and divided by the initial resting skin conductivity, completing the conversion from an absolute physical quantity to a dimensionless relative characteristic index. This mapping mechanism enables good horizontal comparability of monitoring and evaluation data from different time spans, different hardware sampling frequencies, and users with different physical conditions, and forms a processing link from low-level signal interference suppression to high-level feature adaptive purification, ultimately providing remote management with highly confident and interference-resistant quantitative reference data for pain characteristics.

[0186] Example 3: A system for implementing a pain monitoring method, comprising:

[0187] Signal processing module: used to synchronously input digital signals of skin conductivity, temperature, triaxial acceleration and electromyography and convert them into physical amplitudes;

[0188] Impedance penalty module: used to calculate skin conductivity and relative temperature change rate to construct impedance drift penalty factor;

[0189] Electromyography correction module: used to multiply the amplitude of the original electromyography signal using the penalty factor;

[0190] Spasm Calculation Module: Used to calculate the instantaneous rate of change of the amplitude of the corrected electromyographic signal and compare it with the benchmark to generate an electromyographic spasm index;

[0191] Motion quantization module: used to calculate the deviation of the magnitude of the resultant acceleration vector to obtain the kinematic distortion.

[0192] Decoupling modulation module: used to modulate the distortion degree with the spasticity index as damping to generate decoupling weights;

[0193] Baseline extraction module: used to calculate the dynamic baseline offset by first-order difference in the time domain of skin conductivity;

[0194] Feature output module: Used to compensate for baseline offset using decoupled weights and map the output burst pain feature index.

[0195] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0196] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A pain monitoring method, characterized in that, include: Input the user's multimodal physiological signals and convert the digital electromyography signals into raw electromyography signal amplitudes; Calculate the relative rates of change of current skin conductivity and skin temperature relative to the initial resting state, and construct an impedance drift penalty factor based on the absolute value of their product, including: Multiply the relative rate of change of skin conductivity by the relative rate of change of skin temperature, and calculate the absolute value of the product; Add the absolute value to the numerical value 1 as the denominator, and use the numerical value 1 as the numerator. The quotient of the two is the impedance drift penalty factor. The original electromyographic signal amplitude is multiplied using the impedance drift penalty factor to obtain the corrected electromyographic signal amplitude. The instantaneous rate of change of the amplitude of the modified electromyographic signal in the time domain is calculated, and the rate of change is compared with the initial rate of change of the resting electromyographic signal to obtain the electromyographic spasm index; Calculate the resultant vector magnitude of the current triaxial acceleration, extract the relative change of this magnitude from the magnitude of the initial gravitational acceleration in the resting environment to obtain the kinematic distortion, including: Calculate the absolute difference between the current acceleration modulus and the initial gravitational acceleration modulus of the resting environment; The absolute difference is divided by the magnitude of the initial gravitational acceleration at rest to obtain the kinematic distortion. The electromyographic spasm index is introduced into the denominator as the damping coefficient, and the kinematic distortion is used as the numerator. The decoupling weight is obtained by quotienting the damping denominator, including: Add 1 to the electromyographic spasm index to obtain the damping denominator; The kinematic distortion degree is used as the numerator and divided by the damping denominator to obtain the decoupling weight; Calculate the time-domain first-order difference of the current skin conductivity and extract the dynamic baseline offset, including: Calculate the difference between the current skin conductivity and the skin conductivity of the previous sampling period; The dynamic baseline offset is obtained by dividing this difference by the sampling period; Using the dynamic baseline offset as the numerator and the sum of the decoupling weight and the value 1 as the denominator, a quotient is calculated to obtain the compensation offset. This is then combined with the resting initial skin conductivity mapping to output a burst pain characteristic index, including: Add the value 1 to the decoupling weight to obtain the compensation denominator; The compensation offset is obtained by taking the dynamic baseline offset as the numerator and dividing it by the compensation denominator. The compensation offset is multiplied by the sampling period to obtain the product; the product is then divided by the initial resting skin conductivity to obtain the burst pain characteristic index of the system's final output.

2. The pain monitoring method according to claim 1, characterized in that, The input user's multimodal physiological signals, and the conversion of the electromyographic digital signals therein into raw electromyographic signal amplitudes, include: The multimodal physiological signals include: initial resting skin conductivity, initial resting skin temperature, and initial resting electromyographic rate of change; the initial resting environmental gravitational acceleration modulus is obtained; the system subsequently performs real-time data sampling at a fixed sampling period; Obtain the sensor reference voltage and sensor analog-to-digital conversion resolution; Calculate the square of the value corresponding to the analog-to-digital conversion resolution and subtract 1 to obtain the conversion denominator; The voltage reference voltage is divided by the denominator of the conversion to obtain the voltage reference ratio. Multiply the voltage reference ratio by the current value of the digital electromyography signal to obtain the original electromyography signal amplitude at the current time. Synchronously input the skin conductivity, skin temperature, and triaxial acceleration components at the current moment.

3. The pain monitoring method according to claim 2, characterized in that, The process involves calculating the relative rates of change of current skin conductivity and skin temperature relative to the initial resting state, and constructing an impedance drift penalty factor based on the absolute value of their product, including: Calculate the difference between the current skin conductivity and the initial resting skin conductivity, and divide this difference by the initial resting skin conductivity to obtain the relative rate of change of skin conductivity; Calculate the difference between the current skin temperature and the initial resting skin temperature, and divide this difference by the initial resting skin temperature to obtain the relative rate of change of skin temperature.

4. The pain monitoring method according to claim 3, characterized in that, The step of multiplying the original electromyographic signal amplitude using the impedance drift penalty factor to obtain the corrected electromyographic signal amplitude includes: Multiply the original electromyographic signal amplitude at the current moment by the impedance drift penalty factor, and use the product as the corrected electromyographic signal amplitude.

5. A pain monitoring method according to claim 4, characterized in that, The calculation of the instantaneous rate of change of the modified electromyographic signal amplitude in the time domain, and the comparison of this rate of change with the initial rate of change of the resting electromyographic signal to obtain the electromyographic spasm index, includes: Calculate the difference between the current corrected electromyographic signal amplitude and the corrected electromyographic signal amplitude of the previous sampling period; The difference is divided by the sampling period, and the absolute value of the quotient is taken as the instantaneous rate of change of the electromyographic signal. The electromyographic spasm index is obtained by dividing the instantaneous rate of change of the electromyographic signal by the initial rate of change of the resting electromyographic signal.

6. A pain monitoring method according to claim 5, characterized in that, The calculation of the resultant vector magnitude of the current triaxial acceleration, and the extraction of the relative change of this magnitude from the resting initial environmental gravitational acceleration magnitude to obtain the kinematic distortion degree, includes: Calculate the squares of the three-axis acceleration components at the current moment, and then sum the three squares. The square root of the summation result is taken to obtain the current acceleration modulus.

7. A system employing the pain monitoring method of claim 1, characterized in that, include: Signal processing module: used to synchronously input digital signals of skin conductivity, temperature, triaxial acceleration and electromyography and convert them into physical amplitudes; Impedance penalty module: used to calculate skin conductivity and relative temperature change rate to construct impedance drift penalty factor; Electromyography correction module: used to multiply the amplitude of the original electromyography signal using the penalty factor; Spasm Calculation Module: Used to calculate the instantaneous rate of change of the amplitude of the corrected electromyographic signal and compare it with the benchmark to generate an electromyographic spasm index; Motion quantization module: used to calculate the deviation of the magnitude of the resultant acceleration vector to obtain the kinematic distortion. Decoupling modulation module: used to modulate the distortion degree with the spasticity index as damping to generate decoupling weights; Baseline extraction module: used to calculate the dynamic baseline offset by first-order difference in the time domain of skin conductivity; Feature output module: Used to compensate for baseline offset using decoupled weights and map the output burst pain feature index.

Images