Bio-electricity signal real-time monitoring method and system based on massage therapy
By performing timing decomposition and characteristic analysis of the electromyography during massage, the monitoring accuracy problem caused by interference from external factors is solved, and more accurate electromyography signal monitoring and massage status feedback are achieved.
Patent Information
- Application Number
- CN202510963569.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-14
- Publication Date
- 2025-08-08
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
During massage treatment, the monitoring accuracy of electromyography signals caused by interference from external factors is low, and the feedback of monitoring results is inaccurate, making it difficult to accurately identify muscle status.
By obtaining the electromyography during massage, performing timing decomposition to obtain periodic terms and residual terms, analyzing the discrete characteristics and suspected abnormalities of the residual terms, combining neighborhood change characteristics and periodic distribution characteristics, the electromyography is corrected to improve monitoring accuracy.
It improves the monitoring accuracy of bioelectric signals and the accuracy of monitoring results feedback, reduces the impact of external factors on monitoring results, and ensures accurate identification of massage status.
Smart Images
Figure CN120452700A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of medical technology, and in particular to a method and system for real-time monitoring of bioelectric signals based on massage therapy. Background Art
[0002] Massage therapy refers to a method in which a doctor uses his or her hands to apply different forces and techniques to stimulate certain specific parts of the patient's body in order to restore or improve the physical condition and promote recovery from illness. Existing technologies can monitor the patient's bioelectric signals, such as electromyography (EMG), during massage therapy, thereby helping doctors identify muscle status, avoid relying solely on subjective palpation, prevent excessive compensatory muscle contraction caused by excessive manipulation, and reduce the risk of soft tissue injury during massage therapy.
[0003] When monitoring the electromyogram (EMG) during massage, it may be interfered with by external factors, such as motion artifacts between the electrode and the skin, or impedance changes caused by sweating and oil secretion on the skin, which may cause abnormal monitoring signals. Such abnormal signals can easily be confused with reflex muscle tension signals caused by improper massage, ultimately resulting in low accuracy of the monitored bioelectric signals and inaccurate massage status feedback from the monitoring results. Summary of the Invention
[0004] In order to solve the above-mentioned technical problems of low accuracy of monitored bioelectrical signals and inaccurate massage status feedback due to the influence of external factors during massage, the purpose of the present invention is to provide a real-time monitoring method and system for bioelectrical signals based on massage therapy. The technical solutions adopted are as follows: Obtaining electromyogram of the massage subject during the massage process; Decomposing the electromyogram in time series to obtain a period term and a residual term; obtaining a suspected abnormality degree based on a discrete feature of data distribution in the residual term; and obtaining a suspected abnormality moment based on the suspected abnormality degree; Obtaining a neighborhood change characteristic value in the residual term based on the residual distance feature between the suspected abnormal moment and the adjacent moments; obtaining a period length segment of the period term based on the spectrum feature of the period term; obtaining a period distribution characteristic value based on the time interval feature between the suspected abnormal moment in the period term and other suspected abnormal moments in other period length segments, and the quantity feature of other suspected abnormal moments corresponding to the suspected abnormal moment in each period length segment; The abnormality degree of the electromyographic activity at the suspected abnormal moment is obtained according to the neighborhood change characteristic value and the period distribution characteristic value; and the electromyogram is corrected according to the abnormality degree of the electromyographic activity.
[0005] Furthermore, the step of obtaining the suspected abnormality degree according to the discrete characteristics of the data distribution in the residual term includes: The residual points in the residual term are divided into positive residual points and negative residual points; among the positive residual points, the minimum value of the ratio of the number of positive residual points on both sides of the average line of the positive residual points is calculated to obtain a first quantity ratio; among the negative residual points, the minimum value of the ratio of the number of negative residual points on both sides of the average line of the negative residual points is calculated to obtain a second quantity ratio; the product of the first quantity ratio and the second quantity ratio is calculated and negatively correlated to obtain the suspected abnormality degree.
[0006] Furthermore, the step of obtaining the suspected abnormality moment according to the suspected abnormality degree includes: When the suspected abnormality degree exceeds a preset first threshold, the moments of all residual points above the average line of the positive residual points and below the average line of the negative residual points are taken as suspected abnormality moments.
[0007] Furthermore, the step of obtaining a neighborhood change characteristic value in the residual term according to the residual distance feature between the suspected abnormal moment and the adjacent moment includes: Calculate the Euclidean distance between the residual point of the suspected abnormal moment and the previous adjacent moment to obtain a first distance; calculate the Euclidean distance between the residual point of the suspected abnormal moment and the subsequent adjacent moment to obtain a second distance; calculate the inverse of the sum of the first distance and the second distance to obtain a degree of stability; calculate the product of the absolute value of the difference between the first distance and the second distance and the degree of stability to obtain a neighborhood change characteristic value of the suspected abnormal moment.
[0008] Furthermore, the step of obtaining the period length segment of the periodic term according to the spectrum characteristics of the periodic term includes: The periodic term is subjected to Fourier transformation to obtain spectrum data; the inverse of the frequency corresponding to the maximum value in the spectrum data is used as a segmentation length, and the periodic term is divided according to the segmentation length to obtain different periodic length segments.
[0009] Furthermore, the step of obtaining a periodic distribution characteristic value according to the time interval characteristics between the suspected abnormal moment in the periodic item and other suspected abnormal moments in other periodic length segments and the quantity characteristics of other suspected abnormal moments corresponding to the suspected abnormal moment in each periodic length segment includes: Take the moment of the suspected abnormal moment in the period length segment as the target moment, calculate the proportion of other suspected abnormal moments in all target moments, and obtain the first similarity characteristic value; calculate the average value of the phase difference between other suspected abnormal moments closest to the target moment in all period length segments and the target moment, and obtain the average phase difference; calculate the inverse of the average phase difference and the average value of the first similarity characteristic value to obtain the periodic distribution characteristic value of the suspected abnormal moment.
[0010] Furthermore, the step of obtaining the abnormality degree of myoelectric activity at the suspected abnormal moment according to the neighborhood change characteristic value and the period distribution characteristic value includes: The average value of the neighborhood change characteristic value and the period distribution characteristic value is calculated to obtain the abnormality degree of the electromyographic activity at the suspected abnormal moment.
[0011] Furthermore, the step of correcting the electromyogram according to the abnormality of the electromyographic activity includes: The suspected abnormal moment when the electromyographic activity abnormality exceeds a preset second threshold is used as the moment to be corrected; the average value of the adjacent moments before and after the moment to be corrected is calculated in the electromyogram to obtain a correction value; and the correction value replaces the value of the moment to be corrected.
[0012] The present invention also proposes a real-time monitoring system for bioelectric signals based on massage therapy, comprising a memory, a processor, and a computer program stored in the memory and runnable on the processor. The processor executes the computer program to implement any one of the steps of a real-time monitoring method for bioelectric signals based on massage therapy.
[0013] The present invention has the following beneficial effects: In the present invention, obtaining the periodic term and the residual term can be used to determine the moment when the electromyogram (EMG) is abnormal; obtaining the suspected abnormality degree can determine whether there is a moment when the electromyogram (EMG) is abnormal based on the discrete characteristics of the data distribution in the residual term; obtaining the suspected abnormal moment can determine the moment when the electromyogram (EMG) may be abnormal. Since the changing characteristics of the electromyogram abnormality caused by different factors are different, obtaining the neighborhood change characteristic value can reflect the cause of the suspected abnormal moment based on the residual distance characteristics between the suspected abnormal moment and the adjacent moment. Since the periodic characteristics of the electromyogram abnormality caused by different reasons are different, obtaining the periodic distribution characteristic value can reflect the cause of the suspected abnormal moment based on the distribution characteristics of the suspected abnormal moment. Obtaining the electromyogram activity abnormality degree can accurately reflect the cause of the suspected abnormal moment through multiple features, thereby improving the accuracy of judgment; finally, the electromyogram (EMG) is corrected based on the electromyogram activity abnormality degree, so that the corrected electromyogram can accurately represent the changes in the real electromyogram signal, thereby improving the accuracy of bioelectric signal monitoring and the accuracy of massage status feedback from monitoring results. BRIEF DESCRIPTION OF THE DRAWINGS
[0014] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.
[0015] Figure 1 This is a flow chart of a method for real-time monitoring of bioelectric signals based on massage therapy provided by one embodiment of the present invention. DETAILED DESCRIPTION
[0016] To further illustrate the technical means and effects of the present invention to achieve the intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail a method and system for real-time monitoring of bioelectric signals based on massage therapy, including its specific implementation, structure, features, and effects. In the following description, different references to "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
[0017] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
[0018] The following describes in detail a method and system for real-time monitoring of bioelectric signals based on massage therapy provided by the present invention with reference to the accompanying drawings.
[0019] See also Figure 1 , which shows a flow chart of a method for real-time monitoring of bioelectric signals based on massage therapy provided by one embodiment of the present invention, the method comprising the following steps: Step S1, obtaining the electromyogram of the massage subject during the massage process.
[0020] In an embodiment of the present invention, the implementation scenario is to monitor the bioelectrical signals of the massaged subject during massage therapy to improve the accuracy of bioelectrical signal monitoring. First, the electromyogram (EMG) of the massaged subject is obtained during the massage. An electromyograph is used to collect the EMG signals in the massaged area and generate an EMG. The horizontal axis of the EMG is time, and the vertical axis is the EMG signal. In this embodiment of the present invention, the monitoring and acquisition frequency is 10Hz, and an EMG is generated every 10 seconds. The implementer can determine this frequency based on the implementation scenario.
[0021] Step S2: Decompose the electromyogram into time series to obtain periodic terms and residual terms; obtain the suspected abnormality degree according to the discrete characteristics of the data distribution in the residual term; and obtain the suspected abnormal time according to the suspected abnormality degree.
[0022] During massage, whether relieving muscle tension or activating weak muscles, the muscle's electromyographic activity gradually changes as the massage proceeds. For example, when relieving muscle tension or spasms, muscle electrical activity gradually decreases during the massage, indicating a significant relaxation effect. When activating weak and atrophied muscles, massage stimulation activates the target muscle group, increasing muscle excitability, and myoelectric activity gradually increases during the massage. Therefore, the subject's electromyogram (EMG) shows a clear trend during the massage. Furthermore, the massage movements and force applied during massage are cyclical, and thus, the EMG signal changes also exhibit a certain degree of cyclicity. Improper massage can cause reflex muscle tension, resulting in EMG signal fluctuations that do not conform to cyclicity or trend. Therefore, the EMG can be decomposed into a time series to obtain periodic terms and residual terms, which can be used to analyze abnormal EMG characteristics. It should be noted that the seasonal and trend decomposition method based on local weighted regression (STL) in the time series decomposition algorithm is currently available, and the specific decomposition steps are not detailed here. After removing the trend and cycle terms from the original time series, the residual term is obtained. Therefore, the residual term can reflect information that does not conform to trend and cycle characteristics. Based on the residual term, a preliminary analysis of moments when monitoring data anomalies may exist can be performed. When an abnormal EMG signal is present, the residual point at that moment has outlier characteristics and is significantly different from normal residual points. If there is no abnormal EMG signal in the EMG, the data distribution scale on both sides of the residual mean should be close. If an abnormal EMG signal is present, because the abnormal EMG signal is more outlier, the residual mean will be biased towards the abnormal EMG signal value, resulting in a difference in the data distribution scale on both sides of the residual mean. Therefore, the suspected anomaly degree can be obtained based on the discrete characteristics of the data distribution in the residual term.
[0023] Preferably, in an embodiment of the present invention, the step of obtaining the suspected abnormality includes: dividing the residual points in the residual term into positive residual points and negative residual points; since the residual points are positive and negative, it is necessary to distinguish the positive and negative values during analysis, so the positive residual points are positive and the negative residual points are negative. The minimum value of the ratio of the number of positive residual points on both sides of the average line of the positive residual points is calculated among the positive residual points to obtain a first quantity ratio; the average line of the positive residual points is the horizontal line where the average value of all positive residual points is located, and the minimum value of the ratio of the number of positive residual points is the side with fewer numbers as the numerator. When there is no electromyographic abnormality, the number of positive residual points on both sides of the average line should be close, then the first quantity ratio tends to 1; when there is electromyographic abnormality, the residual points caused by the abnormality are relatively discrete, which will pull up the average line, causing the average line to deviate towards the abnormal positive residual points, resulting in a large difference in the number on both sides. Therefore, the more obvious the degree of electromyographic abnormality or the higher the frequency, the more the first quantity ratio tends to 0. The minimum value of the ratio of the number of negative residual points on both sides of the average line of the negative residual points is calculated among the negative residual points to obtain the second number ratio; similarly, when there is no myoelectric abnormality, the number of negative residual points on both sides of the average line should be close, and the second number ratio tends to 1; when there is an electromyographic abnormality, the average line will be pulled down, causing the average line to be biased towards the abnormal negative residual point, resulting in a large difference in the number on both sides. Therefore, the more obvious the degree of electromyographic abnormality or the higher the frequency, the closer the second number ratio is to 0. Calculate the product of the first number ratio and the second number ratio and negatively correlate them to obtain the suspected abnormality degree; in this embodiment of the present invention, use Perform negative correlation mapping, represents an exponential function with a natural constant as the base, and a represents a mapping object. When the ratio of the first quantity to the second quantity is larger, the suspected abnormality is smaller, which means that the possibility of myoelectric abnormality is smaller.
[0024] Furthermore, when the suspected abnormality degree is greater, it is more likely that there is electromyographic abnormality in the electromyogram, so the suspected abnormality moment is obtained according to the suspected abnormality degree; preferably, in an embodiment of the present invention, the step of obtaining the suspected abnormality moment includes: when the suspected abnormality degree exceeds a preset first threshold value, the moment of all residual points above the average line of the positive residual points and below the average line of the negative residual points is taken as the suspected abnormality moment, the residual points at the suspected abnormal moment are relatively outliers, and the possibility of electromyographic abnormality is greater; in an embodiment of the present invention, the preset first threshold value is 0.25, which can be determined by the implementer according to the implementation scenario. When the suspected abnormality degree does not exceed the preset first threshold value, it means that there is no abnormal electromyographic signal in the electromyogram, and the electromyographic signal does not need to be corrected.
[0025] Step S3, in the residual term, obtain the neighborhood change characteristic value based on the residual distance characteristics between the suspected abnormal moment and the adjacent moments; obtain the period length segment of the periodic term based on the spectrum characteristics of the periodic term; obtain the period distribution characteristic value based on the time interval characteristics between the suspected abnormal moment in the periodic term and other suspected abnormal moments in other period length segments, and the number characteristics of other suspected abnormal moments corresponding to the suspected abnormal moment in each period length segment.
[0026] After obtaining the suspected abnormal moment, further analysis is needed to determine whether it represents a true EMG signal. Because the dispersion of EMG signals caused by external factors and those caused by massage is similar, a combined analysis of the characteristics of both scenarios is necessary to improve monitoring accuracy. Abnormal EMG activity caused by massage is similar to a muscle reflex and is a physiological activity characteristic. The EMG signal exhibits a sharp rise followed by a steady or gradual decline. The EMG signals immediately preceding and following the suspected abnormal moment exhibit a neighborhood variation characteristic, with a steep first rise and a gradual decline. However, external interference, such as motion artifacts between the electrode and the skin or impedance changes caused by sweating or oil secretion, do not exhibit physiological activity characteristics. EMG signals at the suspected abnormal moment exhibit sharper changes, making them more prominent than those caused by massage and significantly different from the preceding and following data. Therefore, the neighborhood variation characteristic value is derived from the residual distance between the suspected abnormal moment and its adjacent moments.
[0027] Preferably, in an embodiment of the present invention, the step of obtaining the neighborhood change characteristic value includes: calculating the Euclidean distance between the residual point of the suspected abnormal moment and the previous adjacent moment to obtain a first distance; the larger the first distance, the greater the difference in residual value between the suspected abnormal moment and the previous moment. Calculating the Euclidean distance between the residual point of the suspected abnormal moment and the subsequent adjacent moment to obtain a second distance; the larger the second distance, the greater the difference in residual value between the suspected abnormal moment and the subsequent moment. Calculating the inverse of the sum of the first distance and the second distance to obtain the degree of stability; when the suspected abnormal moment is caused by external interference, the larger the first distance and the second distance, the more the residual value at this moment presents a mutation characteristic and the lower the degree of stability. Calculating the product of the absolute value of the difference between the first distance and the second distance and the degree of stability to obtain the neighborhood change characteristic value of the suspected abnormal moment. When the suspected abnormal moment is caused by massage factors, the residual value will show a characteristic of steep rise and slow fall, the first distance is significantly greater than the second distance, and the absolute value of the difference between the first distance and the second distance is larger; on the contrary, when the suspected abnormal moment is caused by external interference, both the first distance and the second distance are larger, and the absolute value of the difference between the first distance and the second distance is smaller. Therefore, the larger the neighborhood change characteristic value, the more likely the suspected abnormal moment is caused by massage; when the neighborhood change characteristic value is smaller, the more likely the suspected abnormal moment is caused by external interference. The formula for obtaining the neighborhood change characteristic value includes:
[0028] Where F represents the neighborhood change characteristic value at the suspected abnormal moment, represents the first distance, represents the second distance, Indicates the degree of stability, Indicates the absolute value of the difference between the first distance and the second distance.
[0029] Furthermore, since massage movements and force generation exhibit periodicity, if the suspected abnormal moment is similar in different time periods, it means that the suspected abnormal moment represents a true abnormality in the electromyographic signal, while the distribution of suspected abnormal moments caused by external interference is relatively random. First, the periodic term is divided into different periods, and then the period length segment of the periodic term is obtained based on the spectral characteristics of the periodic term; specifically, the periodic term is subjected to Fourier transform to obtain spectral data; it should be noted that Fourier transform belongs to the existing technology, and the specific steps are not repeated here. The inverse of the frequency corresponding to the maximum value in the spectral data is used as the segmentation length. The frequency corresponding to the maximum value in the spectral data reflects the main periodic characteristics, and the segmentation length is the main period length. Then, the periodic term is divided according to the segmentation length to obtain different period length segments, each of which represents a period of change in the electromyographic signal. After obtaining different period length segments, the periodic distribution characteristic value can be obtained based on the time interval characteristics between the suspected abnormal moment in the periodic term and other suspected abnormal moments in other period length segments, and the number characteristics of other suspected abnormal moments corresponding to the suspected abnormal moment in each period length segment.
[0030] Preferably, in an embodiment of the present invention, the step of obtaining a periodic distribution characteristic value includes: taking the moment of the suspected abnormal moment in the period length segment as the target moment, calculating the proportion of the number of other suspected abnormal moments that appear in all target moments, and obtaining a first similar characteristic value; when the first similar characteristic value is larger, it means that there are more other suspected abnormal moments in the target moments of other period length segments, the more periodic distribution the suspected abnormal moment presents, and the more likely it is that massage caused the suspected abnormal moment to appear. Since the suspected abnormal moments of different periods are not exactly at the same moment, the average value of the phase difference between the other suspected abnormal moments closest to the target moment in all period length segments and the target moment is calculated to obtain the average phase difference; when the average phase difference is smaller, it means that some other suspected abnormal moments in other period length segments are closer to the target moment, the more periodic distribution they present, and the more likely it is that massage caused the suspected abnormal moment. It should be noted that when other suspected abnormal moments in other period length segments are at the target moment, the phase difference between the two suspected abnormal moments is limited to 1, which can avoid the situation where the subsequent denominator appears to be 0 or less than 1. Calculate the inverse of the average phase difference and the average of the first similarity eigenvalue to obtain the periodic distribution eigenvalue of the suspected abnormal moment; when the average phase difference corresponding to the suspected abnormal moment is smaller and the first similarity eigenvalue is larger, the periodic distribution eigenvalue is larger, and the suspected abnormal moment is more likely to be caused by massage factors; when the periodic distribution eigenvalue is smaller, the suspected abnormal moment is more likely to be caused by external interference. The formula for obtaining the periodic distribution eigenvalue includes:
[0031] Where R represents the characteristic value of the period distribution, E represents the number of other suspected abnormal moments in all target moments, and N represents the number of period length segments. represents the first similarity eigenvalue, It represents the phase difference between the target time and other suspected abnormal time closest to the target time in the nth cycle length segment. represents the average phase difference.
[0032] Step S4, obtaining the abnormality degree of the myoelectric activity at the suspected abnormal moment according to the neighborhood change characteristic value and the period distribution characteristic value; and correcting the electromyogram according to the abnormality degree of the myoelectric activity.
[0033] Both the neighborhood change characteristic value and the periodic distribution characteristic value can characterize the factors that cause the suspected abnormal moment, so the degree of abnormal myoelectric activity at the suspected abnormal moment is obtained based on the neighborhood change characteristic value and the periodic distribution characteristic value; preferably, in an embodiment of the present invention, the step of obtaining the degree of abnormal myoelectric activity includes: calculating the average value of the neighborhood change characteristic value and the periodic distribution characteristic value to obtain the degree of abnormal myoelectric activity at the suspected abnormal moment; the greater the degree of abnormal myoelectric activity, the more likely it is that the suspected abnormal moment is a true myoelectric activity feature caused by massage; the greater the degree of abnormal myoelectric activity, the more likely it is that the suspected abnormal moment is caused by interference from external factors, and is not a true myoelectric activity feature.
[0034] Furthermore, after obtaining the abnormality degree of electromyographic activity at the suspected abnormal moment in the electromyogram, the electromyogram can be corrected according to the abnormality degree of electromyographic activity; specifically, the suspected abnormal moment when the electromyographic activity abnormality exceeds the preset second threshold is used as the moment to be corrected; in the embodiment of the present invention, the preset second threshold is 0.5, which can be determined by the implementer according to the implementation scenario; the moment to be corrected means that the electromyographic signal at that moment is not a true value. In order to improve the reference reliability of the electromyogram for the massage doctor, the electromyographic signal value at the moment to be corrected needs to be replaced. Then, the average value of the adjacent moments before and after the moment to be corrected is calculated in the electromyogram to obtain a corrected value; the corrected value is replaced with the value of the moment to be corrected; after the correction value is replaced, the influence of external factors on the electromyogram can be reduced, thereby improving the monitoring accuracy of the bioelectric signal and the accuracy of the massage status feedback of the monitoring results.
[0035] In summary, the embodiment of the present invention provides a method for real-time monitoring of bioelectric signals based on massage therapy; the electromyogram is decomposed into time series to obtain periodic terms and residual terms; the suspected abnormal moment is obtained according to the discrete characteristics of the data distribution in the residual term; the neighborhood change characteristic value is obtained according to the residual distance characteristics between the suspected abnormal moment and the adjacent moment; the cycle length segment is obtained according to the spectrum characteristics of the periodic term; the period distribution characteristic value is obtained according to the time interval characteristics between the suspected abnormal moment in the periodic term and other suspected abnormal moments in other period length segments, and the number characteristics of other suspected abnormal moments corresponding to the suspected abnormal moment in each period length segment. The present invention obtains the abnormality of electromyographic activity at the suspected abnormal moment according to the neighborhood change characteristic value and the period distribution characteristic value and corrects the electromyogram, thereby improving the monitoring accuracy of the bioelectric signal and the accuracy of the massage status feedback of the monitoring results.
[0036] The present invention also proposes a real-time monitoring system for bioelectric signals based on massage therapy, comprising a memory, a processor, and a computer program stored in the memory and runnable on the processor. The processor executes the computer program to implement any one of the steps of a real-time monitoring method for bioelectric signals based on massage therapy.
[0037] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0038] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.
Claims
1. A real-time monitoring method for bioelectric signals based on massage therapy, characterized in that: The method comprises the following steps: Obtaining electromyogram of the massage subject during the massage process; Decomposing the electromyogram in time series to obtain a period term and a residual term; obtaining a suspected abnormality degree based on a discrete feature of data distribution in the residual term; and obtaining a suspected abnormality moment based on the suspected abnormality degree; Obtaining a neighborhood change characteristic value in the residual term based on the residual distance feature between the suspected abnormal moment and the adjacent moments; obtaining a period length segment of the period term based on the spectrum feature of the period term; obtaining a period distribution characteristic value based on the time interval feature between the suspected abnormal moment in the period term and other suspected abnormal moments in other period length segments, and the quantity feature of other suspected abnormal moments corresponding to the suspected abnormal moment in each period length segment; The abnormality degree of the electromyographic activity at the suspected abnormal moment is obtained according to the neighborhood change characteristic value and the period distribution characteristic value; and the electromyogram is corrected according to the abnormality degree of the electromyographic activity.
2. The method for real-time monitoring of bioelectric signals based on massage therapy according to claim 1, characterized in that: The step of obtaining the suspected abnormality degree according to the discrete characteristics of the data distribution in the residual term includes: The residual points in the residual term are divided into positive residual points and negative residual points; among the positive residual points, the minimum value of the ratio of the number of positive residual points on both sides of the average line of the positive residual points is calculated to obtain a first quantity ratio; among the negative residual points, the minimum value of the ratio of the number of negative residual points on both sides of the average line of the negative residual points is calculated to obtain a second quantity ratio; the product of the first quantity ratio and the second quantity ratio is calculated and negatively correlated to obtain the suspected abnormality degree.
3. The method for real-time monitoring of bioelectric signals based on massage therapy according to claim 2, characterized in that: The step of obtaining the suspected abnormality moment according to the suspected abnormality degree includes: When the suspected abnormality degree exceeds a preset first threshold, the moments of all residual points above the average line of the positive residual points and below the average line of the negative residual points are taken as suspected abnormality moments.
4. The method for real-time monitoring of bioelectric signals based on massage therapy according to claim 1, characterized in that: The step of obtaining a neighborhood change characteristic value according to the residual distance feature between the suspected abnormal moment and the adjacent moment in the residual term includes: Calculate the Euclidean distance between the residual point of the suspected abnormal moment and the previous adjacent moment to obtain a first distance; calculate the Euclidean distance between the residual point of the suspected abnormal moment and the subsequent adjacent moment to obtain a second distance; calculate the inverse of the sum of the first distance and the second distance to obtain a degree of stability; calculate the product of the absolute value of the difference between the first distance and the second distance and the degree of stability to obtain a neighborhood change characteristic value of the suspected abnormal moment.
5. The method for real-time monitoring of bioelectric signals based on massage therapy according to claim 1, characterized in that: The step of obtaining the period length segment of the periodic term according to the spectrum characteristics of the periodic term comprises: The periodic term is subjected to Fourier transformation to obtain spectrum data; the inverse of the frequency corresponding to the maximum value in the spectrum data is used as a segmentation length, and the periodic term is divided according to the segmentation length to obtain different periodic length segments.
6. The method for real-time monitoring of bioelectric signals based on massage therapy according to claim 1, characterized in that: The step of obtaining a periodic distribution characteristic value according to the time interval characteristics between the suspected abnormal moment in the period item and other suspected abnormal moments in other period length segments, and the quantity characteristics of other suspected abnormal moments corresponding to the suspected abnormal moment in each period length segment, comprises: Take the moment of the suspected abnormal moment in the period length segment as the target moment, calculate the proportion of other suspected abnormal moments in all target moments, and obtain the first similarity characteristic value; calculate the average value of the phase difference between other suspected abnormal moments closest to the target moment in all period length segments and the target moment, and obtain the average phase difference; calculate the inverse of the average phase difference and the average value of the first similarity characteristic value to obtain the periodic distribution characteristic value of the suspected abnormal moment.
7. The method for real-time monitoring of bioelectric signals based on massage therapy according to claim 1, characterized in that: The step of obtaining the abnormality degree of the myoelectric activity at the suspected abnormal moment according to the neighborhood change characteristic value and the period distribution characteristic value comprises: The average value of the neighborhood change characteristic value and the period distribution characteristic value is calculated to obtain the abnormality degree of the electromyographic activity at the suspected abnormal moment.
8. The method for real-time monitoring of bioelectric signals based on massage therapy according to claim 1, characterized in that: The step of correcting the electromyogram according to the abnormality of the electromyographic activity comprises: The suspected abnormal moment when the electromyographic activity abnormality exceeds a preset second threshold is used as the moment to be corrected; the average value of the adjacent moments before and after the moment to be corrected is calculated in the electromyogram to obtain a correction value; and the correction value replaces the value of the moment to be corrected.
9. A real-time monitoring system for bioelectric signals based on massage therapy, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 8.