CN118830850B - Muscle strength recovery prediction method, system and equipment - Google Patents
Muscle strength recovery prediction method, system and equipment Download PDFInfo
- Publication number
- CN118830850B CN118830850B CN202411321832.3A CN202411321832A CN118830850B CN 118830850 B CN118830850 B CN 118830850B CN 202411321832 A CN202411321832 A CN 202411321832A CN 118830850 B CN118830850 B CN 118830850B
- Authority
- CN
- China
- Prior art keywords
- interpolated
- muscle strength
- interval
- sample
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/22—Ergometry; Measuring muscular strength or the force of a muscular blow
- A61B5/224—Measuring muscular strength
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/316—Modalities, i.e. specific diagnostic methods
- A61B5/389—Electromyography [EMG]
- A61B5/397—Analysis of electromyograms
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7271—Specific aspects of physiological measurement analysis
- A61B5/7275—Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H20/00—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
- G16H20/30—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to physical therapies or activities, e.g. physiotherapy, acupressure or exercising
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/02—Preprocessing
- G06F2218/04—Denoising
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Medical Informatics (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Biophysics (AREA)
- Physics & Mathematics (AREA)
- Animal Behavior & Ethology (AREA)
- Molecular Biology (AREA)
- Heart & Thoracic Surgery (AREA)
- Biomedical Technology (AREA)
- Surgery (AREA)
- Pathology (AREA)
- Veterinary Medicine (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physical Education & Sports Medicine (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Primary Health Care (AREA)
- Physiology (AREA)
- Psychiatry (AREA)
- Signal Processing (AREA)
- Epidemiology (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Biology (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
本申请公开了一种肌力恢复预测方法、系统及设备，采集样本肌力数据和样本sEMG信号，获取总待插值个数；根据样本肌力数据的数值波动情况对总待插值个数进行调整，得到各待插值区间的若干个待插值点；获取各待插值区间中每一待插值点的预测值，构建每一待插值点的调整因子，并利用调整因子和样本sEMG信号对每一待插值点的预测值进行修正，得到修正后的样本肌力数据；建立肌力预测模型；采集被试者所需预测肌力处的肌电信号，将被试者所需预测肌力处的肌电信号输入所述肌力预测模型中，得到被试者的肌力预测数值。本申请解决了构建肌力预测模型时，对肌力数据进行插值时使用固定数值造成肌力预测模型构建出现偏差的问题。
The present application discloses a method, system and device for predicting muscle strength recovery, which collects sample muscle strength data and sample sEMG signals, obtains the total number of interpolation values; adjusts the total number of interpolation values according to the numerical fluctuation of the sample muscle strength data, and obtains a number of interpolation points in each interpolation interval; obtains the predicted value of each interpolation point in each interpolation interval, constructs an adjustment factor for each interpolation point, and uses the adjustment factor and the sample sEMG signal to correct the predicted value of each interpolation point to obtain the corrected sample muscle strength data; establishes a muscle strength prediction model; collects the electromyographic signal at the muscle strength point required for prediction of the subject, inputs the electromyographic signal at the muscle strength point required for prediction of the subject into the muscle strength prediction model, and obtains the predicted value of the subject's muscle strength. The present application solves the problem that when constructing a muscle strength prediction model, a fixed value is used when interpolating muscle strength data, causing deviations in the construction of a muscle strength prediction model.
Description
Technical Field
The application relates to the technical field of training prediction, in particular to a method, a system and equipment for predicting muscle strength recovery.
Background
In daily life, the damage of the muscle frequently occurs, and the muscle strength is the maximum force which can be generated by the muscle and is an important index for measuring the health degree of the muscle.
In the field of rehabilitation medicine, research on muscle force mainly uses surface electromyographic signals, i.e. semgs, which are superposition of action potentials of a motion unit in numerous muscle fibers in the time domain. In the existing studies, sEMG signals are commonly used to assess muscle strength. In a common muscle strength prediction method, a muscle strength prediction model is generally constructed according to a plurality of sEMG signals and muscle strength signals acquired in advance, and then the sEMG signals of the muscle to be detected of a tested person are input into the muscle strength prediction model to obtain a muscle strength prediction result.
However, when the muscle strength prediction model is constructed, the sampling frequency of the sEMG signal is different from the sampling frequency of the muscle strength signal, namely, the sampling points in the same time are different, the sampling points of the sEMG signal and the muscle strength signal are larger, the sEMG signal and the muscle strength signal cannot be directly matched synchronously, so that the muscle strength model cannot be directly constructed, the muscle strength signal is required to be interpolated due to the fact that the sampling frequency of the muscle strength signal is lower than that of the sEMG signal, the common interpolation processing method only sets the number of the fixed data inserted into every two adjacent muscle strength signals in time sequence according to the difference of the sampling points of the muscle strength signal and the sEMG signal, the fluctuation condition of muscle strength signals when muscles are in different states is not considered, meanwhile, the inserted numerical value is set according to the numerical value of the muscle strength signals at the two ends of the inserted interval, the influence of the muscle electrical signals is not considered, when the measurement time is continuously increased, abnormal fluctuation of sEMG signals occurs when the muscles are in a fatigue state, at the moment, larger errors exist in the inserted numerical value only according to the values of the front value and the back value on the time sequence of the muscle strength signals, and the accuracy of the muscle strength prediction result is further influenced.
The foregoing is provided merely for the purpose of facilitating understanding of the technical solutions of the present application and is not intended to represent an admission that the foregoing is prior art.
Disclosure of Invention
The application mainly aims to provide a muscle strength recovery prediction method, a system and equipment, and aims to solve the technical problem that the prior art that the inserted numerical value is set only according to the magnitude of the front and back values on the time sequence of a muscle strength signal when the muscle strength signal is interpolated so as to match the muscle strength signal with an electromyographic signal does not consider the influence of fluctuation of the electromyographic signal.
The application provides a muscle strength recovery prediction method, which comprises the steps of collecting sample muscle strength data and sample sEMG signals, obtaining total to-be-interpolated numbers in the sample muscle strength data according to sampling frequency of the sample muscle strength data and sampling frequency of the sample sEMG signals, distributing the total to-be-interpolated numbers according to numerical fluctuation conditions of the sample muscle strength data to obtain actual to-be-interpolated numbers of all to-be-interpolated sections, dividing all to-be-interpolated sections based on the actual to-be-interpolated numbers to obtain a plurality of to-be-interpolated points of all to-be-interpolated sections, obtaining predicted values of all to-be-interpolated points in all to-be-interpolated sections, constructing a regulating factor of each to-be-interpolated point, correcting the predicted value of each to-be-interpolated point by utilizing the regulating factor and the sample sEMG signals to obtain corrected sample muscle strength data, building a muscle strength prediction model based on the corrected sample muscle strength data and the sample sEMG signals, collecting myoelectric signals at the to-be-interpolated sections, and inputting myoelectric signals at the to-be-interpolated points to-be-predicted to-be-interpolated points into the muscle strength prediction model of a human to be-tested.
Optionally, the total to-be-interpolated number is distributed according to the numerical fluctuation condition of the sample muscle force data to obtain the actual to-be-interpolated number of each to-be-interpolated interval, which comprises the steps of determining the numerical fluctuation degree of each to-be-interpolated interval and determining the actual to-be-interpolated number of each to-be-interpolated interval according to the numerical difference of the numerical fluctuation degree of each to-be-interpolated interval.
Optionally, the determining the numerical fluctuation degree of each interval to be interpolated includes determining the numerical fluctuation degree of each interval to be interpolated using the following formula (1):
In the formula, Represent the firstThe degree of numerical fluctuation of the individual intervals to be interpolated,Represent the firstThe first muscle force data on the time sequence of each interval to be interpolated,Represent the firstThe last muscle force data on the time sequence of each interval to be interpolated,The numerical fluctuation degree set of all the intervals to be interpolated in the sample muscle force data is represented,The representation is to take the absolute value,Representing the maximum value in the set of values,Representing the minimum in the set.
Optionally, determining the actual number of to-be-interpolated values of each to-be-interpolated interval according to the numerical difference of the numerical fluctuation degree of the to-be-interpolated interval includes determining the actual number of to-be-interpolated values of each to-be-interpolated interval according to the numerical difference of the numerical fluctuation degree of the to-be-interpolated interval by using the following formula (2):
In the formula, Represent the firstThe actual number of to-be-interpolated values for each to-be-interpolated interval,Represent the firstThe degree of numerical fluctuation of the individual intervals to be interpolated,The number of intervals to be interpolated is indicated,And representing the total number of to-be-interpolated values.
Optionally, the construction of the adjustment factor of each point to be interpolated comprises determining a myoelectricity interval corresponding to each interval to be interpolated, determining trend difference degrees of each interval to be interpolated and the corresponding myoelectricity interval, and constructing the adjustment factor based on the trend difference degrees of each interval to be interpolated and sampling time of each point to be interpolated in each interval to be interpolated.
Optionally, the correcting the predicted value of each point to be interpolated by using the adjustment factor and the sample sEMG signal to obtain corrected sample muscle force data includes correcting the predicted value of each point to be interpolated by using the adjustment factor and the sample sEMG signal by using the following formula (3) to obtain corrected sample muscle force data:
In the formula, Represent the firstThe first interval to be interpolatedPredicted values of the points to be interpolated after correction,Represent the firstThe first interval to be interpolatedThe predicted values of the individual points to be interpolated,Represent the firstThe first interval to be interpolatedThe adjustment factors of the individual points to be interpolated,Represent the firstThe first interval to be interpolatedThe sampling time of the individual points to be interpolated,Represent the firstThe sampling time of the first bit of data in the time sequence in the interval to be interpolated,Represent the firstThe sampling time of the last bit of data in the time sequence in the interval to be interpolated,Representing the first of the sample muscle force dataThe first interval to be interpolatedThe euclidean distance of the corresponding signal on the individual points to be interpolated and the sample sEMG signal,Representing the first of the sample muscle force dataThe first bit of data in the time sequence in the interval to be interpolated and the euclidean distance of the corresponding signal on the sample sEMG signal,Representing the first of the sample muscle force dataThe first bit of data in time sequence in each interval to be interpolated and the Euclidean distance of the corresponding signal on the sample sEMG signal.
Optionally, after the predicted muscle strength value of the tested person is obtained, the predicted muscle strength recovery method further comprises the step of judging the predicted muscle strength value of the tested person by using a muscle strength evaluation method to obtain the predicted muscle strength recovery condition of the tested person.
In addition, in order to achieve the above object, the present application also provides a muscle strength recovery prediction system, including:
The system comprises a data acquisition module, a section dividing module, a model construction module and a muscle strength prediction module, wherein the data acquisition module is used for acquiring sample muscle strength data and sample sEMG signals, acquiring total to-be-interpolated numbers in the sample muscle strength data according to sampling frequency of the sample muscle strength data and sampling frequency of the sample sEMG signals, the section dividing module is used for distributing the total to-be-interpolated numbers according to numerical fluctuation conditions of the sample muscle strength data to obtain actual to-be-interpolated numbers of each to-be-interpolated section, dividing each to-be-interpolated section based on the actual to-be-interpolated numbers to obtain a plurality of to-be-interpolated points of each to-be-interpolated section, the numerical correction module is used for acquiring a predicted value of each to-be-interpolated point in each to-be-interpolated section, constructing an adjusting factor of each to-be-interpolated point, correcting the predicted value of each to-be-interpolated point by utilizing the adjusting factor and the sample sEMG signals to obtain corrected sample muscle strength data, and the model construction module is used for establishing a muscle strength prediction model based on the corrected sample muscle strength data and the sample sEMG signals, and inputting the predicted muscle strength prediction model to the predicted muscle strength of a human to be tested to obtain the predicted muscle strength.
The application also provides a muscle strength recovery prediction device which comprises at least one processor and a memory which is in communication connection with the at least one processor, wherein the memory stores instructions which can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the muscle strength recovery prediction method.
According to the muscle strength recovery prediction method, system and equipment provided by the application, the initial number of to-be-interpolated values is adjusted by analyzing the numerical fluctuation condition of the muscle strength data to obtain the actual number of to-be-interpolated values in each to-be-interpolated region, so that the problem that the interpolation precision is low due to the fact that fixed numbers of data are inserted into adjacent muscle strength signals in pairs in time sequence only according to the sampling frequency difference setting of the muscle strength signals and sEMG signals is solved, the problem that the accuracy is low when the subsequent muscle strength prediction model is constructed is affected, the different sampling precision is set when the muscle is in different states or the state is changed, and the interpolation precision is improved.
In addition, the predicted value of each interpolation point in each interval to be interpolated is obtained, an adjustment factor of each interpolation point is constructed, the predicted value of each interpolation point is corrected by utilizing the adjustment factor and a sample sEMG signal, the problem that the conventional interpolation processing method is poor in interpolation precision due to the fact that the relationship between the sEMG signal and the muscle strength signal is not considered only according to the numerical value of the muscle strength signal, and the influence of the fatigue state of muscles is easy to occur when the human body is in a continuous measurement state is also not considered at the same time is solved, the predicted value to be interpolated of the muscle strength signal is adjusted based on the relationship between the sEMG signal and the muscle strength signal, and meanwhile, the predicted value is further adjusted according to the characteristic of the fatigue state, so that an accurate predicted result can be ensured in the subsequent muscle strength recovery prediction process.
Drawings
FIG. 1 is a flowchart of a method for predicting muscle strength recovery according to one embodiment of the present application;
FIG. 2 is a block diagram of a muscle strength recovery prediction system according to one embodiment of the present application;
fig. 3 is a schematic structural view of a muscle strength recovery prediction apparatus according to an embodiment of the present application.
The achievement of the objects, functional features and advantages of the present application will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.
The application provides a solution, which sets different sampling precision when muscles are in different states or the states are changed, improves the interpolation precision, adjusts the predicted value to be interpolated of the muscle force signals based on the relation between the sEMG signals and the muscle force signals and only obtained based on the muscle force signals, further adjusts the predicted value according to the characteristics of the fatigue state, and can ensure that more accurate predicted results are obtained in the subsequent muscle force recovery prediction process.
The following describes the scheme of the application in detail.
Fig. 1 is a flowchart of a muscle strength recovery prediction method according to an embodiment of the present application, which may be performed by a muscle strength recovery prediction apparatus having data processing capability, with reference to fig. 1, and may include:
step S10, collecting sample muscle force data and sample sEMG signals, and obtaining the total number to be interpolated in the sample muscle force data according to the sampling frequency of the sample muscle force data and the sampling frequency of the sample sEMG signals.
The muscle force data is mechanical force generated by muscles, the strength generated during muscle contraction can be measured and recorded through a mechanical sensor such as a dynamometer, a constant-speed muscle force tester and the like, and the muscle force data can directly reflect the strength and strength output of the muscle contraction. The sEMG signal is a surface electromyographic signal, which can be non-invasively recorded by placing electrodes on the skin surface, and can represent the voltage change of the electrical activity of the muscle, i.e., the sEMG signal is a series of voltage waveforms that change with time, and can specifically reflect the frequency, amplitude and time domain characteristics of the muscle activity.
In a specific implementation process, sEMG signals and sEMG data of a plurality of patients during muscle strength rehabilitation training of a specified action are collected, and for example, any patient can be recorded as a target subject, the sEMG data and sEMG signals generated during muscle strength rehabilitation training of the specified action of the target subject are taken as sample sEMG data and sample sEMG signals, and the sample sEMG data and the sample sEMG signals are taken as examples for explanation.
For example, the specific steps of collecting sample muscle force data and sample sEMG signals include:
When the target subject performs muscle strength rehabilitation training, the force is continuously applied for 15s according to the specified action, the rest is carried out for 5s after the specified action is finished, the repetition time can be set to be 6, the force sensor can be used for collecting muscle strength data at the target muscle to obtain sample muscle strength data when the target subject performs the specified action or rehabilitation training, meanwhile, the electrode is stuck to the surface of the target muscle of the target subject according to the direction of muscle fiber, good contact between the electrode and skin is ensured, the signal quality can be improved by using conductive paste or gel, and the sEMG signal at the target muscle of the target subject is collected by using the electrode to obtain the sample sEMG signal.
In one embodiment, after the sample muscle force data and the sample sEMG signal of the target subject are acquired, the sample sEMG signal may also be denoised.
The sample sEMG signal has noise, artifacts, baseline drift and other interferences due to the influence of the acquisition environment, the detection instrument and the characteristics of the muscle, and the noise removal processing is needed to be carried out on the sample sEMG signal so as to improve the quality and the reliability of the sample sEMG signal. For example, a wavelet transform denoising algorithm may be used to denoise the sample sEMG signal.
In one embodiment, in step S10, obtaining the total number of to-be-interpolated values in the sample muscle force data according to the sampling frequency of the sample muscle force data and the sampling frequency of the sample sEMG signal may specifically include:
S11, determining the sampling point number of the sample muscle force data based on the sampling frequency of the sample muscle force data, and determining the sampling point number of the sample sEMG signal based on the sampling frequency of the sample sEMG signal;
And S12, marking the difference value between the sampling point number of the sample sEMG signal and the sampling point number of the sample muscle force data as the total to-be-interpolated number of the sample muscle force data.
The sampling point number is equal to the sampling frequency multiplied by the sampling time, and the sampling time is the duration time when the target subject performs muscle strength rehabilitation training.
Step S20, the total to-be-interpolated number is distributed according to the numerical fluctuation condition of the sample muscle force data to obtain the actual to-be-interpolated number of each to-be-interpolated interval, and the to-be-interpolated intervals are divided based on the actual to-be-interpolated number to obtain a plurality of to-be-interpolated points of each to-be-interpolated interval.
In an embodiment, in the step, the adjusting the initial number of to-be-interpolated according to the numerical fluctuation condition of the sample muscle force data to obtain the actual number of to-be-interpolated of each to-be-interpolated interval may specifically include:
s21, determining the numerical fluctuation degree of each interval to be interpolated;
S22, distributing the total number of the to-be-interpolated values according to the numerical difference of the numerical fluctuation degree of the to-be-interpolated values, and determining the actual number of the to-be-interpolated values of each to-be-interpolated value.
Two data adjacent in time sequence in the sample muscle force data are taken as an interval to be interpolated.
Specifically, by the firstFor example, the first interval to be interpolated is determinedThe degree of numerical fluctuation of each interval to be interpolatedThe calculation formula of (2) is as follows:
In the formula, Represent the firstThe first muscle force data on the time sequence of each interval to be interpolated,Represent the firstThe last muscle force data on the time sequence of each interval to be interpolated,The numerical fluctuation degree set of all the intervals to be interpolated in the sample muscle force data is represented,The representation is to take the absolute value,Representing the maximum value in the set of values,Representing the minimum in the set.
When the following is performedWhen 0, adjustThe value of (2) is 0.1. Scaling the numerical fluctuation degree of each interval to be interpolated according to a proportion to enable the sum of the numerical fluctuation degrees of all the intervals to be interpolated to be 1, multiplying the scaled numerical fluctuation degree by the total number to be interpolated to obtain the actual number to be interpolated of each interval to be interpolated, and ensuring that the sum of the actual number to be interpolated of each interval to be interpolated is still the total number to be interpolated.
Specifically, by the firstFor example, the first interval to be interpolated is determinedThe actual number of interpolation intervalsThe calculation formula of (2) is as follows:
In the formula, Represent the firstThe degree of numerical fluctuation of the individual intervals to be interpolated,The number of intervals to be interpolated is indicated,And representing the total number of to-be-interpolated values.
It should be noted that, because the degree of change of the head and tail data of different intervals to be interpolated is different, the actual muscle states corresponding to the different intervals to be interpolated are also different, for example, when the muscle state is changed from a force to a relaxed state or from a relaxed state to a force, the muscle force will have larger numerical fluctuation, and when the muscle is in the same state, that is, when the muscle is in a force or a continuous relaxed state, the change of the muscle force is smaller.
Step S30, obtaining the predicted value of each point to be interpolated in each interval to be interpolated, constructing an adjustment factor of each point to be interpolated, and correcting the predicted value of each point to be interpolated by using the adjustment factor and the sample sEMG signal to obtain corrected sample muscle force data.
In the specific implementation process, the least square method can be utilized to fit the sample muscle force data, so as to obtain the predicted value of each point to be interpolated in each interval to be interpolated.
In one embodiment, in step S30, constructing the adjustment factor of each point to be interpolated specifically may include:
s31, determining myoelectricity intervals corresponding to each interval to be interpolated;
s32, determining trend difference degrees of each interval to be interpolated and the corresponding myoelectricity interval;
S33, constructing an adjustment factor based on the trend difference degree of each interval to be interpolated and the sampling time of each point to be interpolated in each interval to be interpolated.
The myoelectricity interval corresponding to the sampling time of each interval to be interpolated on the sample sEMG signal can be recorded as the myoelectricity interval corresponding to each interval to be interpolated.
In a specific implementation process, firstly, according to the characteristic that a sample sEMG signal is generated before muscle contraction exercise, a time error value is preset, sample muscle force data is subjected to time sequence backward based on the preset time error value, namely, the sampling time corresponding to each numerical value on the sample muscle force data is added with the preset time error value, and then the myoelectricity interval corresponding to the sampling time of the first data and the last data in time sequence of each interval to be interpolated is obtained on the sample sEMG signal.
The method is characterized in that when the muscle is in a resting state, the muscle strength is smaller, the sEMG signal amplitude is smaller, when the muscle is in a continuously stressed state, the muscle strength value is continuously increased, the sEMG signal amplitude is increased and is continuously maintained for a period of time, when the muscle is in a resting state again, the muscle strength value is gradually reduced, the sEMG signal amplitude is reduced to the vicinity of the original point again, and therefore the sample sEMG signal and sample muscle strength data change trend at the same muscle are similar when a subject performs muscle strength rehabilitation training according to a specified action, based on the method, the trend difference degree of the sample sEMG signal and the sample muscle strength data can be firstly obtained, and the value pre-fitted in the sample muscle strength data is adjusted according to the trend difference degree.
In one embodiment, determining the trend difference degree between each interval to be interpolated and the corresponding myoelectric interval in step S32 may specifically include:
And determining the DTW distance between each interval to be interpolated and the corresponding myoelectricity interval by using a DTW algorithm, and recording the determined DTW distance as the trend difference degree between each interval to be interpolated and the corresponding myoelectricity interval.
And carrying out linear normalization processing on the trend difference degree of each interval to be interpolated and the corresponding myoelectricity interval, and carrying out linear normalization processing on the sampling time of the points to be interpolated of all the intervals to be interpolated.
It should be noted that, the linear normalization processing is performed on the values in the set, that is, the values in the set are mapped into the (0, 1) range according to the value size.
In one embodiment, in step S33, the constructing the adjustment factor based on the trend difference degree of each interval to be interpolated and the sampling time of each point to be interpolated in each interval to be interpolated may specifically include:
and recording the ratio of the trend difference degree of each point to be interpolated in each interval to be interpolated to the sampling time as the adjustment factor of each point to be interpolated.
It should be noted that, the trend difference degree of each point to be interpolated in each interval to be interpolated is the trend difference degree of the interval to be interpolated where the point to be interpolated is located, when the trend difference degree of the point to be interpolated is larger, it is indicated that the predicted value obtained by fitting the point to be interpolated has a deviation, but when the muscle is in a fatigue state, the trend difference degree of each point to be interpolated in the interval to be interpolated where the fatigue state is located is relatively larger, so that the adjustment is performed by considering the addition sampling time, and therefore, the influence of the fatigue state of the muscle is considered while the adjustment of the value pre-fitted in the sample muscle force data based on the sample sEMG signal can be ensured.
In one embodiment, in step S30, correcting the predicted value of each point to be interpolated by using the adjustment factor and the sample sEMG signal, the obtaining corrected sample muscle force data may specifically include:
And acquiring the electromyographic signals corresponding to each point to be interpolated on the sample sEMG signal, sequentially acquiring the Euclidean distance between each point to be interpolated and the corresponding electromyographic signals, the Euclidean distance between two endpoints of each interval to be interpolated and the electromyographic signals corresponding to the two endpoints of each interval to be interpolated, and correcting the predicted value of each point to be interpolated according to the acquired Euclidean distance difference and combining the sampling time difference between each point to be interpolated and the two endpoints of the interval to be interpolated and the adjustment factor of each point to be interpolated.
Wherein, the sampling time of each interpolation point and the corresponding electromyographic signal is the same.
It should be noted that, when the adjustment factor of the point to be interpolated is large, the lower the confidence that the predicted value of the point to be interpolated is, the higher the necessity of adjustment is. Meanwhile, as the predicted value of each point to be interpolated is obtained by fitting, if the point to be interpolated is close to the end point of the interval to be interpolated in time sequence, the confidence of the predicted value of the point to be interpolated is higher, otherwise, the confidence is lower. Because the two endpoints of the interval where each point to be interpolated is located are actually acquired muscle force data, the sample muscle force data and the fluctuation trend of the sample sEMG signal have similarity, and the distance between the two endpoints of each interval to be interpolated and the electromyographic signals corresponding to the two endpoints can be used as the predicted value adjustment standard of each point to be interpolated in each interval to be interpolated, and the sampling time difference value of each point to be interpolated and the two endpoints of the interval to be interpolated and the adjustment factor of each point to be interpolated are combined to adjust the predicted value of each point to be interpolated.
Exemplary, in the firstThe first interval to be interpolatedTo-be-interpolated pointsFor example, record the firstFirst-bit muscle strength data in time sequence in each interval to be interpolatedRecord the first point as the target starting pointLast bit of data in time sequence in each interval to be interpolatedFor the end point of the target, obtainAnd sample sEMG signalEuclidean distance of electromyographic signals with same sampling time(I.e. isAnd Euclidean distance of corresponding signal on sample sEMG signal), obtainingAnd sample sEMG signalEuclidean distance of electromyographic signals with same sampling time(I.e. isAnd Euclidean distance of corresponding signal on sample sEMG signal), obtainingAnd sEMG signal up and downEuclidean distance of electromyographic signals with same sampling time(I.e. isAnd the Euclidean distance of the corresponding signal on the sample sEMG signal), the following formula can be used for the firstThe first interval to be interpolatedTo-be-interpolated pointsIs corrected by the predicted value of (a):
wherein, Represent the firstThe first interval to be interpolatedPredicted values of the points to be interpolated after correction,Represent the firstThe first interval to be interpolatedThe predicted values of the individual points to be interpolated,Represent the firstThe first interval to be interpolatedThe adjustment factors of the individual points to be interpolated,Represent the firstThe first interval to be interpolatedThe sampling time of the individual points to be interpolated,Represent the firstThe sampling time of the first bit of data in the time sequence in the interval to be interpolated,Represent the firstSampling time of last bit of data in time sequence in each interval to be interpolated.
The smaller the DTW distance, the higher the degree of similarity between the two sequences. Firstly, correcting a predicted value of each point to be interpolated based on the trend difference degree of each interval to be interpolated, when the trend difference degree of the interval to be interpolated is smaller, the change trend of the muscle force signal and the electromyographic signal in the interval to be interpolated is indicated to be similar, and then the confidence coefficient of each predicted value to be interpolated in the interval to be interpolated is higher, otherwise, when the trend difference degree of the interval to be interpolated is larger, the change trend of the muscle force signal and the electromyographic signal in the interval to be interpolated is indicated to have difference, and then the confidence coefficient of each predicted value to be interpolated in the interval to be interpolated is lower, and accordingly, the predicted value needs to be adjusted by using a sample sEMG signal. However, when the measurement time is continuously increased and the muscle is in a fatigue state, abnormal fluctuation of the sEMG signal appears, and the abnormal fluctuation is particularly shown as that in the initial stage of fatigue, in order to maintain the strength output, the muscle recruits more movement units, so that the amplitude of the sEMG signal is increased, and as the fatigue further progresses, the amplitude of the sEMG signal may become larger due to the decrease of the muscle control capability, so that when the muscle is in the fatigue state, the trend difference degree of the interval to be interpolated is also larger.
And step S40, establishing a muscle strength prediction model based on the corrected sample muscle strength data and the sample sEMG signal.
In a specific embodiment, firstly, the corrected sample muscle force data and the corrected sample sEMG signal are processed according to the processing manner of the sEMG signals of the samples acquired in the step S10, and then the sEMG signals and the sEMG data during the muscle force rehabilitation training of the predetermined actions are processed, and secondly, feature extraction is performed on the sEMG signals and the sEMG data, and appropriate features are selected by using pearson correlation coefficients. Finally, a Support Vector Machine (SVM) is used for model training, and a muscle strength prediction model is obtained. In this exemplary embodiment, the kernel function selected by the support vector machine may be a Radial Basis Function (RBF).
The input of the muscle strength prediction model is the muscle electrical signal of the muscle strength prediction position required by the tested person, and the input is the muscle strength prediction value of the tested person.
And S50, acquiring myoelectric signals of the predicted muscle strength of the tested person, and inputting the myoelectric signals of the predicted muscle strength of the tested person into the muscle strength prediction model to obtain the muscle strength prediction value of the tested person.
In the specific implementation process, after the muscle strength prediction model is built, the muscle electrical signals of the muscle strength of the testee are input into the muscle strength prediction model by utilizing the muscle electrical signals of the muscle strength of the testee, the muscle strength prediction model outputs the muscle strength prediction value of the testee, and then the muscle strength recovery condition is judged according to a muscle strength evaluation method commonly used in rehabilitation medicine.
According to the muscle strength recovery prediction method provided by the application, the initial number of to-be-interpolated is adjusted by analyzing the numerical fluctuation condition of the muscle strength data to obtain the actual number of to-be-interpolated in each to-be-interpolated interval, so that the problem that the interpolation accuracy is low and the constructed muscle strength prediction model is low due to the fact that fixed numbers of data are inserted into adjacent muscle strength signals in a time sequence only according to the sampling frequency difference setting of the muscle strength signals and the sEMG signals is solved, the adjustment factor of each to-be-interpolated point is constructed by acquiring the predicted value of each to-be-interpolated point in each to-be-interpolated interval, and the predicted value of each to-be-interpolated point is corrected by utilizing the adjustment factor and the sample sEMG signals, so that the problem that the conventional interpolation processing method is poor in interpolation accuracy due to the fact that the influence of fatigue state is easy to occur to muscles when the human body is in a continuous measurement state is not considered at the same time.
On the basis of the above examples, fig. 2 is a block diagram of a muscular strength recovery prediction system according to an embodiment of the present application, and as shown in fig. 2, the muscular strength recovery prediction system may include a data acquisition module 210, a section division module 220, a numerical correction module 230, a model construction module 240, and a muscular strength prediction module 250, wherein,
The data acquisition module 210 is configured to acquire sample muscle force data and sample sEMG signals, and acquire total to-be-interpolated numbers in the sample muscle force data according to sampling frequencies of the sample muscle force data and the sample sEMG signals;
The interval dividing module 220 is configured to allocate the total number of to-be-interpolated according to the numerical fluctuation condition of the sample muscle force data, obtain an actual number of to-be-interpolated for each to-be-interpolated interval, and divide each to-be-interpolated interval based on the actual number of to-be-interpolated to obtain a plurality of to-be-interpolated points for each to-be-interpolated interval;
The numerical value correction module 230 is configured to obtain a predicted value of each point to be interpolated in each interval to be interpolated, construct an adjustment factor of each point to be interpolated, and correct the predicted value of each point to be interpolated by using the adjustment factor and the sample sEMG signal to obtain corrected sample muscle force data;
The model construction module 240 is configured to establish a muscle strength prediction model based on the corrected sample muscle strength data and the sample sEMG signal;
the muscle strength prediction module 250 is configured to collect a muscle electrical signal at a predicted muscle strength required by the subject, and input the muscle electrical signal at the predicted muscle strength required by the subject into the muscle strength prediction model to obtain a muscle strength prediction value of the subject.
In an exemplary embodiment, the interval dividing module 220 may be further configured to determine a value fluctuation degree of each interval to be interpolated, and determine an actual number of intervals to be interpolated according to a value difference of the value fluctuation degrees of the intervals to be interpolated.
In an exemplary embodiment, the interval division module 220 may determine the numerical fluctuation degree of each interval to be interpolated using the following formula (1):
In the formula (1), the components are as follows, Represent the firstThe degree of numerical fluctuation of the individual intervals to be interpolated,Represent the firstThe first muscle force data on the time sequence of each interval to be interpolated,Represent the firstThe last muscle force data on the time sequence of each interval to be interpolated,The numerical fluctuation degree set of all the intervals to be interpolated in the sample muscle force data is represented,The representation is to take the absolute value,Representing the maximum value in the set of values,Representing the minimum in the set.
In an exemplary embodiment, the interval dividing module 220 may determine the actual number of to be interpolated for each to be interpolated interval using the following formula (2):
In the formula, Represent the firstThe actual number of to-be-interpolated values for each to-be-interpolated interval,Represent the firstThe degree of numerical fluctuation of the individual intervals to be interpolated,The number of intervals to be interpolated is indicated,And representing the total number of to-be-interpolated values.
In an exemplary embodiment, the numerical correction module 230 may be further configured to determine a myoelectric interval corresponding to each interval to be interpolated, determine a trend difference degree between each interval to be interpolated and the corresponding myoelectric interval, and construct an adjustment factor based on the trend difference degree between each interval to be interpolated and a sampling time of each point to be interpolated in each interval to be interpolated.
In an exemplary embodiment, the numerical correction module 230 may be further configured to correct the predicted value of each point to be interpolated using the adjustment factor and the sample sEMG signal by using the following formula (3), to obtain corrected sample muscle force data:
In the formula, Represent the firstThe first interval to be interpolatedPredicted values of the points to be interpolated after correction,Represent the firstThe first interval to be interpolatedThe predicted values of the individual points to be interpolated,Represent the firstThe first interval to be interpolatedThe adjustment factors of the individual points to be interpolated,Represent the firstThe first interval to be interpolatedThe sampling time of the individual points to be interpolated,Represent the firstThe sampling time of the first bit of data in the time sequence in the interval to be interpolated,Represent the firstThe sampling time of the last bit of data in the time sequence in the interval to be interpolated,Represent the firstThe first interval to be interpolatedThe point to be interpolated and the sample sEMG signal are connected with the firstThe first interval to be interpolatedThe euclidean distance of the electromyographic signals with the same sampling time of the points to be interpolated,Represent the firstTime sequence first bit data and sample sEMG signal in each interval to be interpolated andThe euclidean distance of the electromyographic signals with the same sampling time of the first bit data in the time sequence in the interval to be interpolated,Represent the firstTime sequence first bit data and sample sEMG signal in each interval to be interpolated andThe Euclidean distance of the electromyographic signals with the same sampling time of the last bit of data in the time sequence in each interval to be interpolated.
In an exemplary embodiment, the muscular strength recovery prediction system may further include an acquisition module for determining a muscular strength prediction value of the subject using a muscular strength evaluation method to obtain a muscular strength prediction recovery condition of the subject.
It should be understood by those skilled in the art that the division of each module in the embodiment is merely a division of a logic function, and may be fully or partially integrated on one or more actual carriers in practical application, and the modules may be fully implemented in a form of calling by a processing unit through software, may be fully implemented in a form of hardware, or may be implemented in a form of combining software and hardware, and it should be noted that each module in a muscle strength recovery prediction system in this embodiment is in one-to-one correspondence with each step in a muscle strength recovery prediction method in the foregoing embodiment, so that a specific implementation of this embodiment may refer to an implementation of a muscle strength recovery prediction method in the foregoing embodiment, and will not be described herein.
Based on the above example, fig. 3 is a schematic structural diagram of a muscle strength recovery prediction device according to an embodiment of the present application, and as shown in fig. 3, the electronic device may include a processor 310, a communication interface (Communications Interface), a memory 330 and a communication bus 340, where the processor 310, the communication interface 320 and the memory 330 complete communication with each other through the communication bus 340. The processor 310 may call a logic instruction in the memory 330 to execute a muscle strength recovery prediction method, which includes collecting sample muscle strength data and sample sEMG signals, obtaining initial numbers of to-be-interpolated in each section according to sampling frequency of the sample muscle strength data and sampling frequency of the sample sEMG signals, adjusting the initial numbers of to-be-interpolated according to numerical fluctuation of the sample muscle strength data to obtain actual numbers of to-be-interpolated in each section, dividing each section to be-interpolated based on the actual numbers of to-be-interpolated to obtain a plurality of to-be-interpolated points in each section to be-interpolated, obtaining a predicted value of each to-be-interpolated point in each section to construct an adjustment factor of each to-be-interpolated point, correcting the predicted value of each to-be-interpolated point by using the adjustment factor and the sample sEMG signals to obtain corrected sample muscle strength data, establishing a muscle strength prediction model based on the corrected sample muscle strength data and the sample sEMG signals, collecting electrical signals required by a tested person to obtain the electrical signals required by the tested person, and inputting the tested person to the predicted muscle strength required by the tested person into the tested person.
Further, the logic instructions in the memory 330 described above may be implemented in the form of software functional units and may be stored in a computer-readable storage medium when sold or used as a stand-alone product. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. The storage medium includes a U disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, an optical disk, or other various media capable of storing program codes.
On the basis of the above embodiment, in another aspect, the present invention further provides a computer program product, where the computer program product includes a computer program, where the computer program can be stored on a non-transitory computer readable storage medium, and when the computer program is executed by a processor, the computer program can execute the method for predicting muscle strength recovery provided by the above methods, where the method includes collecting sample muscle strength data and sample sEMG signals, and obtaining initial numbers to be interpolated for each interval to be interpolated in the sample muscle strength data according to a sampling frequency of the sample muscle strength data and a sampling frequency of the sample sEMG signals; the method comprises the steps of adjusting the initial interpolation number according to the numerical fluctuation condition of sample muscle force data to obtain the actual interpolation number of each interpolation interval, dividing each interpolation interval based on the actual interpolation number to obtain a plurality of interpolation points of each interpolation interval, obtaining the predicted value of each interpolation point in each interpolation interval, constructing an adjustment factor of each interpolation point, correcting the predicted value of each interpolation point by utilizing the adjustment factor and a sample sEMG signal to obtain corrected sample muscle force data, establishing a muscle force prediction model based on the corrected sample muscle force data and the sample sEMG signal, collecting the myoelectric signals of the muscle force of a tested person, and inputting the myoelectric signals of the muscle force of the tested person into the muscle force prediction model to obtain the muscle force prediction value of the tested person.
On the basis of the above embodiment, in yet another aspect, the present invention further provides a non-transitory computer readable storage medium, on which a computer program is stored, which when executed by a processor, is implemented to perform the method for predicting muscle strength recovery provided by the above methods, the method comprising collecting sample muscle strength data and sample sEMG signals, and obtaining initial numbers of to-be-interpolated intervals in the sample muscle strength data according to a sampling frequency of the sample muscle strength data and a sampling frequency of the sample sEMG signals; the method comprises the steps of adjusting the initial interpolation number according to the numerical fluctuation condition of sample muscle force data to obtain the actual interpolation number of each interpolation interval, dividing each interpolation interval based on the actual interpolation number to obtain a plurality of interpolation points of each interpolation interval, obtaining the predicted value of each interpolation point in each interpolation interval, constructing an adjustment factor of each interpolation point, correcting the predicted value of each interpolation point by utilizing the adjustment factor and a sample sEMG signal to obtain corrected sample muscle force data, establishing a muscle force prediction model based on the corrected sample muscle force data and the sample sEMG signal, collecting the myoelectric signals of the muscle force of a tested person, and inputting the myoelectric signals of the muscle force of the tested person into the muscle force prediction model to obtain the muscle force prediction value of the tested person.
Claims (6)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411321832.3A CN118830850B (en) | 2024-09-23 | 2024-09-23 | Muscle strength recovery prediction method, system and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411321832.3A CN118830850B (en) | 2024-09-23 | 2024-09-23 | Muscle strength recovery prediction method, system and equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118830850A CN118830850A (en) | 2024-10-25 |
| CN118830850B true CN118830850B (en) | 2024-12-06 |
Family
ID=93139082
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202411321832.3A Active CN118830850B (en) | 2024-09-23 | 2024-09-23 | Muscle strength recovery prediction method, system and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118830850B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113017587A (en) * | 2021-03-08 | 2021-06-25 | 自贡市第一人民医院 | High-precision wrist sphygmomanometer |
| CN118383769A (en) * | 2024-03-29 | 2024-07-26 | 华中科技大学 | A sensing system for measuring the work done by human muscles |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9131888B2 (en) * | 2010-09-21 | 2015-09-15 | Alexander B. Grey | Metrics and algorithms for interpretation of muscular use |
| EP3714782A1 (en) * | 2019-03-27 | 2020-09-30 | Koninklijke Philips N.V. | Assessing muscle fatigue |
| CN211484526U (en) * | 2019-12-17 | 2020-09-15 | 郑州大学第二附属医院 | Neurological sensory examination device |
| US20230337989A1 (en) * | 2021-05-12 | 2023-10-26 | Shenzhen Shokz Co., Ltd. | Motion data display method and system |
| CN114947891A (en) * | 2022-03-02 | 2022-08-30 | 北京航空航天大学 | A method for detecting a user's muscle rehabilitation by an arm ring device |
| CN116312951B (en) * | 2023-04-27 | 2023-08-04 | 华南理工大学 | A method and system for evaluating motor function based on multimodal coupling analysis |
| CN116869535A (en) * | 2023-07-19 | 2023-10-13 | 中国科学技术大学 | A method for predicting myoelectric strength suitable for hand function rehabilitation training in children with cerebral palsy |
-
2024
- 2024-09-23 CN CN202411321832.3A patent/CN118830850B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113017587A (en) * | 2021-03-08 | 2021-06-25 | 自贡市第一人民医院 | High-precision wrist sphygmomanometer |
| CN118383769A (en) * | 2024-03-29 | 2024-07-26 | 华中科技大学 | A sensing system for measuring the work done by human muscles |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118830850A (en) | 2024-10-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP5464072B2 (en) | Muscle activity diagnosis apparatus and method, and program | |
| US20010041846A1 (en) | System and method for determining muscle dysfunction | |
| KR101910982B1 (en) | Method and apparatus for eliminating motion artifact of biosignal using personalized biosignal pattern | |
| CN111329476B (en) | A method and device for muscle strength estimation based on microscopic neural drive information | |
| US10349896B2 (en) | Epsilon-tube filter for blunt noise removal | |
| US20130237868A1 (en) | Apparatus and method of controlling threshold for detecting peaks of physiological signals | |
| CN111956203B (en) | Electrocardiosignal parameterization method, model training method, device, equipment and medium | |
| CN118044826B (en) | Time-frequency analysis-based method for aligning across-test electroencephalogram signals and related equipment | |
| TWI598073B (en) | Physiological signal measurement method and physiological signal measurement device | |
| CN115317002B (en) | Method and apparatus for determining electroencephalographic microstates | |
| CN118830850B (en) | Muscle strength recovery prediction method, system and equipment | |
| JP2019030520A (en) | Myoelectricity measurement apparatus, method, and program | |
| Kranjec et al. | Improved assessment of muscle excitation from surface electromyograms in isometric muscle contractions | |
| Lin et al. | Design and implementation of wireless multi-channel EEG recording system and study of EEG clustering method | |
| JP5609985B2 (en) | Arousal level determination device, arousal level determination method, and arousal level determination program | |
| KR20130103259A (en) | Apparatus and method for controlling threshold for detecting peaks of physiological signals | |
| CN114676737B (en) | Dynamic regulation method for sampling frequency of electromyographic signal | |
| CN111588373B (en) | Data processing method and device, computer storage medium and electronic equipment | |
| JP2015217060A (en) | Heart rate detection method and heart rate detection device | |
| CN112773379B (en) | Myoelectric signal processing method, processing equipment and storage medium | |
| CN118576216B (en) | Artifact suppression capability evaluation method, device, electronic device and storage medium | |
| CN119279588B (en) | Emotion state detection system based on electrophysiological data | |
| CN119848472B (en) | Electroencephalogram signal processing system for sleep analysis | |
| CN117257303B (en) | Anxiety detection method and device, electronic equipment and storage medium | |
| JP7549348B2 (en) | DATA PROCESSING APPARATUS, DATA PROCESSING METHOD, AND PROGRAM |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |