Method and system for detecting degree of human soreness based on human bioelectric impedance change

Abstract

The application relates to the technical field of health equipment, and discloses a human body soreness degree detection method and system based on human body bioelectric impedance change, which comprises the following steps: applying periodic mechanical excitation to a specified body part; synchronously collecting a human body bioelectric impedance signal; performing double-path data analysis on the bioelectric impedance signal to obtain a frequency domain correlation index and a time domain fluctuation index; fusing the frequency domain correlation index and the time domain fluctuation index to determine the soreness grade of the body part; and matching and executing a corresponding massage program according to the soreness grade. The system comprises a bioelectric impedance detection module, a mechanical execution component and a data analysis module, and is used for executing the foregoing method. The application realizes objective quantification of the soreness degree by comprehensively analyzing the synchronous response of physiological signals and excitation and overall fluctuation, provides individualized closed-loop massage regulation on this basis, and improves detection accuracy and massage effectiveness.

Description

Technical Field

[0001] This invention relates to the field of health equipment technology, specifically to a method and system for detecting the degree of human pain based on changes in human bioelectrical impedance. Background Technology

[0002] Massage chairs, as health devices that can relieve physical fatigue and promote blood circulation, have been widely welcomed by consumers. With technological advancements and the diversification of user needs, modern massage chairs are becoming increasingly feature-rich. Users are no longer satisfied with preset, fixed massage programs, but expect massage chairs to provide personalized and targeted massage solutions based on their actual physical condition, especially the degree of soreness in different areas. Therefore, how to enable massage chairs to automatically and accurately detect sore areas and match corresponding massage modes has become an important technological development direction in this field.

[0003] To achieve automatic detection of soreness levels, several solutions have been proposed in the prior art. For example, Chinese patents CN 111096757 A and CN 211934041 U disclose a design for detection using a soreness detection handle. These solutions typically employ a complex handle that integrates force transmission components, sensing components, and multiple different types of sensors. Their detection principle usually relies on the user's active physical feedback to massage stimulation; for example, when a user feels soreness in a certain area, they will unconsciously grip the handle more tightly. The system determines the degree of soreness by measuring the magnitude or change in the force applied by the user's grip.

[0004] However, this solution, relying on a complex mechanical sensing handle, has inherent limitations. First, the integration of force transmission components and multiple different types of sensors complicates the overall structure of the handle, increasing both design and manufacturing complexity and the risk of mechanical failure. Second, this complex structure and diverse sensor configurations significantly increase hardware costs, hindering market adoption. Furthermore, the accuracy of this detection method depends to some extent on the user's cooperation and reaction patterns; differences in grip strength and reaction habits among users can introduce uncertainty into the results. Therefore, the industry still needs a simpler, lower-cost, and more objective pain level detection solution. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a method and system for detecting human pain levels based on changes in human bioelectrical impedance. This solves the problems of existing massage chair pain detection solutions relying on user subjective judgment, having complex structures, or being costly, which prevent the automatic and objective quantification of human pain levels to provide truly personalized massage.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solution: a method and system for detecting the degree of human pain based on changes in human bioelectrical impedance, comprising: S1, applying periodic mechanical stimulation to a designated body part;

[0007] S2. Synchronously collect human bioelectrical impedance signals;

[0008] S3. Perform dual-path data analysis on the bioelectrical impedance signal to obtain frequency domain correlation index and time domain fluctuation index;

[0009] S4. By integrating the frequency domain correlation index and the time domain volatility index, the pain level of the specified body part is determined.

[0010] S5. Match and execute the corresponding massage program according to the soreness level.

[0011] Preferably, the application of periodic mechanical excitation includes:

[0012] A sinusoidal control signal is generated, which is used to drive a motor, and the motor generates the periodic mechanical excitation.

[0013] Preferably, after acquiring the human bioelectrical impedance signal and before performing dual-path data analysis, the method further includes:

[0014] The bioelectrical impedance signal is filtered, amplified, and converted from analog to digital to obtain a digital time series signal.

[0015] Preferably, the frequency domain correlation analysis includes:

[0016] Perform a Fourier transform on the bioelectrical impedance signal to obtain the amplitude of the bioelectrical impedance signal at the excitation frequency;

[0017] Perform a Fourier transform on the excitation signal of the periodic mechanical excitation to obtain the amplitude of the excitation signal at the excitation frequency;

[0018] The amplitude of the bioelectrical impedance signal at the excitation frequency is normalized to the amplitude of the excitation signal at the excitation frequency to obtain the frequency domain correlation index.

[0019] Preferably, the time-domain volatility analysis includes:

[0020] Calculate the arithmetic mean of the bioelectrical impedance signal;

[0021] Calculate the sum of squares of the differences between each sampling point in the bioelectrical impedance signal and the arithmetic mean;

[0022] The time-domain volatility index is obtained based on the sum of squares.

[0023] Preferably, prior to the time-domain volatility analysis, the method further includes:

[0024] The bioelectrical impedance signal is preprocessed by median filtering.

[0025] Preferably, the fusion includes:

[0026] The frequency domain correlation index is compared with a first preset threshold to obtain a preliminary frequency domain level.

[0027] The time-domain volatility index is compared with a second preset threshold to obtain a preliminary time-domain level.

[0028] The pain level is determined based on the preliminary frequency domain level and the preliminary time domain level.

[0029] Preferably, determining the pain level includes:

[0030] When both the preliminary frequency domain level and the preliminary time domain level are severe, the aches and pains are determined to be severe aches and pains.

[0031] When both the preliminary frequency domain level and the preliminary time domain level are mild, the aches and pains are determined to be mild aches and pains.

[0032] Preferably, determining the pain level further includes:

[0033] If the criteria for severe and mild aches and pains are not met, the aches and pains are classified as moderate aches and pains.

[0034] Preferably, a human pain level detection system based on changes in human bioelectrical impedance includes:

[0035] Bioelectrical impedance detection module, used to collect human bioelectrical impedance signals;

[0036] Mechanical actuators are used to apply periodic mechanical excitation to designated body parts;

[0037] The data analysis module is electrically connected to the bioelectrical impedance detection module and the mechanical actuation component;

[0038] The data analysis module is configured to:

[0039] The excitation applied to the mechanical actuator is synchronously controlled to the signal acquisition of the bioelectrical impedance detection module;

[0040] Dual-path data analysis was performed on the bioelectrical impedance signal to obtain frequency domain correlation index and time domain fluctuation index;

[0041] By combining the frequency domain correlation index and the time domain volatility index, the pain level of the specified body part is determined.

[0042] Based on the level of soreness, the mechanical actuator is controlled to perform the corresponding massage program.

[0043] This invention provides a method and system for detecting the degree of human pain based on changes in human bioelectrical impedance. It has the following beneficial effects:

[0044] 1. This invention combines the synchronous response intensity of physiological signals and external stimuli with the overall fluctuation of the signal itself, so that the detection results can more accurately reflect the user's true physiological state and improve the reliability of pain level determination.

[0045] 2. This invention directly uses the detected pain level to match and execute the corresponding massage program, realizing the automation and personalization of massage plans. The system can provide users with targeted massages of different intensities, techniques and ranges based on the quantified pain results, avoiding the limitations of the traditional one-size-fits-all fixed program. This closed-loop adjustment method of personalized massage allows massage resources to be used more effectively, thereby improving the actual effect of massage and user satisfaction.

[0046] 3. This invention integrates a bioelectrical impedance detection module into the conventional components of a massage chair. Users only need to naturally hold the handle while receiving a massage to complete the collection of physiological signals. The entire detection process is integrated with the massage experience, without requiring users to wear additional sensors or perform complex operations, thus ensuring the ease of use of the product and its practical value in consumer products. Attached Figure Description

[0047] Figure 1 This is a system composition block diagram of the present invention;

[0048] Figure 2 This is a flowchart of the human body pain level detection method of the present invention;

[0049] Figure 3 This is a schematic diagram of the data flow for signal processing and analysis in this invention;

[0050] Figure 4 This is the logic diagram for determining the pain level fusion of the present invention.

[0051] Among them, 10 is the bioelectrical impedance detection module; 20 is the signal processing module; and 30 is the data analysis module. Detailed Implementation

[0052] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0053] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0054] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0055] (Example 1)

[0056] See attached document Figure 1 This invention provides a system for detecting the degree of human pain based on changes in human bioelectrical impedance. This system can be used to execute the method disclosed in this invention. The system includes: a bioelectrical impedance detection module 10, a signal processing module 20, and a data analysis module 30. The signal processing module 20 is electrically connected between the bioelectrical impedance detection module 10 and the data analysis module 30.

[0057] In a specific physical implementation, the bioelectrical impedance detection module 10 can be embodied as a handle structure including conductive electrode pads. The signal processing module 20 can consist of discrete analog circuit elements and an analog-to-digital converter. The data analysis module 30 can be a microcontroller, a digital signal processor, or an embedded system motherboard. In a preferred embodiment, the analog-to-digital converter portion of the signal processing module 20 and the data analysis module 30 can be integrated into the same microcontroller chip to simplify the system structure and reduce interference in signal transmission.

[0058] See attached document Figure 2 This invention provides a method for detecting the degree of human pain based on changes in human bioelectrical impedance. Referring to the accompanying drawings, the method may include the following steps:

[0059] S1. System initialization and test preparation;

[0060] S2. Apply periodic mechanical stimulation to the designated body part and simultaneously collect the bioelectrical impedance signal of the human body;

[0061] S3. Perform dual-path data analysis on the collected bioelectrical impedance signals. The dual-path data analysis includes frequency domain correlation analysis and time domain fluctuation analysis.

[0062] S4. Integrate the results of dual-path data analysis to determine the pain level of a specified body part;

[0063] S5. Based on the determined level of soreness, match and execute the corresponding massage program.

[0064] The overall flow of the method of the present invention is described below. After the method is started, the system initialization and detection preparation step S1 is executed first, in which the data analysis module 30 controls the mechanical execution part of the massage chair to move to the preset body part to be detected, and confirms that the bioelectrical impedance detection module 10 has established electrical contact with the user.

[0065] Next, in step S2, the data analysis module 30 generates a sinusoidal control signal of a specific frequency to drive the kneading motor of the massage chair, thereby applying periodically changing mechanical stimulation to a designated body part. During the same time period of this stimulation application, the bioelectrical impedance detection module 10 captures the dynamic changes in human bioelectrical impedance in real time through electrode plates in contact with the user's palm.

[0066] The captured raw bioelectrical impedance signal is transmitted to the signal processing module 20, which filters, amplifies, and performs analog-to-digital conversion on the signal to generate a discrete digital signal sequence, which is then sent to the data analysis module 30.

[0067] In step S3, the data analysis module 30 performs two analyses in parallel on the received digital signal sequence. The first is frequency domain correlation analysis, which calculates the frequency domain correlation index using a fast Fourier transform. Its definition is:

[0068] ;

[0069] in, It is a frequency domain correlation index; To collect bioelectrical impedance signals at excitation frequency The amplitude at that point; To apply the excitation signal to the kneading motor at the same excitation frequency S5 The amplitude at that point. This index is used to quantify the synchronous response strength of bioelectrical impedance signals and mechanical excitation at frequency.

[0070] The second is time-domain volatility analysis, which calculates the sample variance of the signal sequence. Its definition is:

[0071] ;

[0072] in, This represents the sample variance of the preprocessed signal sequence. The first one in the preprocessed discrete bioelectrical impedance signal sequence One sampling point; This is the arithmetic mean of the sequence; This represents the total number of sampling points for the sequence. This metric is used to quantify the overall fluctuation amplitude of the bioelectrical impedance signal during excitation.

[0073] Subsequently, in step S4, the data analysis module 30, according to preset judgment rules, analyzes the frequency domain correlation index. and sample variance The analysis results are combined to determine a specific level of pain, such as severe, moderate, or mild.

[0074] Finally, in step S5, the data analysis module 30 selects a corresponding massage program from the preset program library based on the final soreness level determined in step S4, and controls the mechanical actuators of the massage chair to execute it. After completing the detection and massage of the affected area, the system can repeatedly execute steps S2 to S5 to complete the detection of soreness levels in other body parts.

[0075] See attached document Figure 2 In a specific embodiment, the execution process of step S1 will be described in detail.

[0076] This step begins after the data analysis module 30 receives the pain level detection start command. This command can be generated by the user through the user interface connected to the massage chair.

[0077] Upon receiving the start command, the data analysis module 30 first executes the user interaction and hardware readiness sub-steps. The data analysis module 30 issues explicit instructions to the user via an output device, such as a display screen or voice broadcast module, prompting the user to hold the bioelectrical impedance detection module 10 in a specific posture. This specific posture requires stable and sufficient electrical contact between the user's palms and the two conductive electrode pads on the bioelectrical impedance detection module 10.

[0078] In parallel or after a user command is issued, the data analysis module 30 executes the sub-step of positioning the massage mechanism. The data analysis module 30 reads a preset sequence of detection sites from its internal memory. This sequence defines the body parts requiring pain level detection and their execution order; for example, the first detection site is the neck. The data analysis module 30 then sends a control signal to the motor driver controlling the massage mechanism, driving the massage mechanism's robotic arm to move precisely along the guide rail to the physical coordinate position corresponding to the first detection site.

[0079] Before formally proceeding to signal acquisition step S2, the data analysis module 30 must perform a system status verification sub-step to ensure the validity of subsequent data acquisition. This verification process includes verifying the contact status of the bioelectrical impedance detection module 10. Specifically, the data analysis module 30 acquires initial, static bioelectrical impedance measurements through the signal processing module 20. The system has a preset effective impedance threshold range. If the initial measurement value is outside this range, for example, appearing as infinity or close to zero, the data analysis module 30 determines that the user has not established effective electrical contact with the electrode pads and can repeatedly issue a gripping command to the user until the measurement value falls within the effective range. The upper and lower limits of the effective impedance threshold range are set to distinguish between normal human skin contact resistance and abnormal conditions. For example, the lower limit can be set to several hundred ohms to exclude the possibility of accidental short circuits in the electrode pads; the upper limit can be set to several megaohms to exclude open circuits caused by the user not making contact or poor contact. The typical bioelectrical impedance value of a dry human hand in contact with skin is in the range of tens to hundreds of kilohms, so this threshold range can be effectively used for state determination.

[0080] Simultaneously, the system's status confirmation sub-step also includes verifying the status of the massage mechanism. The data analysis module 30 queries the status information of the massage mechanism's motor driver to confirm that the mechanism has stably stopped at the designated target detection position and that the motor driver has no faults or error reports.

[0081] Step S1 is considered complete only after the contact state of the bioelectrical impedance detection module 10 and the position and state of the massage mechanism have been verified, and the entire process can then proceed to the subsequent step S2. This ensures that the application position of the mechanical excitation is accurate and that the baseline for the acquisition of the bioelectrical impedance signal is stable and effective.

[0082] See attached document Figure 2 and attached Figure 3 In a specific embodiment, the execution process of step S2 will be described in detail. This step is initiated immediately after step S1 is completed, and its core lies in applying a precisely controlled physical stimulus, and during this stimulus, synchronously and with high fidelity capturing the physiological response signals of the human body.

[0083] Step S2 first executes the generation and application of periodic mechanical excitation. The data analysis module 30, within its internal digital signal generator, generates a discrete-time domain digital sine wave signal sequence. The sequence The mathematical expression can be:

[0084] ;

[0085] in, The preset amplitude determines the maximum range of variation in massage intensity; The preset excitation frequency, for example, can be set to 0.5 Hz, which is in a low range to ensure that the human body can produce a recognizable physiological following response; The operating frequency for digital signal generation and system sampling of the data analysis module 30 can be set to, for example, 100Hz; For discrete-time indices; This is the initial phase, which can usually be set to 0.

[0086] The data analysis module 30 will analyze the numerical sequence. The signal is converted into an analog voltage signal by a digital-to-analog converter or output via pulse width modulation. This signal is then transmitted to the driver that controls the kneading motor. Based on this sinusoidally changing input signal, the motor driver adjusts the driving voltage or current supplied to the kneading motor accordingly, causing the motor's rotational speed or output torque to also change sinusoidally. Ultimately, the pressure applied to the massage head on the user's designated body part achieves a periodic, smoothly varying mechanical excitation.

[0087] During the same time period when the aforementioned mechanical excitation is applied, the system performs synchronous acquisition and conversion of bioelectrical impedance signals in parallel. The two electrode pads on the bioelectrical impedance detection module 10 form a bioelectrical measurement circuit with the user's palm skin. When the user experiences a physiological response due to massage stimulation, the overall bioelectrical impedance of this measurement circuit changes dynamically.

[0088] The changing bioelectrical impedance signal is first processed by signal processing module 20. Signal processing module 20 contains a sampling resistor that forms a voltage divider circuit with the human body, thereby linearly converting the weak bioelectrical impedance change into a voltage signal change. This voltage signal is then fed into a differential amplifier circuit or instrumentation amplifier, which uses a precise 2.5V reference voltage as a reference to suppress common-mode interference and set the DC bias point of the signal.

[0089] The signal, after differential amplification, is then fed into a second-order active low-pass filter. This filter removes high-frequency noise above a specific cutoff frequency, such as electromagnetic interference and high-frequency components of electromyographic signals, while retaining the signal at the excitation frequency. The relevant useful signal frequency band. The signal-to-noise ratio of the filtered and amplified analog signal is significantly improved.

[0090] Finally, the conditioned analog signal is input to the analog-to-digital converter (ADC) within the signal processing module 20. The ADC operates at a sampling frequency synchronized with the generation of the excitation signal. The analog signal is sampled and quantized to convert it into a discrete digital time series signal with specific bandwidth. This number sequence The dynamic response process of the human bioelectrical impedance was fully recorded during the entire periodic mechanical excitation and transmitted to the data analysis module 30 for processing in subsequent step S3. The entire execution of step S2 lasted for a preset duration. For example, 10 seconds, to ensure that a sufficient number of data points are collected for accurate frequency and time domain analysis.

[0091] Sampling frequency The selection of the sampling frequency (Hz) must satisfy the Nyquist sampling theorem, meaning its value should be greater than twice the highest frequency of the useful signal. Considering the low mechanical excitation frequency in this invention and the fact that human physiological responses are mainly concentrated in the low-frequency range, a sampling frequency of 100Hz is sufficient to capture the dynamic changes of the signal without distortion and provides ample margin for subsequent digital filtering processing. (Time length) The selection of [the appropriate method] requires, on the one hand, ensuring that a sufficient number of excitation cycles are collected (e.g., when [the excitation cycle is not specified]). =0.5Hz, =10s, five complete cycles can be collected to improve the frequency resolution and statistical stability of the results of subsequent FFT analysis; on the other hand, it is also necessary to control it within a reasonable range to avoid excessive detection time.

[0092] See attached document Figure 2 and attached Figure 3 In a specific embodiment, the execution process of step S3 will be described in detail. This step occurs when the data analysis module 30 receives the data generated in step S2, which has a length of... Bioelectrical impedance digital signal sequence Then it starts. The data analysis module 30 analyzes the sequence. The two sub-steps, frequency domain correlation analysis and time domain volatility analysis, are performed in parallel.

[0093] This analytical sub-step aims to precisely quantify the response strength of bioelectrical impedance signals to mechanical excitation at a specific frequency.

[0094] First, the data analysis module has 30 pairs of lengths. Bioelectrical impedance signal sequence Perform a Fast Fourier Transform (FFT). The FFT is an efficient algorithm for calculating the Discrete Fourier Transform, and its formula is as follows:

[0095] ;

[0096] in, It is the index of the time-domain sample. It is a discrete frequency index in the frequency domain. It corresponds to the frequency index. The frequency domain coefficients in complex form.

[0097] Simultaneously, the data analysis module 30 analyzes the ideal digital sine wave signal sequence generated internally in step S2 for driving the motor. Perform the same FFT operation to obtain its frequency domain representation. .

[0098] Next, the excitation frequency needs to be determined. The precise location in the discrete frequency domain spectrum. The corresponding frequency index. Calculated using the following formula:

[0099] ;

[0100] in, The sampling frequency of the signal. The number of points in the FFT. The function represents taking the nearest integer.

[0101] After determining the frequency index Subsequently, the data analysis module 30 respectively from and Extract the complex coefficient at that index position. and And calculate their magnitudes, i.e., amplitudes. and .

[0102] Finally, the frequency domain correlation index is calculated. This index is defined as the normalized ratio of the response amplitude of the bioelectrical impedance signal at the excitation frequency to the amplitude of the excitation signal itself:

[0103] ;

[0104] By normalizing the amplitude of the excitation signal, the influence of system parameter variations on the results can be eliminated, thus... It more purely reflects the relative intensity of the human body's physiological response. Higher The value represents the bioelectrical impedance signal at the excitation frequency. There is a significant concentration of energy at this point, which directly corresponds to the strong and synchronous physiological response of the human body to massage stimulation at this frequency.

[0105] This analytical sub-step aims to assess, holistically, the instability or fluctuation of bioelectrical impedance signals under massage stimulation.

[0106] First, the data analysis module pairs 30 original bioelectrical impedance signal sequences. Perform preprocessing to obtain a smoother signal sequence. This preprocessing can be performed using a window size of [size missing]. Median filters are effective at removing isolated impulse noise from signals while preserving edge information relatively well. The window size of a median filter... The selection of the window size requires a trade-off between noise filtering effectiveness and signal detail preservation. An excessively large window can over-smooth the signal, potentially leading to the loss of useful physiological fluctuation information; an excessively small window may fail to effectively filter noise. In this embodiment, According to the sampling frequency The selection can be made, for example, to set the number of sampling points that can cover a time span of 50 milliseconds to 100 milliseconds, so as to effectively filter out occasional spike interference without affecting the response trend caused by second-level periodic excitation.

[0107] After obtaining the preprocessed signal sequence Next, the data analysis module 30 calculates the arithmetic mean of the sequence. :

[0108] ;

[0109] Subsequently, based on the mean Calculate the sample variance of the sequence. :

[0110] ;

[0111] variance It directly quantifies the degree of dispersion of signal data points relative to their mean. In this application, a larger... The values ​​indicate that during the massage, the user's physiological state experienced drastic and irregular fluctuations, which are directly related to the increased activity of the autonomic nervous system caused by soreness and discomfort.

[0112] After completing the two sub-steps of frequency domain correlation analysis and time domain volatility analysis, the data analysis module 30 obtained two key quantitative indicators: frequency domain correlation index. and sample variance These two metrics will be used as inputs for the fusion determination in the subsequent step S4.

[0113] See attached document Figure 2 and attached Figure 4 In a specific embodiment, the execution process of step S4 will be described in detail. This step occurs after the data analysis module 30 obtains the frequency domain correlation index through step S3. and sample variance Then proceed. The core of this step lies in mapping these two numerical indicators from different analytical dimensions to the final, discrete pain level based on a set of preset and explicit logical rules.

[0114] First, the preliminary grading sub-step is performed. The data analysis module 30 reads two sets of four preset numerical thresholds from its internal memory. The first set of thresholds is used for... The division includes high threshold. and low threshold ,in The second set of thresholds is used for The division includes high threshold. and low threshold ,in These thresholds can be calibrated using prior experimental data or set based on expert experience. A specific threshold calibration method may include the following steps: recruiting multiple test subjects and having them subjectively rate the degree of soreness under different massage intensities using a standardized subjective perception scale, categorized into mild, moderate, and severe levels. Simultaneously, bioelectrical impedance signals are collected from each test subject under the corresponding conditions, and the results are calculated. and Value. Collect all test takers' (subjective ratings). value, (Value) data pairs. Finally, through statistical analysis, the optimal method for distinguishing different levels of pain was determined. .

[0115] Data analysis module 30 will calculate the results and and By comparison, a preliminary pain level is generated from frequency domain analysis. The judgment logic is as follows:

[0116] like Then determine It is severe;

[0117] like Then determine Moderate;

[0118] like Then determine It is mild.

[0119] In parallel, the data analysis module 30 will calculate the... and and By comparing the results, a preliminary pain level is generated based on time-domain analysis. The judgment logic is as follows:

[0120] like Then determine S5 indicates severe severity.

[0121] like Then determine It is moderate.

[0122] like Then determine It is mild.

[0123] After obtaining two initial soreness levels and Then, the sub-step of the fusion judgment logic is executed to obtain a single, final pain level. The purpose of this fusion judgment is to improve the accuracy and robustness of the judgment results through double verification, and to avoid misjudgments caused by individual physiological differences or transient interference due to a single indicator. The fusion rule is defined as a set of strict conditional statements:

[0124] Severe aches and pains assessment:

[0125] If and only if The diagnosis result is severe S5 and S5 When the assessment result is also severe, the final pain level is... It was determined to be severe aches and pains. This logic ensures that the strongest aches and pains level can only be determined when the signal simultaneously exhibits a strong response highly synchronized with the stimulus and dramatic overall fluctuations.

[0126] Mild aches and pains assessment:

[0127] If and only if The diagnosis was mild and When the assessment result is also mild, the final pain level is... It was determined to be mild soreness. This, along with the logic, ensures that the weakest level of soreness can only be determined when both the synchronous response of the signal and the overall fluctuation are weak.

[0128] Moderate aches and pains assessment:

[0129] In all other cases where the above criteria for severe and mild aches and pains are not met, the final aches and pains level is determined as follows: The pain was determined to be moderate. This encompasses seven of the nine possible combinations, excluding (severe, S5 severe) and (mild, S5 mild), such as (severe, S5 moderate), (severe, S5 mild), (moderate, S5 severe), (moderate, S5 moderate), (moderate, S5 mild), (mild, S5 severe), and (mild, S5 moderate). This assessment strategy categorizes all non-extreme cases with discrepancies in indicators into the intermediate level, reflecting a more robust and prudent classification principle.

[0130] By executing the above-mentioned fusion judgment logic, the data analysis module 30 generates a unique and definitive final pain level for the currently detected body part. This result will be used for personalized procedure matching in the subsequent step S5.

[0131] See attached document Figure 2 In a specific embodiment, the execution process of step S5 will be described in detail. This step occurs after the data analysis module 30 determines the final soreness level of the current detection site through step S4. Start up afterward.

[0132] First, the sub-step of program mapping is executed. The data analysis module 30 has a preset pain level-massage program mapping table stored in its internal memory. This table is a data structure, such as a lookup table or hash table, that establishes a one-to-one correspondence between pain levels and specific massage program parameter sets.

[0133] The massage program parameter set may include, but is not limited to, the following parameters:

[0134] Massage techniques: Define the basic movements performed by the massage mechanism, such as kneading, tapping, rubbing, acupressure, patting, etc.

[0135] Massage intensity: Defines the level of pressure applied to the body by the massage head, such as adjustable from 1 to 5.

[0136] Massage speed / frequency: Defines the speed or frequency of massage movements;

[0137] Massage range: Defines the range of motion of the massage mechanism on this part of the body;

[0138] Auxiliary functions: Define whether to enable additional functions, such as heating, airbag compression, etc.

[0139] For example, the mapping table can be defined as follows:

[0140] like For severe soreness, the mapped parameter set is: {Technique: S5 deep kneading + continuous acupressure, S5 intensity: S5 level 4, S5 speed: S5 slow, S5 range: S5 key area small range, S5 auxiliary function: S5 heating on}.

[0141] like S5 moderate soreness, the mapped parameter set is: {Technique: S5 regular kneading + rhythmic tapping, S5 intensity: S5 level 3, S5 speed: S5 medium speed, S5 range: S5 standard range, S5 auxiliary function: S5 heating off};

[0142] like For mild soreness, the mapped parameter set is: {Technique: S5 gentle massage + soothing patting, S5 intensity: S5 level 2, S5 speed: S5 fast, S5 range: S5 large range, S5 auxiliary function: S5 heating off}.

[0143] Data analysis module 30 Used as a query key, the corresponding massage program parameter set is retrieved from the mapping table.

[0144] After obtaining the parameter set, the execution program performs sub-steps of the process loop. The data analysis module 30 parses the retrieved parameter set into a series of specific control commands and sends them sequentially to the controllers of various subsystems in the massage chair. For example, it sends commands to the motor driver of the massage mechanism to set the technique, intensity, and speed, and sends an activation command to the heating module controller. Upon receiving the commands, each controller drives the corresponding actuator to begin performing a targeted massage program on the current body part. This massage program lasts for a preset duration.

[0145] In the complete whole-body soreness detection process, after the targeted massage of the current area is completed or at the end of its execution phase, the system will automatically enter the next cycle. The data analysis module 30 will query the detection area sequence loaded in step S1 to determine the next body area to be detected. Subsequently, the system process will return to step S2 and re-execute the entire process of periodic excitation application, signal acquisition, data analysis, level determination, and program matching on the new area.

[0146] This cycle will continue until all areas in the detection sequence have been detected and treated with targeted massage. At this point, the complete and personalized whole-body aches and pains detection and relief process is finished. The system can store the generated aches and pains levels for each area of ​​the body for user health tracking or to generate subsequent comprehensive massage plans. The aches and pains map is a data structure that associates the identifier of a body part with the corresponding final aches and pains level. The data is stored in a correlated manner. This map can not only be used to display the results after the current massage session, but also to form a time series after multiple massages, which can be used to track the long-term trend of soreness in specific areas. In addition, the system can design a comprehensive global massage plan based on the distribution of the entire map. For example, it can prioritize longer, deeper massages for areas with the highest soreness levels, while providing shorter, more routine relaxation massages for areas with lower soreness levels, thereby achieving intelligent scheduling of massage resources.

[0147] The steps S1 to S5 of the method and system for detecting human pain intensity based on changes in human bioelectrical impedance provided by the present invention have been described in detail above through specific embodiments. This embodiment applies a sinusoidal mechanical excitation of a specific frequency, simultaneously acquires bioelectrical impedance signals from the distal parts of the human body, and combines a dual-path data processing strategy of frequency domain correlation analysis and time domain fluctuation analysis. Finally, through fusion judgment, it achieves quantitative grading and personalized response to the pain intensity of specific parts of the body.

[0148] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Those skilled in the art, inspired by the technical concepts disclosed in this invention, can make various equivalent changes or modifications.

[0149] For example, the waveform of periodic mechanical excitation is not limited to a sine wave; it can also be a triangular wave, square wave, sawtooth wave, or any other non-sinusoidal periodic waveform that can elicit a periodic physiological response. Correspondingly, when performing frequency domain correlation analysis, in addition to analyzing the fundamental frequency response, the response intensity of its harmonic components can also be analyzed. Excitation frequency It is not limited to 0.5Hz; it can be selected within the frequency range that can produce a stable physiological following effect in the human body, such as the range of 0.1Hz to 2Hz.

[0150] Similarly, frequency domain correlation analysis is not limited to using Fast Fourier Transform to obtain frequency domain amplitude. Any technique capable of measuring the correlation strength between two signals at a specific frequency, such as calculating the cross-correlation between a signal and a standard sine and cosine reference signal in the time domain, or employing signal processing methods based on the lock-in amplification principle, can be used to achieve the objectives of this invention. Time domain volatility analysis is also not limited to calculating sample variance; other statistics that can characterize the degree of dispersion or volatility of a signal, such as calculating the standard deviation, mean absolute deviation, or root mean square value of the AC component of the signal, are also equivalent technical means of this invention.

[0151] Furthermore, the fusion judgment logic is not limited to the hard threshold condition judgment rule disclosed in this embodiment. This logic could be a weighted sum-based scoring mechanism, that is, weighting and summing the normalized frequency domain indicators and time domain indicators, and classifying the levels based on the total score. This logic could also be based on a decision tree model or a simple classifier obtained from training data, which uses frequency domain correlation indicators and time domain volatility indicators as input features and outputs the pain level.

[0152] In summary, any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this invention should be included within the scope of protection of this invention.

[0153] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for detecting the degree of human pain based on changes in human bioelectrical impedance, characterized in that, include: S1. Apply periodic mechanical stimulation to a designated body part; S2. Synchronously collect human bioelectrical impedance signals; S3. Perform dual-path data analysis on the bioelectrical impedance signal to obtain frequency domain correlation index and time domain fluctuation index; S4. By integrating the frequency domain correlation index and the time domain volatility index, the pain level of the specified body part is determined. S5. Match and execute the corresponding massage program according to the soreness level; The frequency domain correlation analysis includes: Perform a Fourier transform on the bioelectrical impedance signal to obtain the amplitude of the bioelectrical impedance signal at the excitation frequency; Perform a Fourier transform on the excitation signal of the periodic mechanical excitation to obtain the amplitude of the excitation signal at the excitation frequency; The amplitude of the bioelectrical impedance signal at the excitation frequency is normalized to the amplitude of the excitation signal at the excitation frequency to obtain the frequency domain correlation index.

2. The method for detecting the degree of human pain based on changes in human bioelectrical impedance according to claim 1, characterized in that, The application of periodic mechanical excitation includes: A sinusoidal control signal is generated, which is used to drive a motor, and the motor generates the periodic mechanical excitation.

3. The method for detecting the degree of human pain based on changes in human bioelectrical impedance according to claim 1, characterized in that, After collecting human bioelectrical impedance signals and before performing dual-path data analysis, the following is also included: The bioelectrical impedance signal is filtered, amplified, and converted from analog to digital to obtain a digital time series signal.

4. The method for detecting the degree of human pain based on changes in human bioelectrical impedance according to claim 1, characterized in that, The time-domain volatility analysis includes: Calculate the arithmetic mean of the bioelectrical impedance signal; Calculate the sum of squares of the differences between each sampling point in the bioelectrical impedance signal and the arithmetic mean; The time-domain volatility index is obtained based on the sum of squares.

5. The method for detecting the degree of human pain based on changes in human bioelectrical impedance according to claim 4, characterized in that, Prior to the time-domain volatility analysis, the following is also included: The bioelectrical impedance signal is preprocessed by median filtering.

6. The method for detecting the degree of human pain based on changes in human bioelectrical impedance according to claim 1, characterized in that, The fusion includes: The frequency domain correlation index is compared with a first preset threshold to obtain a preliminary frequency domain level. The time-domain volatility index is compared with a second preset threshold to obtain a preliminary time-domain level. The pain level is determined based on the preliminary frequency domain level and the preliminary time domain level.

7. The method for detecting the degree of human pain based on changes in human bioelectrical impedance according to claim 6, characterized in that, Determining the level of soreness includes: When both the preliminary frequency domain level and the preliminary time domain level are severe, the aches and pains are determined to be severe aches and pains. When both the preliminary frequency domain level and the preliminary time domain level are mild, the aches and pains are determined to be mild aches and pains.

8. The method for detecting the degree of human pain based on changes in human bioelectrical impedance according to claim 7, characterized in that, Determining the level of soreness also includes: If the criteria for severe and mild aches and pains are not met, the aches and pains are classified as moderate aches and pains.

9. A system for detecting human pain intensity based on changes in human bioelectrical impedance, comprising, according to any one of claims 1-8, a method for detecting human pain intensity based on changes in human bioelectrical impedance, including: Bioelectrical impedance detection module (10) is used to collect human bioelectrical impedance signals; Mechanical actuators are used to apply periodic mechanical excitation to designated body parts; The data analysis module (30) is electrically connected to the bioelectrical impedance detection module (10) and the mechanical actuation component; The data analysis module (30) is configured to: Synchronous control of the excitation applied to the mechanical actuator and the signal acquisition of the bioelectrical impedance detection module (10); Dual-path data analysis was performed on the bioelectrical impedance signal to obtain frequency domain correlation index and time domain fluctuation index; By combining the frequency domain correlation index and the time domain volatility index, the pain level of the specified body part is determined. Based on the level of soreness, the mechanical actuator is controlled to perform the corresponding massage program.

Images