Pelvic floor muscle early screening training method based on data analysis

Abstract

The invention discloses a pelvic floor muscle early screening training method based on data analysis, and the method comprises the steps: a user completes a specified rating action, and obtains SEMG (Surface Electromyography) data; the method comprises the following steps: processing original data of a surface electromyographic signal, calculating an evaluation index of the electromyographic signal, obtaining index information in a time domain aspect, carrying out fast Fourier transform on the signal to obtain frequency domain related information, and calculating the evaluation index of the electromyographic signal according to the frequency domain related information. The related information comprises a median frequency MF and an average power frequency MPF; comparing and screening the obtained signal index with an expected index; according to comparison with a preset value, a pelvic floor muscle state report is produced according to a rating result; the pelvic floor muscle function evaluation method has the technical advantages that subjectivity and uncertainty possibly existing in a traditional evaluation method can be avoided, and the real function state of the pelvic floor muscle can be comprehensively reflected.

Description

Technical Field

[0001] The present invention relates to a method for early screening and training of pelvic floor muscles. More specifically, it relates to a method for early screening and training of pelvic floor muscles based on data analysis, belonging to the field of medical systems. Background Art

[0002] The pelvic floor muscles, as an important supporting structure for pelvic organs, are like an invisible "elastic net", firmly holding up organs such as the bladder, uterus, and rectum, maintaining their normal positions and physiological functions. However, under the influence of many factors in modern life, pelvic floor muscle dysfunction problems are becoming increasingly common. Postpartum women experience the "ordeal" of pregnancy and childbirth, and their pelvic floor muscles suffer great stretching and damage; as the elderly age, the muscle elasticity decreases and the fascia relaxes, and the function of the pelvic floor muscles gradually degenerates; people who sit for a long time, are obese, or engage in high-intensity physical labor have their pelvic floor muscles under long-term pressure and over-fatigue, and also face the risk of pelvic floor muscle function impairment.

[0003] Early and accurate screening of the functional status of pelvic floor muscles, combined with scientific and effective training methods, is a crucial measure for preventing and treating pelvic floor muscle dysfunction and improving the quality of life of patients. A pelvic floor muscle early screening system and training method based on data analysis have emerged. It integrates advanced sensing technology, complex data processing algorithms, and professional medical knowledge, opening up a new path for pelvic floor muscle health management.

[0004] Currently, there are mainly two ways for the main pelvic floor muscle assessment programs: one is the traditional way of relying solely on doctors' manual observation, feeling the movement of the pelvic floor muscles with fingers, and rating the muscle strength according to the contraction state of the muscles. The other is to adopt the Glazer protocol. Patients need to complete specified actions according to the doctor's instructions, and analyze and rate according to the electromyographic signals taken by the pelvic floor muscles during the movement process. For example, patent application publication number CN117275667A discloses a system and method for optimizing the fatigue degree of pelvic floor muscles based on big data, including a reading module, a data recognition module, a data calculation module, a decision-making processing module, a plan adjustment module, and a log recording module. This invention reads the plan data, obtains the template data in the contraction stage according to the component information, judges the abdominal muscle participation detection function, collects the electromyographic signals of the patient's pelvic floor muscles and abdominal muscles through a vaginal electrode and processes them to obtain the corresponding pelvic floor muscle data and abdominal muscle data. According to the template data in the contraction stage, the pelvic floor muscle data, and the abdominal muscle data, it judges whether the pelvic floor muscles are in a fatigued state. According to the judgment of the pelvic floor muscle fatigue state, it processes the subsequent training plan data to make the patient's training plan data reach the template value, and finally records and saves the data flow situation in the whole process one by one. However, the assessment of pelvic floor muscles often depends not only on the fatigue assessment index, but also requires a more comprehensive assessment. Summary of the Invention

[0005] To solve the above-mentioned problems in the prior art, the present invention provides a method for early screening and training of pelvic floor muscles based on data analysis, which has the technical features of being able to avoid the subjectivity and uncertainty that may exist in traditional evaluation methods and being able to comprehensively reflect the true functional state of the pelvic floor muscles.

[0006] To achieve the above object, the present invention is realized through the following technical solutions:

[0007] A method for early screening and training of pelvic floor muscles based on data analysis according to the present invention, the method comprising the following steps:

[0008] Step a: The user uses a powered non-invasive pelvic floor muscle detection device to complete a specified rating action, and obtains the real-time value of the surface electromyogram (SEMG) of the pelvic floor muscles during the action.

[0009] Step b: First, process the original data of the surface electromyogram signal, calculate the evaluation indexes of the electromyogram signal, obtain the index information in the time domain, and the index information includes the mean absolute value (MAV), root mean square (RMS), maximum peak time (t1), and integrated electromyogram value (iEMG). Secondly, perform a fast Fourier transform on the signal to obtain the frequency domain related information, and the related information includes the median frequency (MF) and the mean power frequency (MPF).

[0010] Step c: Compare the obtained signal indexes with the expected indexes for multi-faceted screening, and the multi-faceted screening includes muscle strength, muscle endurance, reaction speed, coordination, and fatigue.

[0011] Step d: Generate a pelvic floor muscle status report according to the comparison with the preset value; if an abnormality occurs, provide targeted rehabilitation training guidance based on the comparison between the signal indexes and the expected indexes.

[0012] Step e: Upload the data to the cloud to generate a data set, which is convenient for the user to more accurately realize the early screening and training of pelvic floor muscles based on data analysis.

[0013] Preferably, the non-invasive pelvic floor muscle detection device includes a signal acquisition module, a signal processing module, a signal detection and judgment module, a report generation module, and a data storage module that are communicatively connected. The signal processing module includes a signal amplifier. The signal acquisition module includes disposable gel electrodes. The signal amplifier is connected to 9 disposable gel electrodes, and the 9 electrodes are divided into 3 groups, and each group includes a positive electrode, a negative electrode, and a reference electrode.

[0014] Preferably, the specified rating actions in step a include rapid contraction actions, slow continuous contraction actions, and intermittent contraction actions; the user collects signals under the guidance of professional medical staff when using the original equipment to complete the specified actions.

[0015] Among them, the quick contraction action requires the user to instantly tighten the pelvic floor muscles to the maximum contraction, lasting for about 3 - 5 seconds, just like simulating the emergency response of the pelvic floor muscles in the scenarios of sudden coughing and sneezing in daily life, so as to capture the high-intensity myoelectric signals corresponding to the instantaneous explosive power of the muscles and accurately evaluate the nerve conduction speed and rapid activation ability of the pelvic floor muscles;

[0016] For the slow and continuous contraction action, the user needs to exert force slowly and gradually reach the maximum contraction degree after 5 - 10 seconds, and then maintain it for 10 - 15 seconds, analogous to the working state of the pelvic floor muscles under continuous pressure during long-term standing and walking with load, which is used to consider the muscle endurance and the ability to maintain stable tension;

[0017] The intermittent contraction action is set to a cyclic rhythm of "contract for 3 seconds and then relax for 2 seconds", repeating 8 - 10 times, accurately mapping the dynamic cooperation situation of the frequent alternating contraction and relaxation of the pelvic floor muscles during daily walking and climbing stairs, and testing its coordination and flexibility.

[0018] Preferably, obtaining the original surface electromyogram SEMG data in step a includes:

[0019] First, perform 50HZ notch filtering on it to eliminate power frequency interference; then perform band-pass filtering on it through an infinite impulse response system to complete the processing of the original signal; finally, extract the eigenvalue of the user's surface pelvic floor muscle surface electromyogram SEMG signal.

[0020] Preferably, in the time domain, assuming that the original SEMG signal is processed to obtain {x{i}|(i = 1, 2, 3...N)}, where N represents the number of sample points, the feature extraction calculation formula for this one-dimensional signal is as follows:

[0021] 1) The mean absolute value MAV in the index information is:

[0022] In this formula, N is the number of surface electromyogram signal data points collected, and x(i) is the i-th data in the signal data sequence, that is, first find the absolute value of the amplitude of the electromyogram signal point by point, sum up these absolute values, and then divide by the total number of sample points, so as to obtain the average level of the electromyogram signal amplitude in this period;

[0023] 2) The root mean square RMS in the index information is:

[0024] During the operation, first square the amplitude of each sampling point in the electromyogram signal sequence, sum up all the squared values and take the average, and then take the square root to obtain the RMS value;

[0025] 3) The integrated electromyogram value iEMG in the index information is: That is, directly sum up the absolute values of the electromyogram signal amplitudes;

[0026] 4) Maximum peak time t1: Precisely lock the moment when the amplitude of the EMG signal first reaches the maximum value in the time-domain waveform, which is the maximum peak time t1; it measures the time taken by the pelvic floor muscles from the moment of receiving the contraction instruction to the full burst and reaching the maximum contraction force in terms of time dimension.

[0027] Preferably, perform a fast Fourier transform on the original SEMG signal in the frequency domain to obtain the frequency components and amplitude information of the signal. Let the power spectral density of the user's surface EMG signal be PSD(f), and use two indicators, the median frequency MF and the mean power frequency MPF in the relevant information, to characterize the characteristics of the user's SEMG spectrum or power spectrum:

[0028] 1) Median frequency MF: The median frequency refers to the frequency that divides the power spectral density function PSD of the EMG signal into two equal-area parts;

[0029] 2) Mean power frequency MPF, which represents the average distribution position of the EMG signal power on the frequency axis.

[0030] Preferably, the multi-faceted screening in step c is specifically as follows:

[0031] 1) Muscle strength screening: Obtain the data of the pelvic floor muscle contraction strength, which is reflected by obtaining three indicators, the mean absolute value MAV, the root mean square RMS, and the integrated EMG value iEMG of the user during the completion of the specified process;

[0032] The mean absolute value MAV reflects the average contraction force, the root mean square RMS highlights the energy release intensity, and the integrated EMG value iEMG presents the total amount of electrical activity; and compare with the normal muscle strength range generated by the large-sample statistics of the same age and gender population. If the measured mean absolute value MAV and root mean square RMS of the patient are lower than the lower limit, and the integrated EMG value iEMG is small, it indicates that the pelvic floor muscle contraction strength is seriously insufficient;

[0033] 2) Muscle endurance screening: Muscle endurance considers the continuous working ability of the pelvic floor muscles and is evaluated by the indicators of the median frequency MF and the mean power frequency MPF generated during the user's exercise; if the median frequency MF decreases slowly and the mean power frequency MPF is stable, it is considered that the healthy pelvic floor muscles contract for a long time; if MF drops steeply and MPF deviates significantly, it is considered that the endurance is poor; for example, long-distance drivers sit for a long time, and the pelvic floor muscles are chronically fatigued, MF continues to decline, and MPF drifts to low frequencies;

[0034] 3) Reaction speed screening: Reaction speed is related to the emergency response ability of the pelvic floor muscles, and is based on the maximum peak time t1; according to a large sample of healthy people, appropriate reaction duration thresholds are set for different movement patterns; when the abdominal pressure rises instantaneously during coughing and sneezing, the pelvic floor muscles should contract rapidly, usually within dozens of milliseconds; a normal threshold is set. If it far exceeds the threshold, such as reaching hundreds of milliseconds, it indicates abnormal nerve conduction and muscle activation.

[0035] 4) Coordination screening: The coordination of the pelvic floor muscles is judged by observing parameters, including the waveform, frequency and amplitude of the electromyogram signal;

[0036] Under normal circumstances, when the pelvic floor muscles contract synergistically, the start time, peak time and end time of the electromyogram signals recorded in each channel corresponding to different electrode positions should be relatively synchronous. If the signal of a certain channel is significantly delayed or advanced, it indicates poor coordination between the pelvic floor muscles at this site and other sites.

[0037] 5) Fatigue screening: Comprehensive time-domain and frequency-domain full-course index changes are used to achieve fatigue screening; in the time domain, the mean absolute value MAV, root mean square RMS, and decay slope of the integrated electromyogram value iEMG are observed, and in the frequency domain, the median frequency MF, drift direction and speed of the mean power frequency MPF are observed; if the index attenuates linearly and moderately, it is normal progressive fatigue of the pelvic floor muscles. If the index drops exponentially linearly, it indicates abnormal fatigue.

[0038] Preferably, step d specifically includes: comprehensively integrating the detailed results of multi-dimensional screening to closely integrate the user's key information, including age, BMI, and mental health status. The system generates a pelvic floor muscle status report, which includes the user's basic information, the measured values of each index, the details of the comparison with the expected index, and the health rating, and is accompanied by intuitive charts and professional interpretations, making it clear at a glance for the user; the health rating is subdivided into multiple levels, and corresponds to the functional levels of the pelvic floor muscles accurately from "excellent" to "severely abnormal";

[0039] If it is screened and determined that there is an abnormality in the pelvic floor muscle status, the system will implement a customized recovery training guidance plan based on the built-in intelligent algorithm and the rehabilitation expert knowledge base. The training plan fully takes into account the dual needs of physical and mental rehabilitation, carefully designs an exclusive set of movements, covering diverse items such as modified Kegel exercises and pelvic floor muscle yoga; reasonably plans the training frequency (recommended 3-5 times a week) and duration (20-30 minutes each time); recommends suitable training aids (such as biofeedback devices, yoga bricks, etc.); is equipped with a real-time feedback mechanism, uses portable devices to monitor the electromyogram signal in real time, and prompts the accuracy of the movement in real time through voice and images, escorting the training throughout the process. At the same time, it highly concerns the mental health of users. For patients with tense pelvic floor muscles caused by high stress and anxiety, relaxation training and psychological counseling are skillfully incorporated, adhering to the concept of treating the body and mind simultaneously to promote recovery.

[0040] Step e: Along with the widespread application of the system and the continuous influx of massive user data, the system uploads the data to the cloud, enabling more intelligent optimization of the training guidance strategy, realizing personalized and precise customization of the training plan, helping users further understand the state of the pelvic floor muscles, and meeting the unique rehabilitation needs of each user.

[0041] Preferably, the rehabilitation training guidance described in step d includes Kegel exercises and pelvic floor muscle yoga;

[0042] The non-invasive pelvic floor muscle detection device further includes a voice module, a graphic module, a dynamic capture module, and an infrared induction module that are communicatively connected to the signal processing module. Through the combination of the voice module, graphic module, dynamic capture module, and infrared induction module in the system, real-time feedback on the accuracy of the actions for rehabilitation training guidance is achieved through voice and image prompts.

[0043] Beneficial effects: Conduct a preliminary pelvic floor muscle screening for users, enabling users to generally understand the state of their own pelvic floor muscles; analyze the pelvic floor muscle electromyogram signals from the time domain and frequency domain, evaluate the pelvic floor muscles from multiple angles, making the evaluation results more comprehensively reflect the true functional state of the pelvic floor muscles. If there are abnormalities in the pelvic floor muscle state, based on the evaluation results, provide users with more professional evaluation guidance, and upload user data to the cloud, enabling more intelligent optimization of the training guidance strategy. Brief Description of the Drawings

[0044] Figure 1 It is a schematic flowchart of the present invention. Detailed Embodiments

[0045] The following further describes the present invention in conjunction with the drawings of the specification, but the present invention is not limited to the following embodiments.

[0046] Prior Art: In traditional pelvic floor muscle palpation examinations, doctors feel the contraction of the pelvic floor muscles by inserting their fingers into the vagina or rectum. The results of this examination rely to a large extent on the doctor's experience and subjective judgment. There may be differences in the judgment criteria of different doctors for muscle contraction strength, duration, and symmetry. For example, regarding the assessment of muscle contraction strength, some doctors may consider a certain degree of contraction normal, while another doctor may feel that the contraction strength is insufficient, which leads to a lack of unified standards for objectivity and accuracy in the examination results. Secondly, the traditional method requires doctors to perform palpation examinations, which not only causes discomfort to patients but also poses an infection risk. For the Glazer protocol, it mainly assesses pelvic floor muscle function based on electromyographic signals. It only provides information about the muscle strength, endurance, etc. of the pelvic floor muscles from the time domain, without analyzing the state of the pelvic floor muscles in combination with the frequency domain, and the reference values are too one-sided. Secondly, the Glazer protocol cannot accurately detect the coordination and fatigue degree of the pelvic floor muscles, and the Glazer protocol has no big data support and cannot combine the user's own state and a large amount of user data to reflect a more accurate state of the pelvic floor muscles.

[0047] Principle / Technical Solution of the Present Invention: By using multiple electromyographic signal indicators in the time domain (MAV, RMS, maximum peak duration t1) and frequency domain (MF, MPF), a multi-dimensional pelvic floor muscle function evaluation system is formed. The comprehensive evaluation can more comprehensively and accurately reflect the state of multiple key aspects such as the muscle strength, muscle endurance, reaction speed, coordination, and fatigue of the pelvic floor muscles. According to the multi-dimensional screening results, combined with information such as the user's age, BMI, and mental health status, a pelvic floor muscle status report is generated. The report content is rich, including basic information, measured values, comparison results, and health ratings, enabling users to clearly understand the health status of their own pelvic floor muscles.

[0048] As Figure 1 shown is a specific embodiment of a pelvic floor muscle early screening and training method based on data analysis. This embodiment is a pelvic floor muscle early screening and training method based on data analysis, and this method includes the following steps:

[0049] Step a: The user uses the powered non-invasive pelvic floor muscle detection device to complete the specified rating actions and obtain the real-time values of the surface electromyogram (SEMG) of the pelvic floor muscles during the actions. The non-invasive pelvic floor muscle detection device includes a signal acquisition module, a signal processing module, a signal detection and judgment module, a report generation module, and a data storage module that are communicatively connected. The non-invasive pelvic floor muscle detection device also includes a voice module, a graphics module, a dynamic capture module, and an infrared sensing module that are communicatively connected to the signal processing module. Through the combination of the voice module, the graphics module, the dynamic capture module, and the infrared sensing module in the system, real-time voice and image prompts are realized to provide feedback on the accuracy of the actions for rehabilitation training guidance. The signal processing module includes a signal amplifier, and the signal acquisition module includes disposable gel electrodes. The signal amplifier is connected to 9 disposable gel electrodes, and the 9 electrodes are divided into 3 groups, with each group containing a positive electrode, a negative electrode, and a reference electrode. Such a layout can capture more comprehensive multi-dimensional activity information of the pelvic floor muscle group, providing a rich data basis for subsequent signal processing. The use of disposable gel electrodes in close contact with the user's skin ensures high-quality signal acquisition. These electrodes not only ensure the clarity of the signals but also, through their disposable design, effectively avoid the risk of cross-infection and improve the hygiene of use;

[0050] The specified rating actions include rapid contraction actions, slow and sustained contraction actions, and intermittent contraction actions. When the user completes the specified actions using the original device under the guidance of professional medical staff, signal acquisition is carried out. Among them, for the rapid contraction action, the user is required to instantaneously tighten the pelvic floor muscles to the maximum tightness and maintain it for about 3 - 5 seconds, just like simulating the emergency response of the pelvic floor muscles in daily life during sudden coughing or sneezing, thereby capturing the high-intensity myoelectric signals corresponding to the instantaneous explosive power of the muscles to accurately evaluate the nerve conduction velocity and rapid activation ability of the pelvic floor muscles. For the slow and sustained contraction action, the user needs to exert force slowly, gradually reach the maximum contraction degree in 5 - 10 seconds, and then maintain it for 10 - 15 seconds, analogous to the working state of the pelvic floor muscles under continuous pressure during long-term standing or walking with a load, to assess the muscle endurance and the ability to maintain stable tension. The intermittent contraction action is set to a cyclic rhythm of "contract for 3 seconds and then relax for 2 seconds", repeating 8 - 10 times, accurately mapping the dynamic cooperation situation of the pelvic floor muscles during frequent alternating contraction and relaxation during daily walking and climbing stairs, and testing its coordination and flexibility;

[0051] Step b: First, process the original data of the surface electromyogram signal, calculate the evaluation indexes of the electromyogram signal, obtain index information in the time domain, and the index information includes the mean absolute value MAV (abbreviation: MAV), root mean square RMS (abbreviation: RMS), maximum peak time t1 (abbreviation: t1), and integrated electromyogram value iEMG (abbreviation: iEMG). Secondly, perform a fast Fourier transform on the signal to obtain frequency-domain related information, and the related information includes the median frequency MF (abbreviation: MF) and mean power frequency MPF (abbreviation: MPF).

[0052] Step c: Compare the obtained signal indexes with the expected indexes and conduct multi-faceted screening. The multi-faceted screening includes muscle strength, muscle endurance, reaction speed, coordination, and fatigue.

[0053] Among them, 1) Muscle strength screening: Obtain data on the contraction strength of the pelvic floor muscles, which is reflected by obtaining the three indexes of the mean absolute value MAV, root mean square RMS, and integrated electromyogram value iEMG of the user during the completion of the specified process.

[0054] The mean absolute value MAV reflects the average contraction force, the root mean square RMS highlights the intensity of energy release, and the integrated electromyogram value iEMG presents the total amount of electrical activity. Compare with the normal muscle strength range generated by large-sample statistics of the same age and gender population. If the measured mean absolute value MAV and root mean square RMS of the patient are lower than the lower limit, and the integrated electromyogram value iEMG is on the small side, it indicates that the contraction strength of the pelvic floor muscles is seriously insufficient.

[0055] 2) Muscle endurance screening: Muscle endurance considers the continuous working ability of the pelvic floor muscles and is evaluated by the indexes of median frequency MF and mean power frequency MPF generated during the user's movement. If the median frequency MF drops slowly and the mean power frequency MPF is stable, it is considered that the healthy pelvic floor muscles contract for a long time. If MF drops steeply and MPF deviates greatly, it is considered that the endurance is poor. For example, long-distance truck drivers sit for a long time, and the pelvic floor muscles are chronically fatigued, MF continues to decline, and MPF drifts to low frequencies.

[0056] 3) Reaction speed screening: Reaction speed is related to the emergency ability of the pelvic floor muscles and is based on the maximum peak time t1. According to a large number of healthy population samples, set appropriate reaction duration thresholds for different movement modes. When the abdominal pressure rises instantaneously during coughing and sneezing, the pelvic floor muscles should contract quickly, usually within a few tens of milliseconds. Set a normal threshold. If it far exceeds the threshold, such as reaching several hundred milliseconds, it indicates abnormal nerve conduction and muscle activation.

[0057] 4) Coordination screening: Judge the coordination of the pelvic floor muscles by observing parameters, and the parameters include the waveform, frequency, and amplitude of the electromyogram signal.

[0058] Under normal circumstances, when the pelvic floor muscles contract synergistically, the start time, peak time, and end time of the electromyographic signals recorded in each channel corresponding to different electrode positions should be relatively synchronized. If the signal of a certain channel is significantly delayed or advanced, it indicates poor coordination between the pelvic floor muscles at that site and other sites.

[0059] 5) Fatigue screening: Comprehensive changes in time-domain and frequency-domain full-course indicators are used to achieve fatigue screening; in the time domain, the mean absolute value MAV, root mean square RMS, and attenuation slope of the integrated electromyogram value iEMG are observed, and in the frequency domain, the median frequency MF, drift direction and speed of the mean power frequency MPF are observed; if the indicators attenuate linearly and moderately, it is normal progressive fatigue of the pelvic floor muscles, and if the indicators drop exponentially linearly, it indicates abnormal fatigue.

[0060] Step d: Generate a pelvic floor muscle status report based on the comparison with the preset value through the rating result; if an abnormality occurs, targeted rehabilitation training guidance is carried out based on the comparison between the signal indicators and the expected indicators.

[0061] Specifically: Integrate the detailed results of multi-dimensional screening to closely integrate the user's key information, which includes age, BMI, and mental health status. The system generates a pelvic floor muscle status report, which includes the user's basic information, measured values of each indicator, details of the comparison with the expected indicators, and health ratings, and is accompanied by intuitive charts and professional interpretations to make it clear at a glance for the user; the health ratings are subdivided into multiple levels, and from "excellent" to "severely abnormal" to accurately anchor the functional level of the pelvic floor muscles.

[0062] If the screening determines that the pelvic floor muscle status is abnormal, the system can adopt built-in intelligent algorithms and a rehabilitation expert knowledge base to implement a customized rehabilitation training guidance plan. The specific algorithms and expert data are stored according to needs. The specific training plan can fully consider the dual needs of physical and mental rehabilitation, such as covering multiple items such as modified Kegel exercises and pelvic floor muscle yoga. A specific recommendation: Reasonably plan the training frequency (it is recommended to do it 3-5 times a week) and duration (20-30 minutes each time); recommend suitable training aids (such as biofeedback devices, yoga bricks, etc.); carry a real-time feedback mechanism, use portable devices to monitor electromyographic signals in real time, and prompt the accuracy of movements in real time through voice and images to escort the training throughout the process. At the same time, pay high attention to the user's mental health. For patients with tense pelvic floor muscles caused by high stress and anxiety, skillfully integrate relaxation training and psychological counseling sessions, and adhere to the concept of treating the body and mind together to promote rehabilitation.

[0063] Step e: Upload the data to the cloud to generate a dataset, which is convenient for users to more accurately achieve early screening training of the pelvic floor muscles based on data analysis. By uploading the data to the cloud, the training guidance strategy can be more intelligently optimized, and the training plan can be customized personalized and precisely, helping users further understand the pelvic floor muscle status and meet the unique rehabilitation demands of each user.

[0064] In a preferred embodiment, obtaining the original surface electromyogram (SEMG) data in step a includes:

[0065] First, perform 50 Hz notch filtering on it to eliminate power frequency interference; then perform band-pass filtering on it through an infinite impulse response system to complete the processing of the original signal; finally, extract the eigenvalue of the surface pelvic floor muscle SEMG signal of the user.

[0066] In a preferred embodiment, in the time domain, assume that the processed original SEMG signal obtains {x{i}|(i = 1, 2, 3...N)}, where N represents the number of sample points. The feature extraction calculation formula for this one-dimensional signal is as follows:

[0067] 1) The mean absolute value (MAV) in the index information is:

[0068] In this formula, N is the number of surface electromyogram signal data points collected, and x(i) is the i-th data in this signal data sequence. That is, first find the absolute value of the amplitude of the electromyogram signal point by point, sum up these absolute values, and then divide by the total number of sample points, so as to obtain the average level of the amplitude of the electromyogram signal during this period;

[0069] 2) The root mean square (RMS) in the index information is:

[0070] During the operation, first square the amplitude of each sampling point in the electromyogram signal sequence, sum up all the squared values and take the average, and then take the square root to obtain the RMS value;

[0071] 3) The integrated electromyogram value (iEMG) in the index information is: That is, directly sum up the absolute values of the electromyogram signal amplitude;

[0072] 4) The maximum peak time t1: Accurately lock the moment when the amplitude of the electromyogram signal first reaches the maximum value in the time domain waveform, which is the maximum peak time t1; it measures the time taken by the pelvic floor muscle from the moment of receiving the contraction instruction to the full burst and reaching the maximum contraction force from the time dimension.

[0073] In the frequency domain, perform a fast Fourier transform on the original SEMG signal to obtain the frequency components and amplitude information of the signal. Assume that the power spectral density of the user's surface electromyogram signal is PSD(f), and use the two indicators of the median frequency (MF) and the mean power frequency (MPF) in the relevant information to characterize the characteristics of the user's SEMG spectrum or power spectrum:

[0074] 1) Median frequency MF: The median frequency is the frequency that divides the power spectral density function PSD of the electromyogram signal into two equal-area parts;

[0075] 2) Mean Power Frequency (MPF), which represents the average distribution position of the myoelectric signal power on the frequency axis.

[0076] A specific division in the present invention for comparing the acquired signal indicators with the expected indicators:

[0077] (1) Muscle Strength Scoring Criteria (Muscle Strength Screening)

[0078] Muscle strength scoring mainly depends on three core indicators: MAV (Mean Activity Potential), RMS (Root Mean Square), and iEMG (Integrated Electromyogram). According to these indicators, different scoring intervals are defined, and the user data is compared with the large sample statistics of the normal population.

[0079] Scoring range:

[0080] A. Normal: The values of MAV, RMS, and iEMG are within the normal range (determined according to the reference data of the same age group and gender).

[0081] B. Slight Deficiency: MAV and RMS are at the lower limit of the normal range, and iEMG is slightly lower, indicating slightly weak muscles.

[0082] C. Moderate Deficiency: MAV and RMS are significantly lower than the lower limit of the normal range, and iEMG is smaller, indicating insufficient muscle strength.

[0083] D. Severe Deficiency: MAV and RMS are significantly lower than the lower limit of the normal range, and iEMG is much lower than the normal range, indicating severely insufficient pelvic floor muscle contraction strength.

[0084] Scoring Standard: Normal: 80 - 100 points; Slight Deficiency: 60 - 79 points; Moderate Deficiency: 40 - 59 points; Severe Deficiency: 0 - 39 points

[0085] (2) Muscle Endurance Scoring Criteria (Muscle Endurance Screening)

[0086] Muscle endurance screening evaluates the continuous working ability of the pelvic floor muscles by observing MF (Frequency Center) and MPF (Main Frequency Point).

[0087] Scoring range:

[0088] A. Normal: MF decreases steadily and slowly, and MPF is stable or fluctuates slightly.

[0089] B. Slight Fatigue: MF gradually decreases, MPF shifts slightly, and endurance decreases slightly.

[0090] C. Moderate Fatigue: MF decreases significantly, MPF shifts greatly, indicating pelvic floor muscle fatigue.

[0091] D. Severe fatigue: The MF drops rapidly, and the MPF drifts towards low frequencies, indicating long-term fatigue of the pelvic floor muscles and poor recovery ability.

[0092] Scoring criteria: Normal: 80 - 100 points; Mild fatigue: 60 - 79 points; Moderate fatigue: 40 - 59 points; Severe fatigue: 0 - 39 points

[0093] (3) Response speed scoring criteria (response speed screening)

[0094] The response speed assesses the emergency response ability of the pelvic floor muscles, based on the maximum peak time t1. According to the sample of healthy people, the response time threshold can be set.

[0095] Scoring range:

[0096] A. Normal: The response time t1 is within dozens of milliseconds.

[0097] B. Mild delay: The response time t1 exceeds the normal range but is still within 100 milliseconds.

[0098] C. Moderate delay: The response time t1 is between 100 - 200 milliseconds, indicating abnormal nerve or muscle activation.

[0099] D. Severe delay: The response time t1 exceeds 200 milliseconds, indicating severe impairment of the nerve conduction and activation ability of the pelvic floor muscles.

[0100] Scoring criteria: Normal: 80 - 100 points; Mild delay: 60 - 79 points; Moderate delay: 40 - 59 points; Severe delay: 0 - 39 points

[0101] (4) Coordination scoring criteria (coordination screening)

[0102] The coordination screening determines the collaborative working ability of the pelvic floor muscles by observing parameters such as the waveform, frequency, and amplitude of the electromyogram signals of each channel.

[0103] Scoring range:

[0104] A. Normal: The signal synchronization of each channel is good, and the start time, peak time, and end time are relatively consistent.

[0105] B. Mild incoordination: The signals of some channels are slightly lagging or advancing, with small - scale problems in coordination.

[0106] C. Moderate incoordination: The signals of multiple channels are significantly incoordinated, and there may be problems with the coordination of some pelvic floor muscle groups.

[0107] D. Severe incoordination: The multi - channel signals are incoordinated, and there are serious obstacles to the collaborative work of the pelvic floor muscle groups.

[0108] Scoring Criteria: Normal: 80 - 100 points; Mild Incoordination: 60 - 79 points; Moderate Incoordination: 40 - 59 points; Severe Incoordination: 0 - 39 points

[0109] (5) Fatigue Scoring Criteria (Fatigue Screening)

[0110] Fatigue screening comprehensively evaluates the fatigue status of the pelvic floor muscles mainly through changes in time-domain and frequency-domain indicators, with a focus on observing the attenuation slopes of MAV, RMS, iEMG and the drift direction and speed of MF and MPF.

[0111] Scoring Range:

[0112] A. Normal: MAV, RMS, and iEMG gradually attenuate, and MF and MPF slowly shift to lower frequencies, showing a linear and gentle attenuation trend.

[0113] B. Mild Fatigue: MAV, RMS, and iEMG show a slightly accelerated attenuation, MF and MPF shift slightly, and endurance decreases.

[0114] C. Moderate Fatigue: MAV, RMS, and iEMG significantly attenuate, MF and MPF accelerate their shift, and pelvic floor muscle fatigue is more obvious.

[0115] D. Severe Fatigue: MAV, RMS, and iEMG rapidly attenuate, MF and MPF quickly drift to lower frequencies, and fatigue rapidly worsens.

[0116] Scoring Criteria: Normal: 80 - 100 points; Mild Fatigue: 60 - 79 points; Moderate Fatigue: 40 - 59 points; Severe Fatigue: 0 - 39 points

[0117] (6) Comprehensive Scoring

[0118] Total Score = Muscle Strength Score × 0.2 + Muscle Endurance Score × 0.2 + Coordination Score × 0.2 + Fatigue Score × 0.2 + Reaction Speed Score × 0.2.

[0119] Comprehensive Scoring Range: Normal: 80 - 100 points; Mild Abnormality: 60 - 79 points; Moderate Abnormality: 40 - 59 points; Severe Abnormality: 0 - 39 points

[0120] Among them, this comprehensive score can adjust the scoring weights of each part according to the patient's age, MBI, childbearing status, and medical history to customize a more suitable scoring system for the patient.

[0121] (7) Generating Training Guidelines

[0122] a. Quick Contraction Training (Improving Muscle Strength and Reaction Speed)

[0123] Training Objectives: Improve the quick activation ability and emergency response ability of the pelvic floor muscles, and improve nerve conduction and the instantaneous explosive power of muscles.

[0124] Action demonstration: Under the guidance of medical staff, the user instantly tightens the pelvic floor muscles to the extreme, holds for 3 - 5 seconds, and simulates situations such as sudden coughing and sneezing.

[0125] Training frequency: Each training session consists of 8 - 10 repetitions, with a 1 - 2 - minute rest interval between each.

[0126] Training intensity: According to the scoring criteria, during rapid contractions, it is necessary to reach medium - high intensity for MAV and RMS. If the reaction time t1 exceeds 150 milliseconds, the reaction speed can be improved by increasing the training frequency or shortening the relaxation time.

[0127] Progression method: Gradually increase the intensity of contraction, extend the duration of high - intensity contraction (5 seconds), and gradually increase the challenge of the training.

[0128] Training effect evaluation: If the reaction time t1 continuously remains below 100 milliseconds and both MAV, RMS, and iEMG show an increase, it indicates that the rapid activation ability of the pelvic floor muscles has been improved.

[0129] If MAV or RMS is significantly lower than the normal range, continue with low - intensity, short - time rapid contraction training.

[0130] b. Slow and sustained contraction training (to improve muscle endurance and stability)

[0131] Training objective: Enhance the continuous contraction ability and endurance of the pelvic floor muscles, and be able to maintain a stable tension for a long time.

[0132] Action demonstration: Slowly tighten the pelvic floor muscles, gradually reach the maximum contraction degree in 5 - 10 seconds, and hold for 10 - 15 seconds, then relax.

[0133] Training frequency: Each training session consists of 5 - 8 repetitions, with each contraction lasting 15 - 20 seconds and a relaxation time of 30 - 60 seconds.

[0134] Training intensity: According to the scoring criteria, MF and MPF should remain stable or change slowly during the contraction. If MF significantly decreases or MPF deviates, it is necessary to increase the time and intensity of slow contraction.

[0135] Progression method: Increase the contraction duration, gradually extend the holding time (20 seconds), and enhance the intensity of each contraction to challenge the endurance of the pelvic floor muscles.

[0136] Training effect evaluation: If MF steadily decreases and MPF changes smoothly, it indicates that the endurance of the pelvic floor muscles is improving. If MF and MPF show abnormal fluctuations, it may be necessary to increase the training intensity or extend the holding time of each contraction.

[0137] c. Intermittent contraction training (to improve coordination and flexibility)

[0138] Training objective: Enhance the coordination and flexibility of the pelvic floor muscles during dynamic movements, and simulate alternating contractions and relaxations in daily activities.

[0139] Action demonstration: Perform 8 - 10 times of alternating contractions and relaxations according to the rhythm of "contract for 3 seconds - relax for 2 seconds".

[0140] Training frequency: Each training session consists of 8 - 10 times, and maintain the synchrony and coordination of signals during the training process.

[0141] Training intensity: According to the scoring criteria, pay attention to the synchrony and volatility of the electromyogram signals. If the coordination is poor, the contraction frequency can be appropriately reduced and focus on training the precision of each contraction.

[0142] Progression method: Gradually increase the frequency of the contraction - relaxation rhythm (for example, increase to "contract for 4 seconds - relax for 3 seconds"), and gradually increase the number of training sessions to improve the coordination of dynamic collaboration.

[0143] Training effect evaluation: If the signal synchrony of each channel is good and the fluctuations of RMS and iEMG are small, it indicates good coordination and flexibility. If there are obvious signal lags or advances, it means that the coordination of the pelvic floor muscles needs to be improved, and the training frequency should be continued to increase and pay attention to the signal synchrony.

[0144] The technical solution of the present invention can perform early screening through multi - dimensional electromyogram signal analysis before obvious symptoms are caused by pelvic floor muscle dysfunction. This helps to detect potential problems in advance, such as early identification of insufficient pelvic floor muscle strength in postpartum women, decreased pelvic floor muscle endurance in the elderly, etc., facilitating timely intervention measures and effectively preventing the occurrence of serious pelvic floor muscle dysfunction diseases such as urinary incontinence and pelvic organ prolapse. The evaluation and guidance are completely based on the data analysis of the pelvic floor muscle electromyogram signals, avoiding the subjectivity and uncertainty that may exist in traditional evaluation methods. Specifically, through rigorous mathematical calculations and comparison with expected indicators, it provides scientific and objective data support for the judgment of the pelvic floor muscle state and the formulation of training programs, making the entire health management process more reliable. From data acquisition, signal processing, screening and evaluation to training guidance, it provides a basis for realizing one - stop service. Users only need to complete the specified actions as required to obtain a comprehensive pelvic floor muscle state report and personalized training program, without having to move between different institutions or devices, greatly improving the efficiency and convenience of health management.

[0145] Description of each module in the present invention:

[0146] Signal acquisition module: The core component of this module is a disposable gel electrode. These electrodes are made of high-quality gel materials that can closely adhere to the user's skin, effectively reducing interference and ensuring clear and accurate surface electromyogram (SEMG) signals are collected. For example, when the user performs a pelvic floor muscle contraction action, the signal acquisition module will immediately capture these tiny muscle activity signals, providing raw data for subsequent signal processing. In addition, since the electrodes are disposable and replaced after each use, the risk of cross-infection is effectively avoided, ensuring the user's hygiene and safety.

[0147] Signal processing module: This module contains a signal amplifier for amplifying and preprocessing the collected SEMG signals. The amplifier is designed to clearly capture even weak muscle activity signals and convert them into digital signals for further analysis. For example, when the signal acquisition module captures weak contraction signals of the pelvic floor muscles, the signal amplifier will quickly amplify these signals to ensure that the signal quality meets the requirements of subsequent processing.

[0148] Signal detection and judgment module: This module uses advanced algorithms to analyze the processed SEMG signals to detect key parameters such as the contraction intensity and duration of the muscles. For example, it can accurately determine whether the contraction force of the pelvic floor muscles meets the standard when the user performs a certain rating action, providing precise feedback to the user. At the same time, this module can also predict possible pelvic floor muscle dysfunction based on the user's muscle activity pattern, providing a scientific basis for doctors' diagnosis.

[0149] Report generation module: This module is responsible for converting the analysis results of the signal detection and judgment module into an easy-to-understand report. The report content may include the user's pelvic floor muscle function score, muscle activity charts, and training suggestions, etc. For example, the report may show the contraction force and duration of the pelvic floor muscles when the user completes a specific action, as well as the deviation from the normal range, helping the user understand their pelvic floor muscle function status and formulate a personalized training plan.

[0150] Data storage module: This module is used to store information such as the user's detection data, training records, and generated reports. For example, each user's detection data will be automatically saved for future comparison and analysis. This not only helps the user track their training progress but also provides long-term health monitoring data for doctors.

[0151] In addition, auxiliary modules such as a voice module, a graphics module, a dynamic capture module, and an infrared sensing module are also equipped.

[0152] Voice module: For example, when the user is training, the voice module will broadcast the training instructions in real time to remind the user to maintain the correct posture and rhythm. At the same time, it can also give timely encouragement and feedback based on the user's training progress and performance.

[0153] Graphics module: This module displays intuitive graphics and animations on the device's display to help users better understand the training requirements and movement details. For example, it shows the contraction and relaxation process of the pelvic floor muscles, as well as the effects of different movements on the pelvic floor muscles.

[0154] Motion capture module: This module uses camera or sensor technology to capture the user's motion trajectory in real time. For example, it can monitor key parameters such as the body's tilt angle and arm swing amplitude when the user is performing a certain action, thereby ensuring that the user's action meets the specifications.

[0155] Infrared sensor module: This module automatically adjusts training parameters and feedback methods by sensing the user's body position and movement status. For example, when the user approaches the device, the infrared sensor module automatically starts the training mode; when the user leaves, the training is paused and the data is saved.

[0156] Finally, it should be noted that the present invention is not limited to the above embodiments, and there are many variations. All variations that can be directly derived or associated with the content disclosed by ordinary technicians in this field should be considered as the protection scope of the present invention.

Claims

1. A pelvic floor muscle early screening training method based on data analysis, characterized in that The method comprises the following steps: Step a: The user uses a powered pelvic floor muscle non-invasive detection device to complete a specified rating action and obtain the real-time value of the surface electromyography (SEMG) of the pelvic floor muscle during the action; Step b: firstly process the raw data of the surface electromyographic signal, calculate the evaluation index of the electromyographic signal, and obtain the index information in the time domain, including the mean absolute value MAV, the root mean square RMS, the maximum peak time t1, and the integrated electromyographic value iEMG. Secondly, perform fast Fourier transform on the signal to obtain the frequency domain related information, including the median frequency MF and the mean power frequency MPF. Step c: Compare the acquired signal indicators with the expected indicators and conduct multi-faceted screening, including muscle strength, muscle endurance, reaction speed, coordination, and fatigue; Step d: generating a pelvic floor muscle status report based on the rating results compared with the preset values; If an abnormality occurs, targeted recovery training guidance will be provided based on the comparison between the signal indicators and the expected indicators; Step e: Upload the data to the cloud to generate a data set, so that users can more accurately implement early screening training of pelvic floor muscles based on data analysis.

2. The method for early screening and training of pelvic floor muscles based on data analysis according to claim 1, characterized in that: The non-invasive pelvic floor muscle detection device includes a communication-connected signal acquisition module, a signal processing module, a signal detection and judgment module, a report generation module, and a data storage module. The signal processing module includes a signal amplifier, and the signal acquisition module includes a disposable gel electrode. The signal amplifier is connected to 9 disposable gel electrodes, and the 9 electrodes are divided into 3 groups, each group including a positive electrode, a negative electrode, and a reference electrode.

3. A pelvic floor muscle early screening training method based on data analysis according to claim 1 or 2, characterized in that: The designated rating actions in step a include rapid contraction actions, slow continuous contraction actions, and intermittent contraction actions; signal collection is performed when the designated actions are completed; The rapid contraction action requires the user to instantly tighten the pelvic floor muscles to the maximum for about 3-5 seconds, capturing the high-intensity electromyographic signals corresponding to the instantaneous explosive force of the muscles, and accurately evaluating the nerve conduction speed and rapid activation ability of the pelvic floor muscles. The slow continuous contraction movement requires the user to exert force slowly, gradually reach the maximum contraction level over 5-10 seconds, and then maintain it for 10-15 seconds. It is used to consider muscle endurance and the ability to maintain stable tension. The intermittent contraction movement is set to a cyclic rhythm of "contract for 3 seconds and then relax for 2 seconds", and repeated 8-10 times.

4. The method for early screening and training of pelvic floor muscles based on data analysis according to claim 1, characterized in that: Acquiring the original surface electromyography (SEMG) data in step a includes: Firstly, 50HZ notch processing is performed to eliminate power frequency interference; then the original signal is processed by bandpass filtering; finally, the characteristic value of the user's pelvic floor muscle surface electromyography (SEMG) signal is extracted.

5. A pelvic floor muscle early screening training method based on data analysis according to claim 1, 2 or 4, characterized in that: In the time domain, assume that the original SEMG signal is processed to obtain {x{i}|(i=1.2.3…N)}, where N represents the number of sample points. The feature extraction calculation formula of the one-dimensional signal is as follows: 1) The average absolute value MAV in the indicator information is: In this formula, N is the number of surface electromyographic signal data points collected, and x(i) is the i-th data in the signal data sequence, that is, the absolute value of the electromyographic signal amplitude is first calculated point by point, and these absolute values ​​are accumulated and summed, and then divided by the total number of sample points, so that the average level of the electromyographic signal amplitude in this period can be obtained; 2) The root mean square RMS in the indicator information is: During the calculation, the amplitude of each sampling point in the electromyographic signal sequence is first squared, all square values ​​are accumulated and averaged, and then the square root is taken to obtain the RMS value; 3) The integrated electromyographic value iEMG in the indicator information is: That is, the absolute value of the amplitude of the electromyographic signal is directly accumulated and summed; 4) Maximum peak time t1: The time when the amplitude of the electromyographic signal reaches the maximum value for the first time is accurately locked in the time domain waveform, which is the maximum peak time t1; It measures the time it takes for the pelvic floor muscles to burst into full force and reach maximum contraction force from the moment they receive the contraction command.

6. The method for early screening and training of pelvic floor muscles based on data analysis according to claim 5, characterized in that: In the frequency domain, the original SEMG signal is subjected to fast Fourier transform to obtain the frequency component and amplitude information of the signal. The power spectrum density of the user's surface electromyography signal is assumed to be PSD(f), and the two indicators of the relevant information median frequency MF and the average power frequency MPF are used to characterize the characteristics of the user's SEMG spectrum or power spectrum: 1) Median frequency MF: The median frequency refers to the frequency that divides the power spectral density function PSD of the electromyographic signal into two equal area parts; 2) Mean power frequency MPF, Represents the average distribution position of the EMG signal power on the frequency axis.

7. The method for early screening and training of pelvic floor muscles based on data analysis according to claim 1, characterized in that: The multifaceted screening in step c is as follows: 1) Muscle strength screening: Obtain the data of pelvic floor muscle contraction strength, which is reflected by obtaining the average absolute value MAV, root mean square RMS, and integrated electromyographic value iEMG when the user completes the specified process; The average absolute value MAV reflects the average contraction force, the root mean square RMS highlights the energy release intensity, and the integrated electromyographic value iEMG presents the total amount of electrical activity; And compare the normal muscle strength range generated by statistics of a large sample of people of the same age and gender. If the patient's measured average absolute value MAV and root mean square RMS are lower than the lower limit, and the integrated electromyographic value iEMG is too small, it indicates that the pelvic floor muscle contraction strength is seriously insufficient; 2) Muscle endurance screening: It is evaluated by the median frequency MF and mean power frequency MPF indicators generated during the user's exercise; if the median frequency MF decreases slowly and the mean power frequency MPF is stable, it is considered that the healthy pelvic floor muscles are contracted for a long time; if MF drops sharply and MPF deviates greatly, it is considered that the endurance is poor; 3) Reaction speed screening: using the maximum peak time t1 as the basis; setting a normal threshold; if it far exceeds the threshold, it indicates abnormal nerve conduction and muscle activation. 4) Coordination screening: The coordination of the pelvic floor muscles is judged by observing parameters, including the waveform, frequency and amplitude of the electromyographic signal; Under normal circumstances, when the pelvic floor muscles contract in coordination, the start time, peak time, and end time of the electromyographic signals recorded by each channel corresponding to different electrode positions should be relatively synchronized. If a channel signal is significantly delayed or advanced, it means that the pelvic floor muscles at that part are not well coordinated with other parts. 5) Fatigue screening: Comprehensive changes in indicators in the time domain and frequency domain are used to achieve fatigue screening; in the time domain, the attenuation slope of the mean absolute value MAV, root mean square RMS, and integrated electromyographic value iEMG is observed; in the frequency domain, the drift direction and speed of the median frequency MF and mean power frequency MPF are observed; if the indicators decay linearly and gently, it indicates normal pelvic floor muscle fatigue, and if the indicators drop linearly and exponentially, it indicates abnormal fatigue.

8. The method for early screening and training of pelvic floor muscles based on data analysis according to claim 1, characterized in that: Step d specifically includes: comprehensive multi-dimensional screening of detailed results to achieve close integration of user key information, including age, BMI, and mental health status. The system generates a pelvic floor muscle status report, which includes user basic information, actual measured values ​​of each indicator, details of comparison with expected indicators, health rating, and intuitive charts; the health rating is subdivided into multiple levels, from "excellent" to "severely abnormal" to correspond to the functional level of the pelvic floor muscles.

9. The method for early screening and training of pelvic floor muscles based on data analysis according to claim 2, characterized in that: The recovery training guidance described in step d includes Kegel exercises and pelvic floor muscle yoga; The non-invasive pelvic floor muscle detection device also includes a voice module, a graphics module, a motion capture module, and an infrared sensing module that are communicatively connected to the signal processing module. The voice module, graphics module, motion capture module, and infrared sensing module in the system are combined to achieve real-time prompts in voice and images to achieve feedback on the accuracy of movements for recovery training guidance.

Images