Gynecological postoperative patient rehabilitation training intelligent guiding system and training method

Abstract

The invention relates to the technical field of medical rehabilitation, in particular to an intelligent guiding system for rehabilitation training of a gynecological postoperative patient and a training method.The system collects the muscle activation condition and stress distribution of the patient in the rehabilitation training process in real time through a surface myoelectricity sensor and a flexible pressure sensor training pad, a three-dimensional visualization system is constructed, and the rehabilitation training effect of the patient is improved. And an individual recovery cycle and a posture offset error are quantitatively calculated to adjust a training state, so that the scientificity, individualization and intelligence levels of postoperative rehabilitation training are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of medical rehabilitation, and particularly to an intelligent guidance system and training method for the rehabilitation training of patients after gynecological surgery. Background Art

[0002] Patients after gynecological surgery often suffer from reduced motor function due to surgical trauma, muscle atrophy, or nerve function impairment, especially in the functional recovery of the pelvic floor muscles and abdominal core muscle groups. Traditional rehabilitation training mostly relies on subjective experience and manual guidance, lacking objective quantitative evaluation, resulting in unstable training effects and difficult individualized adjustment.

[0003] In the prior art, the invention patent with publication number CN108939436A proposes an active lower limb training system for coordinated movement of the healthy and affected sides and its operation method: by installing a pressure sensor on the sole of the healthy side to detect the pressure on the sole of the healthy side, when the pressure sensor starts to have pressure, the healthy side starts to move; when the value of the pressure sensor reaches the threshold, the movement of the healthy side ends. This prior art only relies on data from a single dimension such as a pressure sensor and cannot effectively integrate movement patterns, muscle activation, and force distribution, lacking intuitive visual feedback and intelligent guidance.

[0004] In the prior art, the invention patent with publication number CN118902478A proposes a method for identifying gait phase and speed based on electromyography and deep learning: the computer uses the CNN-LSTM model based on electromyography signal data and plantar pressure data to obtain gait speed and gait phase. The method of deep learning artificial intelligence relies on complex parameter settings and model training processes and requires high hardware configuration, which is not simple and efficient.

[0005] Therefore, there is an urgent need for an intelligent rehabilitation training system that can real-time monitor muscle activation and force distribution, perform three-dimensional visual analysis, and can simply and efficiently calculate the individual recovery period to adjust the training status, so as to improve the scientific, individualized, and intelligent level of postoperative rehabilitation training. Summary of the Invention

[0006] The purpose of the present invention is to provide an intelligent rehabilitation training guidance system, which can real-time collect the muscle activation and force distribution of patients during the rehabilitation training process through surface electromyography sensors (sEMG) and flexible pressure sensor pads, and construct a three-dimensional visual feedback system to achieve intelligent training guidance.

[0007] The technical solution of the present invention, an intelligent guidance system for the rehabilitation training of patients after gynecological surgery, is characterized by including:

[0008] A surface electromyography sensor module, which is used to attach to the surface of the target muscle to real-time collect electromyography signals;

[0009] A flexible pressure sensor training pad, which is a training pad for body contact during patient training, and detects pressure distribution data in the body contact area;

[0010] A data fusion and 3D visualization system, which is used to construct a 3D human motion state model based on the collected electromyogram signals and pressure distribution data;

[0011] A quantization feedback module, which is used to calculate the individual recovery period based on the temporal characteristics of electromyogram signals and pressure distribution data, and then calculate the posture deviation error in combination with the training support method to adjust the training state.

[0012] Optionally, the surface electromyogram sensor module includes multiple electrodes that can be adhered to the surfaces of target muscle groups such as the pelvic floor muscles, transverse abdominal muscles, and gluteus maximus muscles.

[0013] Optionally, the flexible pressure sensor training pad adopts a multi-point distributed sensor matrix to detect real-time pressure distribution data during the training process.

[0014] Optionally, the data fusion and 3D visualization system adopts heat map mapping.

[0015] Optionally, the data fusion and 3D visualization system has a 3D coordinate system, where the XY plane is the plane where the training pad is located, which is used to represent the force application position and center of gravity offset detected by the flexible pressure sensor, and the z-axis is the muscle activation degree of different training parts represented by the electromyogram signal.

[0016] Optionally, the offset part obtained from the posture deviation error is marked in the heat map.

[0017] Optionally, the final recovery rate is obtained by taking the average of the recovery rates calculated through several training times with a difference of Δt.

[0018] A training method for a postoperative gynecological patient rehabilitation training intelligent guidance system based on any one of the above, including the following steps:

[0019] Step 1: Paste surface electromyogram sensors on the surface of the target muscle, and place a flexible pressure sensor training pad in the training area;

[0020] Step 2: Collect muscle activation signals and pressure sensor data during the training process;

[0021] Step 3: Use the data fusion and visualization system to generate a real-time 3D heat map to intuitively display the muscle activation degree and force distribution;

[0022] Step 4: Analyze the training data according to the quantization feedback module, calculate the individual recovery period, and adjust the training plan.

[0023] Compared with the existing technologies, the beneficial effects of the present invention are as follows:

[0024] 1. Real-time monitoring of the myoelectric activities of the pelvic floor muscles and core muscle groups during the training of patients to prevent incorrect compensation;

[0025] 2. Compared with only setting pressure sensors on the soles of the feet, the flexible pressure sensor training pad can more comprehensively detect the body force distribution of patients and judge whether the training posture is correct;

[0026] 3. Construct a personalized three-dimensional rehabilitation training model, provide dynamic visual feedback, and optimize the training actions;

[0027] 4. Conduct quantitative feedback, calculate the individual recovery period and posture deviation error, and adjust the training status;

[0028] 5. The algorithm is simple and efficient, with a small amount of calculation, low hardware requirements, and high system accuracy and stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 Schematic diagram of the overall system framework in an embodiment of the present invention;

[0030] Figure 2 Schematic diagram of rehabilitation training in an embodiment of the present invention;

[0031] Figure 3 Schematic diagram of the steps of the training method in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0032] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments.

[0033] Referring to the schematic diagram of the overall system framework shown in the appended Figure 1 and the schematic diagram of rehabilitation training shown in the appended Figure 2 wherein 1 is a flexible pressure sensor training pad, 2 is a surface electromyography sensor module, and 3 is a distributed sensor matrix, the technical solution of the present invention,

[0034] An intelligent guidance system for the rehabilitation training of gynecological postoperative patients, characterized by comprising:

[0035] A surface electromyography sensor module for attaching to the surface of the target muscle to collect myoelectric signals in real time;

[0036] A flexible pressure sensor training pad, which is a training pad for the patient's body contact during training and detects the pressure distribution data of the body contact area;

[0037] A data fusion and 3D visualization system for constructing a 3D human motion state model based on the collected electromyography signals and pressure distribution data.

[0038] A quantization feedback module for calculating an individual recovery period based on the temporal characteristics of electromyography signals and pressure distribution data, and then calculating the posture deviation error in combination with the training support method to adjust the training state.

[0039] Optionally, the surface electromyography sensor module includes multiple electrodes that can be adhered to the surfaces of target muscle groups such as the pelvic floor muscles, transverse abdominal muscles, and gluteus maximus muscles.

[0040] Specifically, a surface electromyography (sEMG) sensor is a non-invasive bioelectrical signal acquisition device mainly used to monitor the electrical activities of target muscles, help evaluate the muscle function status, and guide rehabilitation training.

[0041] The surface electromyography sensor module mainly consists of the following parts:

[0042] (1) An electrode sensing unit using Ag / AgCl (silver / silver chloride) electrodes, attached to the skin surface to collect bioelectrical signals (in mV level) generated during muscle contraction. Applicable parts: Muscle groups related to postoperative rehabilitation such as pelvic floor muscles, rectus abdominis muscles, and gluteal muscles.

[0043] (2) A signal amplifier. Since the amplitude of electromyography signals is relatively low (generally 0 - 5 mV), a high-gain amplifier (10 - 1000 times) is required. It is designed with a high common-mode rejection ratio (CMRR) to reduce external interference.

[0044] (3) Filtering and signal processing, including low-pass filtering (typically 10 - 500 Hz): Removing motion artifacts and environmental noise; high-pass filtering (> 10 Hz): Removing baseline drift and improving signal stability; band-pass filtering (20 - 450 Hz): Retaining effective electromyography signals.

[0045] (4) Data acquisition and wireless transmission. Sampling rate: 500 - 2000 Hz, ensuring high-precision signal acquisition. Using Bluetooth / wireless Wi-Fi to transmit data to a computer or mobile device in real time.

[0046] Optionally, the flexible pressure sensor training pad adopts a multi-point distributed sensor matrix to detect real-time pressure distribution data during the training process.

[0047] Specifically, the flexible pressure sensor training mat is an intelligent training device based on a distributed pressure sensing matrix. Its size can be set to the dimensions of a conventional yoga mat. The distributed pressure sensing matrix is arranged in the middle layer of the mat, enabling real-time monitoring of the force distribution in the areas where the patient's body makes contact. It is used for posture correction, load assessment, and rehabilitation progress tracking during postoperative rehabilitation training.

[0048] The flexible pressure sensor training mat consists of the following key components:

[0049] (1) Distributed pressure sensing matrix, working principle: Using piezoresistive, piezoelectric, or capacitive sensors to detect the pressure changes when the body contacts the training mat. Resolution: The size of a unit pressure sensing element (pixel) is usually 1 cm 2 , and the distance between two pressure sensing elements is 1 - 3 times the size of a unit pressure sensing element.

[0050] (2) Sensing material, which can be piezoresistive (materials: conductive polymers such as PEDOT:PSS, carbon nanotubes, graphene), or piezoelectric (materials: PVDF thin films, PZT ceramics, etc.).

[0051] (3) Data acquisition and wireless transmission, ADC (analog - to - digital converter): Converting the analog pressure signal into a digital signal, using Bluetooth / Wi - Fi to transmit data to a computer or mobile device in real - time. Refresh rate: 30 - 100 Hz to ensure real - time monitoring of pressure changes.

[0052] Optionally, the data fusion and 3D visualization system uses heat map mapping.

[0053] Optionally, the data fusion and 3D visualization system has a 3D coordinate system, where the XY plane is the plane where the training mat is located, used to represent the force - applied positions and center - of - gravity offsets detected by the flexible pressure sensors, and the z - axis represents the muscle activation levels of different training parts indicated by electromyography signals.

[0054] Specifically, the data fusion and 3D visualization system is used to receive, process, and visualize the data collected by surface electromyography (sEMG) and flexible pressure sensors. Through a 3D human model and heat map mapping, it intuitively shows the muscle activation degree and force distribution of the patient, realizing intelligent rehabilitation training guidance.

[0055] Specifically, the data fusion and 3D visualization system consists of a data acquisition module, a data fusion algorithm, and a 3D visualization engine:

[0056] (1) Data acquisition module

[0057] Surface electromyography sensor (sEMG): Collects muscle activation electrical signals to determine the degree of muscle contraction.

[0058] Flexible pressure sensor: Collects the force distribution on the training pad (N / m 2 ), and judges the contact area and force change.

[0059] Time synchronization module: Ensures that the electromyography and pressure data are aligned on the same time axis to improve the analysis accuracy.

[0060] (2) Data fusion algorithm

[0061] Fuses the sEMG and flexible pressure data to reflect the force patterns of different muscle groups:

[0062] XY coordinates (horizontal plane): Correspond to the force application positions of the pressure sensors.

[0063] Z coordinates (vertical direction): Represent the muscle activation intensity (mV).

[0064] Fusion method: Adopts weighted mapping and interpolation algorithms to make the data show smooth changes on the 3D model.

[0065] (3) 3D visualization engine

[0066] Uses a 3D model combined with a heat map mapping to visually display the training effect:

[0067] 3D human body model:

[0068] Adopts a standard human skeletal model (the height and weight parameters can be adjusted to fit the individual).

[0069] Key muscle groups (such as pelvic floor muscles and rectus abdominis) are color-coded for easy observation.

[0070] Heat map mapping:

[0071] The color gradient represents the degree of muscle activation (red = high activation, blue = low activation).

[0072] The force distribution is displayed in color or grid deformation to identify high-pressure areas and avoid incorrect training.

[0073] Interactive rotation perspective allows patients and doctors to view the data from different angles.

[0074] Example:

[0075] Correct training action → The heat map color of the target muscle group is uniform and the pressure distribution is reasonable.

[0076] Incorrect training method → Abnormal activation (too high electromyography) occurs in non-target muscle groups, or the force is uneven.

[0077] The quantization feedback module is used to calculate the individual recovery cycle based on the temporal characteristics of the electromyogram signal and the pressure distribution data, and then calculate the posture deviation error in combination with the training support method to adjust the training state.

[0078] The specific formula for calculating the individual recovery cycle is as follows:

[0079]

[0080] where S target is the ideal rehabilitation target value; S current is the current state value; r is the recovery rate; F is the individualization factor.

[0081] Specifically, in order to calculate the individual recovery cycle, we first determine that the core goal of calculating the individual recovery cycle (RecoveryCycle, RC) is to accurately predict the time required for the patient to recover from the current state to the target state, so as to dynamically adjust the rehabilitation training plan. Therefore, parameters (target state, current state, recovery rate, individualization factor) are extracted from the three key dimensions of the rehabilitation process, patient individual differences, and recovery rate characteristics to ensure that the calculation is both scientific and adaptable to individual differences.

[0082] 1. Target state S target Determines the rehabilitation endpoint of the patient, that is, the ultimate goal of training, and the target value can be determined through medical standards, preoperative data, and healthy population data.

[0083] With different target states, the recovery cycle will change significantly. For example: the goal of a mild surgery patient is to resume daily activities (faster recovery); competitive athletes may need to recover to a high-intensity training level (longer recovery time).

[0084] The following are examples of sensor variable data for the target state:

[0085] Variable <![CDATA[Target state (S target )]]> <![CDATA[Center of Pressure shift (CoP dev )]]> 0 (fully balanced) Symmetry (Sym) 1 (left - right balanced) <![CDATA[Muscle RMS signal (EMG RMS )]]> ≥0.5 (fully activated)

[0086] Therefore, S target =(0, 1, 0.5).

[0087] 2. Current state S current Reflects the patient's real-time rehabilitation progress, that is, the current functional level.

[0088] By collecting surface electromyogram (EMG) and flexible pressure sensor data through sensors, the patient's current motor ability, muscle activation, balance, etc. can be obtained. When calculating the recovery cycle, it is necessary to know where the patient starts to recover in order to measure the progress.

[0089] The following are examples of sensor variable data for the current state:

[0090] Variable <![CDATA[Current state (S current )]]> <![CDATA[Center of Pressure (CoP dev )]]> 3.6 (poor balance) Symmetry (Sym) 0.6 (still biased towards one side) <![CDATA[Muscle RMS signal (EMG RMS )]]> 0.36 (partially activated)

[0091] Therefore, S current (t) = (3.6, 0.6, 0.3).

[0092] 3. The calculation process of the recovery rate r is as follows: The first-order difference method is used to calculate the recovery rate r of the state variable

[0093]

[0094] where S t and S t-1 are the state variables obtained by sensors with a training time difference of Δt.

[0095] Optionally, the final recovery rate is obtained by taking the average of the recovery rates calculated through several training times with a difference of Δt.

[0096] Specifically, the recovery rate r represents the improvement speed of the patient after training. The calculation method is to observe the data change trend of the past few trainings and estimate the recovery speed through the time slope.

[0097] If k is large, it indicates a fast recovery and the required rehabilitation cycle RC will be shortened; if k is small, it indicates a slow recovery and RC will be extended.

[0098] We assume that there is rehabilitation training data for 6 weeks, collected once a week. The data is as follows:

[0099]

[0100]

[0101] Use the first-order difference method to calculate the change rate of the state variable:

[0102]

[0103] where Δt = 1 (unit: week).

[0104] (1) Center of gravity offset recovery rate r CoP

[0105]

[0106] r CoP = [-0.5, -0.8, -0.7, -0.7, -0.6]

[0107] Take the mean:

[0108]

[0109] (2) Symmetry recovery rate r Sym

[0110]

[0111] r Sym = [0.05, 0.08, 0.09, 0.09, 0.07]

[0112] Take the average value:

[0113]

[0114] (3) Muscle RMS value recovery rate r EMG

[0115]

[0116] r EMG = [0.05, 0.07, 0.06, 0.07, 0.05]

[0117] Take the average value:

[0118]

[0119] Finally, the recovery rate is the average value of the recovery rates of multiple sensor data:

[0120]

[0121] 4. Individualized factor F, used to correct the recovery ability of different individuals, because age and BMI (Body Mass Index) will affect the postoperative recovery speed. Its calculation formula is as follows:

[0122]

[0123] Among them, taking a person aged 25 with a BMI of 22 as the reference value, any individual factor exceeding the reference will affect the recovery rate, and the weight factor 0.05 is used to moderately adjust the recovery period to avoid excessive influence of age or BMI.

[0124] Optionally, 0.05, 25, and 22 can all be adjusted according to actual test data to ensure the rationality of the prediction.

[0125] Suppose there are two patients:

[0126] Patient A: 35 years old, BMI = 28;

[0127] Patient B: 60 years old, BMI = 35;

[0128] Calculate the individualized factor F respectively:

[0129]

[0130]

[0131] Conclusion:

[0132] The recovery period of patient A increased by 13% (due to slightly higher BMI);

[0133] The recovery period of patient B increased by 20% (due to older age and higher BMI).

[0134] 5. Calculate the recovery period

[0135]

[0136] where S target = (0, 1, 0.5);

[0137] S current (t) = (3.6, 0.6, 0.3);

[0138] r = 0.265; F = 1.13 (calculated with a 35-year-old patient and a BMI of 28 as an example);

[0139] Component calculation: ΔS = (0 - 3.6, 1 - 0.6, 0.5 - 0.3) = (-3.6, 0.4, 0.2)

[0140] Finally, the difference S target - S current , is the average state difference

[0141] Finally

[0142] 6. Adjust the training state according to the recovery period (RC)

[0143] The calculated RC represents the time required for the patient to recover from the current state to the target state. According to the length of the RC, we can adjust the training plan in stages.

[0144] (1) Long recovery period (RC > 10 weeks): The training load is lighter, focusing on restoring muscle perception

[0145] Applicable population: Those in the early postoperative period or with slow rehabilitation progress

[0146] Adjustment strategy:

[0147] Reduce the training intensity: Reduce the training duration, 5 - 10 minutes per training session, and gradually increase to 15 - 20 minutes.

[0148] Optimize the training movements: Prioritize low-load isometric training (such as static contractions) to reduce muscle fatigue.

[0149] Enhanced Perception Training: Enhance the patient's awareness of correct posture through biofeedback (such as 3D model visualization).

[0150] Improve Symmetry Control: If the pressure sensor detects a large center of gravity shift, guide the patient to perform corrective exercises.

[0151] (2) Medium Recovery Period (5 weeks < RC ≤ 10 weeks): Moderate training load, strengthen muscle control ability

[0152] Target Population: Patients in the mid - stage of postoperative rehabilitation

[0153] Adjustment Strategy:

[0154] Appropriately increase the training intensity: Increase the training duration to 20 - 30 minutes, combined with dynamic muscle strength training (such as light resistance exercise).

[0155] Improve muscle endurance training: Adopt isotonic training (such as slow pelvic floor muscle contraction and relaxation).

[0156] Optimize symmetry: If there is a large difference in muscle activation between the left and right, adjust the training intensity through biofeedback.

[0157] Adjust the training rhythm: Increase intermittent training to improve muscle recovery ability.

[0158] (3) Short Recovery Period (RC ≤ 5 weeks): High training load, enhance strength and functionality

[0159] Target Population: In the late stage of rehabilitation, approaching the target state

[0160] Adjustment Strategy:

[0161] Increase the training intensity: Gradually transition to high - intensity functional training (such as resistance training, stability training).

[0162] Improve pressure balance: Monitor the pressure distribution of the training mat to ensure uniform force application and improve symmetry.

[0163] Strengthen coordination and dynamic balance: Guide the patient to perform dynamic training, such as single - leg standing, gait training.

[0164] Reduce dependence on biofeedback: Let the patient gradually get rid of external assistance and enhance self - control ability.

[0165] Preferably, considering the impact of postoperative weeks on recovery, according to the non - linear characteristics of postoperative recovery:

[0166] Early recovery is relatively fast (1 - 4 weeks after surgery), due to the strong initial repair mechanism, the recovery rate is high;

[0167] Mid - stage recovery slows down (5 - 12 weeks after surgery), tissue repair enters a stable stage, and the recovery rate decreases;

[0168] Recovery is even slower in the later stage (more than 12 weeks after surgery), mainly relying on long-term training and adaptation, and the recovery rate further decreases.

[0169] Therefore, the influence of the number of weeks after surgery on the recovery rate can be expressed as r×F 术后 , where the influence factor F of the number of weeks after surgery 术后

[0170] F 术后 = e -0.15T

[0171] At this time, the improved calculation formula for the individual recovery period

[0172]

[0173] T represents the number of weeks after surgery (unit: week).

[0174] The exponential decay model e -0.15T reflects the physiological characteristics of faster recovery in the initial stage of rehabilitation and gradually slower recovery in the later stage:

[0175] 0 weeks after surgery: F 术后 = 1.00 (maximum recovery rate).

[0176] 4 weeks after surgery: F 术后 ≈ 0.55 (recovery speed increase).

[0177] 12 weeks after surgery: F 术后 ≈ 0.17 (longer recovery speed).

[0178] Function: The recovery speed is fast in the early stage because the physiological repair ability is strong in the initial stage after surgery; the recovery speed becomes slow in the later stage because the tissue repair is approaching stability and the training effect decreases.

[0179] Combined with the training support method to calculate the posture deviation error, specifically:

[0180] Based on the collected electromyographic signals, pressure data and training support methods, we calculate the posture deviation error (PDE). The main steps are as follows:

[0181] (1) Temporal feature extraction

[0182] A. Electromyographic signal feature extraction

[0183] Calculate the electromyographic time-frequency features by short-time Fourier transform (STFT) or continuous wavelet transform (CWT), and calculate the muscle activation sequence error:

[0184] Among them: is the starting time of the activation of the patient's muscle i; The starting time of muscle activation in the standard posture.

[0185] B. Pressure distribution feature extraction

[0186] Calculate the cosine similarity of the pressure distribution:

[0187]

[0188] where P real is the pressure distribution vector at the current moment; P ref is the pressure distribution vector corresponding to the standard training support method; if the similarity is low, it indicates that the posture support method is incorrect.

[0189] Calculate the center of pressure (CoP) offset:

[0190] E CoP = ||CoP real - CoP ref ||

[0191] where CoP real is the current center of gravity position vector; CoP ref is the center of gravity position vector corresponding to the standard training support method; this error measures the stability of the patient during training, and the larger the offset, the less stable the posture.

[0192] (2) Calculate the overall posture offset error in combination with the training support method

[0193] The training support method determines the standard force application area and the center of gravity position. Therefore, calculate the posture offset error in different postures: E pose = w1E EMG + w2E CoP + w3(1 - Sim pressure ) where:

[0194] E EMG reflects the time error of the muscle activation pattern;

[0195] E CoP measures the stability of the training center of gravity;

[0196] Sim pressure evaluates the force application method of the patient during training;

[0197] w1, w2, w3 are the corresponding weight factors, which can be calculated and determined by the method of variance contribution.

[0198] Mark the offset part obtained from the posture offset error in the heat map. Specifically:

[0199] 1. 3D human model construction, using OpenPose, MediaPipe or LSTM-based 3D models for human pose reconstruction. Identify and locate key joint points (JointKeypoints), such as shoulders, elbows, hips, knees, and ankles.

[0200] 2. Mark the offset areas obtained by combining the offset error in the heatmap.

[0201] Perform heat marking on the abnormal electromyogram area:

[0202] Calculate the time difference between the actually activated muscles and the standard activated muscles.

[0203] If the error is too large (such as incorrect muscle activation), mark the muscle points in red.

[0204] If the muscle is overly tense (such as too high EMG amplitude), mark it in yellow.

[0205] Perform heat marking on the abnormal pressure area:

[0206] Excessive offset of the force application point (such as abnormal CoP) → Highlight the force application area of the training pad in red.

[0207] Uneven force application (such as excessive unilateral force application) → Display the degree of unevenness with a gradient color (yellow → red).

[0208] Overall posture offset marking:

[0209] Combined with the posture matching algorithm, mark the areas where the posture offset error exceeds a certain threshold. A training method for an intelligent guidance system for the rehabilitation training of gynecological postoperative patients based on any one of the above, including the following steps:

[0210] Step 1: Paste surface electromyogram sensors on the surface of the target muscles and place a flexible pressure sensor training pad in the training area;

[0211] Step 2: Collect muscle activation signals and pressure sensor data during the training process;

[0212] Step 3: Use the data fusion and visualization system to generate a real-time three-dimensional heatmap to visually display the muscle activation degree and force distribution;

[0213] Step 4: Analyze the training data according to the quantization feedback module, calculate the individual recovery period, and adjust the training plan.

[0214] Through the above steps, the present invention can:

[0215] 1. Real-time monitor the electromyogram activities of the pelvic floor muscles and core muscle groups during the training process of patients to prevent incorrect compensation;

[0216] 2. Detect the body force distribution of the patient to determine whether the training posture is correct;

[0217] 3. Construct a personalized three-dimensional rehabilitation training model, provide dynamic visual feedback, and optimize the training actions;

[0218] 4. Conduct quantitative feedback, calculate the individual recovery period, and adjust the training status;

[0219] 5. The algorithm is simple and efficient, with a small amount of calculation, low hardware requirements, and high system accuracy and stability.

[0220] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (such as infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that the computer can access or a data storage device such as a server or data center that includes one or more collections of available media. The available media can be magnetic media (such as floppy disks, hard disks, magnetic tapes), optical media (such as DVDs), or semiconductor media. The semiconductor media can be a solid-state drive.

[0221] Those of ordinary skill in the art can realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be implemented by electronic hardware or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professionals can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0222] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0223] In several embodiments provided in the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division of an underwater topographic change analysis system and method for waterways. In actual implementation, there can be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of the devices or units can be in electrical, mechanical, or other forms.

[0224] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0225] In addition, in each embodiment of the present application, the functional units can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit.

[0226] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present application. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store program codes.

[0227] The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed in the present application can easily think of changes or substitutions, which should all be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0228] The above are only specific embodiments of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered by the protection scope of the present application.

Claims

1. An intelligent guidance system for the rehabilitation training of gynecological postoperative patients, characterized in that, Comprising: A surface electromyography sensor module for attaching to the surface of a target muscle to collect electromyography signals in real time; A flexible pressure sensor training mat for a training mat that comes into contact with the patient's body during training to detect pressure distribution data in the area of contact with the body; A data fusion and three-dimensional visualization system for constructing a three-dimensional human motion state model based on the collected electromyography signals and pressure distribution data; A quantization feedback module for calculating an individual recovery period based on the temporal characteristics of the electromyography signals and pressure distribution data, and then calculating a posture deviation error in combination with the training support method to adjust the training state.

2. The system according to claim 1, characterized in that The surface electromyography sensor module includes a plurality of electrodes that can be adhered to the surfaces of target muscle groups such as the pelvic floor muscle, transverse abdominal muscle, and gluteus maximus muscle.

3. The system according to claim 1, wherein The flexible pressure sensor training mat uses a multi-point distributed sensor matrix to detect real-time pressure distribution data during training.

4. The system according to claim 1, characterized in that, The data fusion and three-dimensional visualization system uses heat map mapping.

5. The system according to claim 4, characterized in that, The data fusion and three-dimensional visualization system has a three-dimensional coordinate system, where the XY plane is the plane where the training mat is located, used to represent the force application position and center of gravity deviation detected by the flexible pressure sensor, and the z-axis is the muscle activation degree of different training parts represented by the electromyography signal.

6. The system according to claim 4 or 5, characterized in that, Mark the offset part obtained from the posture deviation error in the heat map.

7. A training method for a postoperative gynecological patient rehabilitation training intelligent guidance system according to any one of claims 1-6, comprising the following steps: Step 1: Paste a surface electromyography sensor on the surface of the target muscle and place a flexible pressure sensor training mat in the training area; Step 2: Collect muscle activation signals and pressure sensor data during training; Step 3: Use the data fusion and visualization system to generate a real-time three-dimensional heat map to visually display the muscle activation degree and force distribution; Step 4: Analyze the training data according to the quantization feedback module, calculate the individual recovery period, and adjust the training plan.

Images