Evaluation method and system for massage effect

Abstract

The invention discloses a massage effect evaluation method and system, and the method comprises the following steps: S1, data collection: obtaining a plurality of physiological parameters; s2, data preprocessing: carrying out filtering, normalization and feature extraction processing on the physiological parameters; s3, effect evaluation: inputting the data after feature extraction into an effect evaluation model, and adaptively outputting an evaluation result by the effect evaluation model according to the massage manipulation; and S4, optimization iteration: optimizing the effect evaluation model by adopting a gradient descent method so as to improve the evaluation effect. The massage effect is evaluated by fusing the EMG feature, the skin temperature feature, the pressure distribution feature, the HRV feature and other parameter data, and the evaluation effect is more accurate.

Description

Technical Field

[0001] The present invention relates to the technical field of traditional Chinese medicine massage, and particularly to an evaluation method and system for massage effects. Background Art

[0002] Massage has the functions of warming and dredging meridians, relieving muscle spasms, relaxing and relieving pain, etc. Manipulative massage can correct the anatomical shape and position of displaced soft tissues or spinal small joints, restore the abnormal line of force to normal, adjust the internal and external balance, relieve clinical symptoms such as pain and numbness, improve blood circulation, enhance muscle strength, improve the elasticity of ligaments and eliminate inflammation. Indirect massage techniques such as rubbing and kneading on the outside of the spinal cord can affect the spinal cord and nerve roots to produce friction, expecting to improve microcirculation, promote edema washing and relieve adhesions, so that the compressed nerves can be repaired and the dysfunction can be alleviated and disappeared.

[0003] However, the existing evaluation of massage effects mainly relies on the judgment of doctors' experience or the subjective feedback of patients, lacking objective quantitative indicators. Some studies have tried to use a single physiological parameter (such as electromyogram signal) for evaluation, but there are the following problems: 1. A single physiological parameter cannot comprehensively reflect the massage effect; 2. Lack of a dynamic feedback mechanism and unable to adjust the evaluation model according to real-time data; 3. The differential effects of different massage techniques and intensities during the massage process are not considered. Summary of the Invention

[0004] In view of the above technical problems, the present invention provides an evaluation method and system for massage effects.

[0005] The present invention is implemented by the following technical solutions: An evaluation method for massage effects includes the following steps: S1: Data collection, obtaining a variety of physiological parameters; S2: Data preprocessing, performing filtering, normalization, and feature extraction on the physiological parameters; S3: Effect evaluation, inputting the data after feature extraction into an effect evaluation model, and the effect evaluation model adaptively outputs an evaluation result according to the massage technique; S4: Optimization and iteration, using the gradient descent method to optimize the effect evaluation model to improve the evaluation effect.

[0006] Further, the data collection includes: Obtaining electromyogram signal data through an EMG sensor; Obtaining local skin temperature change data through an infrared thermal imaging module; Obtaining massage force and distribution data through a flexible pressure sensor; Obtaining heart rate variability data through a heart rate monitoring module.

[0007] Further, step S2 includes the following sub-steps: S21: Filter the obtained physiological parameters by using one or more of low-pass filtering, and / or high-pass filtering, and / or band-pass filtering, and / or moving average filtering, and / or Gaussian smoothing filtering; S22: Normalize the filtered data to eliminate the amplitude differences between different signals; S23: Extract features from the normalized data, and respectively extract EMG features, skin temperature features, pressure distribution features, and HRV features among the physiological parameters.

[0008] Further, step S23 specifically includes the following sub-steps: Extract EMG features, that is, extract the root mean square value: ; In the formula, M represents the number of samples, and EMG i represents the value of the i-th sample; Extract skin temperature features, that is, extract the temperature change rate: ; In the formula, T2 represents the temperature after massage, T1 represents the temperature before massage, and t represents the time window; Extract pressure distribution features, that is, extract the pressure entropy value: ; In the formula, represents the value of the i-th pressure point, represents the total pressure; Extract HRV features, that is, extract the SDNN standard deviation: ; In the formula, N represents the total number of heart rate intervals, represents the i-th heart rate interval, represents the average value of all heart rate intervals.

[0009] Further, step S3 includes the following sub-steps: S31: Input the EMG feature data, skin temperature feature data, pressure distribution feature data, and HRV feature data after feature extraction into the effect evaluation model respectively; S32: The effect evaluation model adaptively outputs an evaluation value according to the massage technique: ; In the formula, represents the EMG feature data, represents the skin temperature feature data, Represents pressure distribution characteristic data, Represents HRV characteristic data, Both are dynamic weight coefficients; S33: Compare the evaluation value S with the scoring standard S', and output the evaluation result, where the evaluation result includes poor, good, and excellent.

[0010] Furthermore, the effect evaluation model adaptively adjusts the dynamic weight coefficients according to different massage techniques , and the massage techniques include pressing method, kneading method, pushing method, and pulling method.

[0011] Furthermore, step S4 is specifically: after each evaluation, compare the evaluation result with the actual evaluation data, and optimize the weight coefficients of the effect evaluation model by the gradient descent method: ; wherein, represents the learning rate, represents the actual evaluation data, S represents the evaluation value, and X represents the feature vector.

[0012] An evaluation system for massage effect, used to implement the above-mentioned evaluation method for massage effect, including a data acquisition module, a data preprocessing module, an effect evaluation module, and an optimization iteration module. Among them, the data acquisition module is used to obtain various physiological parameters; the data preprocessing module is used to filter, normalize, and extract features from the physiological parameters; the effect evaluation module is used to input the data after feature extraction into the effect evaluation model, and the effect evaluation model adaptively outputs the evaluation result according to the massage technique; the optimization iteration module is used to optimize the effect evaluation model by the gradient descent method to improve the evaluation effect.

[0013] The beneficial effects of the present invention are as follows: The present invention evaluates the massage effect by fusing parameter data such as EMG features, skin temperature features, pressure distribution features, and HRV features, and the evaluation effect is more accurate; and combined with the dynamic weight adjustment mechanism, it overcomes the defect of poor adaptability of the traditional fixed weight model, and also dynamically optimizes the sensitivity and specificity of the evaluation model by comparing real-time data with actual data, realizing the continuous optimization of the evaluation model. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained according to the structures shown in these drawings.

[0015] Figure 1 This is the flow chart of the present invention. Detailed implementation manners

[0016] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some but not all of the embodiments of the present invention. Components of the embodiments of the present invention generally described and illustrated in the drawings herein can be arranged and designed in a variety of different configurations.

[0017] It should be noted that like reference numerals and letters denote like items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings.

[0018] The following will, with reference to the accompanying drawings, elaborate on some implementation manners of the present invention. Without conflict, the following embodiments and the features in the embodiments can be combined with each other.

[0019] See Figure 1 , an evaluation method for tuina effect, comprising the following steps: S1: Data collection, obtaining a variety of physiological parameters; S2: Data preprocessing, performing filtering, normalization and feature extraction on the physiological parameters; S3: Effect evaluation, inputting the data after feature extraction into an effect evaluation model, and the effect evaluation model adaptively outputs an evaluation result according to the tuina technique; S4: Optimization iteration, using the gradient descent method to optimize the effect evaluation model to improve the evaluation effect.

[0020] In this embodiment, the data collection includes: obtaining electromyogram (EMG) signal data through an EMG sensor, such as obtaining EMG signal data through a wearable EMG sensor (attached to the target muscle group); obtaining local skin temperature change data through an infrared thermal imaging module; obtaining tuina force and distribution data through a flexible pressure sensor; obtaining heart rate variability (HRV) data through a heart rate monitoring module.

[0021] In this embodiment, step S2 includes the following sub-steps: S21: Using one or more of low-pass filtering and / or high-pass filtering and / or band-pass filtering and / or moving average filtering and / or Gaussian smoothing filtering to filter the obtained physiological parameters; S22: Performing normalization processing on the filtered data to eliminate the amplitude difference between different signals, thereby improving the accuracy of feature extraction; S23: Perform feature extraction on the normalized data, and respectively extract EMG features, skin temperature features, pressure distribution features, and HRV features from the physiological parameters.

[0022] Specifically, for example, bandpass filtering (20-500Hz) can be used to remove motion artifacts for EMG data, sliding average filtering can be used for skin temperature change data, Gaussian smoothing filtering can be used for massage intensity and distribution data, and low-pass filtering can be used for heart rate variability HRV data to remove abnormal RR intervals (such as threshold method: remove data points that exceed the mean ± 3 times the standard deviation). Furthermore, feature extraction is performed on the normalized data as follows: Extract EMG features, that is, extract the root mean square value: ; In the formula, M represents the number of samples, EMG i represents the value of the i-th sample; Extract skin temperature features, that is, extract temperature change rate: ; In the formula, T2 represents the temperature after massage, T1 represents the temperature before massage, and t represents the time window; Extract the pressure distribution characteristics, that is, extract the pressure entropy value (reflecting the uniformity of force): ; In the formula, represents the value of the i-th pressure point, Indicates total pressure; Extract HRV features, that is, extract SDNN standard deviation, as an indicator of sympathetic nerve activity: ; Where N is the total number of heartbeat intervals. represents the i-th heartbeat interval, Represents the average value of all heartbeat intervals.

[0023] In this embodiment, step S3 includes the following sub-steps: S31: inputting the extracted EMG feature data, skin temperature feature data, pressure distribution feature data and HRV feature data into the effect evaluation model respectively; S32: The effect evaluation model outputs evaluation values ​​according to the adaptive massage technique: ; In the formula, Represents EMG feature data, Represents skin temperature characteristic data, Represents pressure distribution characteristic data, Represents HRV feature data, All are dynamic weight coefficients, and the initial values can be set according to the manipulation type; S33: Compare the evaluation value S with the scoring standard S', and output the evaluation result, where the evaluation result includes poor, good, and excellent.

[0024] Specifically, the scoring standard S' can be set to 2.75 and 4.25 respectively. When the evaluation value S is less than 2.75, the output evaluation result is poor; when the evaluation value S is greater than or equal to 2.75 and less than 4.25, the output evaluation result is good; when the evaluation value S is greater than or equal to 4.25, the output evaluation result is excellent. It can be imagined that the scoring standard S' can also be adjusted according to the actual situation, such as set to 2, 4, etc.

[0025] In this embodiment, the effect evaluation model adaptively adjusts the dynamic weight coefficients according to different massage manipulations , and the massage manipulations include pressing method, kneading method, pushing method, and pulling method. For example, for the pressing method, any one of the dynamic weight coefficients can be increased or decreased; similarly, for the kneading method, pushing method, and pulling method, any one of the dynamic weight coefficients can be increased or decreased.

[0026] For example: when targeting the pressing method, the dynamic weight coefficient can be increased by 10%, the dynamic weight coefficient can be increased by 15%, the dynamic weight coefficient can be decreased by 5%, and the dynamic weight coefficient can be decreased by 10%.

[0027] Or, when targeting the kneading method, the dynamic weight coefficient can be decreased by 7%, the dynamic weight coefficient can be increased by 12%, the dynamic weight coefficient can be increased by 10%, and the dynamic weight coefficient can be decreased by 5%.

[0028] Or, when targeting the pushing method, the dynamic weight coefficient can be increased by 8%, the dynamic weight coefficient can be decreased by 6%, the dynamic weight coefficient can be decreased by 5%, and the dynamic weight coefficient can be decreased by 5%.

[0029] Or, when targeting the pulling method, the dynamic weight coefficient can be decreased by 5%, the dynamic weight coefficient can be decreased by 10%, and the dynamic weight coefficient 6%, increase the dynamic weight coefficient 5%.

[0030] In this embodiment, step S4 is specifically as follows: After each evaluation, compare the evaluation result with the actual evaluation data, and optimize the weight coefficient of the effect evaluation model by the gradient descent method: ; Among them, represents the learning rate, represents the actual evaluation data, S represents the evaluation value, and X represents the feature vector.

[0031] The present invention also provides an evaluation system for the massage effect to implement the above-mentioned evaluation method for the massage effect, including a data acquisition module, a data preprocessing module, an effect evaluation module, and an optimization iteration module. Among them, the data acquisition module is used to obtain various physiological parameters; the data preprocessing module is used to perform filtering, normalization, and feature extraction processing on the physiological parameters; the effect evaluation module is used to input the data after feature extraction into the effect evaluation model, and the effect evaluation model adaptively outputs an evaluation result according to the massage technique; the optimization iteration module is used to optimize the effect evaluation model by using the gradient descent method to improve the evaluation effect.

[0032] Taking the example of kneading massage on the subject's waist for 10 minutes Respectively obtain the physiological parameters of the subject, including EMG features, skin temperature features, pressure distribution features, and HRV features. After filtering, normalization, and feature extraction, we get: RMS = 0.25, ΔT = 1.8, Hp = 0.62, SDNN = 35.

[0033] Select the kneading method weight coefficient (α = 0.3, β = 0.4, γ = 0.2, δ = 0.1).

[0034] The comprehensive score S = 0.3 * 0.25 + 0.4 * 1.8 + 0.2 * (1 - 0.62) + 0.1 * 35 = 2.14.

[0035] Since 2.14 is less than 2.75, the output evaluation result is poor.

[0036] Compare the output result with the actual clinical evaluation result (if it is 2.5), and update the weight coefficient according to the following formula: Δθ = 0.01 * (2.5 - 2.14) * X.

[0037] The present invention evaluates the massage effect by fusing parameter data such as EMG features, skin temperature features, pressure distribution features, and HRV features, and the evaluation effect is more accurate. Moreover, by combining a dynamic weight adjustment mechanism, the defect of poor adaptability of the traditional fixed-weight model is overcome. Additionally, by comparing real-time data with actual data, the sensitivity and specificity of the evaluation model are dynamically optimized, realizing the continuous optimization of the evaluation model.

[0038] For the foregoing embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that the present application is not limited by the described action sequence, because according to the present application, some steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are preferred embodiments, and the actions involved are not necessarily essential to the present application.

[0039] In the above embodiments, the basic principles, main features, and advantages of the present invention are described. Those skilled in the art of this industry should understand that the present invention is not limited by the above embodiments. What is described in the above embodiments and the specification only illustrates the principle of the present invention. Without departing from the spirit and scope of the present invention, any modifications and changes made by those skilled in the art that do not depart from the spirit and scope of the present invention shall fall within the protection scope of the appended claims of the present invention.

Claims

1. A method for evaluating massage effect, characterized in that: The steps include: S1: Data acquisition, obtaining various physiological parameters; S2: Data preprocessing, filtering, normalization and feature extraction of physiological parameters; S3: Effect evaluation, inputting the feature-extracted data into an effect evaluation model, and the effect evaluation model outputs evaluation results according to the massage technique; S4: Optimize iteration and use the gradient descent method to optimize the effect evaluation model to improve the evaluation effect.

2. A method for evaluating massage effect according to claim 1, characterized in that: The data collection includes: Obtain electromyographic signal data through EMG sensors; Acquire local skin temperature change data through infrared thermal imaging module; Obtain massage force and distribution data through flexible pressure sensors; Obtain heart rate variability data through the heart rate monitoring module.

3. A method for evaluating massage effect as claimed in claim 1, characterized in that: Step S2 includes the following sub-steps: S21: filtering the acquired physiological parameters by using one or more of low-pass filtering, high-pass filtering, band-pass filtering, sliding average filtering, and / or Gaussian smoothing filtering; S22: performing normalization processing on the filtered data to eliminate the amplitude differences between different signals; S23: Perform feature extraction on the normalized data, and respectively extract EMG features, skin temperature features, pressure distribution features, and HRV features from the physiological parameters.

4. A method for evaluating massage effect as claimed in claim 3, characterized in that: Step S23 specifically includes the following sub-steps: Extract EMG features, that is, extract the root mean square value: ； Where M represents the number of samples, EMG i Represents the value of the i-th sample; Extract skin temperature features, that is, extract temperature change rate: ； In the formula, T2 represents the temperature after massage, T1 represents the temperature before massage, and t represents the time window; Extract the pressure distribution characteristics, that is, extract the pressure entropy value: ； In the formula, represents the value of the i-th pressure point, Indicates total pressure; Extract HRV features, that is, extract SDNN standard deviation: ； Where N is the total number of heartbeat intervals. represents the i-th heartbeat interval, Represents the average value of all heartbeat intervals.

5. A method for evaluating massage effect as claimed in claim 1, characterized in that: Step S3 includes the following sub-steps: S31: inputting the extracted EMG feature data, skin temperature feature data, pressure distribution feature data and HRV feature data into the effect evaluation model respectively; S32: The effect evaluation model outputs evaluation values ​​according to the adaptive massage technique: ； In the formula, Represents EMG feature data, Represents skin temperature characteristic data, Represents pressure distribution characteristic data, Represents HRV feature data, and All are dynamic weight coefficients; S33: Compare the evaluation value S with the scoring standard S' and output an evaluation result, which includes poor, good and excellent.

6. A method for evaluating massage effect as claimed in claim 5, characterized in that: The effect evaluation model adaptively adjusts the dynamic weight coefficient according to different massage techniques. The massage techniques include pressing, kneading, pushing and pulling.

7. A method for evaluating massage effect as claimed in claim 1, characterized in that: Step S4 is specifically as follows: after each evaluation is completed, the evaluation result is compared with the actual evaluation data, and the weight coefficient of the effect evaluation model is optimized by the gradient descent method: ； in, represents the learning rate, represents the actual evaluation data, S represents the evaluation value, and X represents the feature vector.

8. A massage effect evaluation system, used to implement a massage effect evaluation method according to any one of claims 1 to 7, characterized in that: It includes a data acquisition module, a data preprocessing module, an effect evaluation module and an optimization iteration module, wherein the data acquisition module is used to obtain a variety of physiological parameters; the data preprocessing module is used to filter, normalize and extract features from the physiological parameters; the effect evaluation module is used to input the data after feature extraction into the effect evaluation model, and the effect evaluation model outputs the evaluation result adaptively according to the massage technique; the optimization iteration module is used to optimize the effect evaluation model by using the gradient descent method to improve the evaluation effect.

Images