Method and system for analyzing biomechanical activity and exposure to biomechanical risk factors for a subject in the context of physical activity - Patents.com
The integration of MMG sensors and IMU units with data processing and wireless communication addresses the limitations of subjective ergonomics assessments by providing real-time, objective analysis of biomechanical stress and predicting muscle fatigue to prevent injuries.
Patent Information
- Application Number
- JP2022509651
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Priority Date
- 2019-08-14
- Filing Date
- 2020-08-13
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2040-08-13
AI Technical Summary
Existing methods for evaluating biomechanical activity and exposure to risk factors in physical ergonomics are subjective, time-limited, and lack objective quantification of muscle activity, particularly in uncontrolled environments, leading to inaccurate assessments of muscle fatigue and potential long-term health risks.
A system combining MMG sensors and IMU units with data processing and wireless communication to objectively measure muscle activity and movement, integrating these signals to predict physiological rest times and automate ergonomic assessments.
Provides real-time, objective analysis of biomechanical stress, enabling accurate prediction of muscle fatigue and recommending physiological breaks to prevent injuries, suitable for uncontrolled environments and applications like exoskeleton validation and sports training.
Smart Images

Figure 0007588635000009 
Figure 0007588635000010 
Figure 0007588635000011
Abstract
Description
[Technical field]
[0001] The present invention relates to a method for analyzing a subject's biomechanical activity and exposure to biomechanical risk factors in relation to physical activity, and to a system for implementing this method. [Background technology]
[0002] The field of the invention is particularly concerned with physical ergonomics and the ergonomic evaluation of workstations and physical assistive devices (exoskeletons, cobots, robots). Summary of the Invention
[0003] In physical ergonomics, the methods used are essentially based on analysing the movements and postures of the worker. A preventive specialist or ergonomist observes the situation and completes an assessment grid according to criteria specific to each company. Some are based on an older method called RULA (Right Upper Limb Assessment), widely used for the lack of better ones, in which the observer evaluates the worker's posture at the critical moments of the task by assessing the main joint angles mobilised.
[0004] It is generally accepted in industry that some personnel spend almost all their time completing this type of grid. Another fundamental limitation of this type of grid is the subjective criteria for obtaining a strength score, which can then move the overall score from an acceptable level to a critical level by changing a single evaluation parameter. Therefore, to obtain a reliable evaluation, the evaluation must be performed by a specialized, knowledgeable eye. Problems then appear related to the need to accurately and objectively quantify the worker's muscular activity in order to make the ergonomic analysis reliable.
[0005] Furthermore, the grid is filled in a limited,short time, making it impossible to measure the effect of fatigue on the,posture of a worker who can develop different compensations,not discernible by the naked eye.,Today, there are no tools to monitor biomechanical risk factors of,workers in real situations during the entire work activity.
[0006] A final point concerns the arrival of exoskeletons and other robotic devices for physical assistance in the new Industry 4.0, where validation from an ergonomic point of view is a real challenge: in fact, exoskeletons can adopt postures that are acceptable from the point of view of the ruler's law, while overloading certain muscle groups, which may then cause accidents or occupational diseases in the long term.
[0007] Furthermore, monitoring the muscle activity of the human body is an important function in many applications related to health, sports and robotics. Characterizing the muscle state (e.g. fatigue) and the evolution of this state according to whole-body movements and evolving the whole-body movements according to this state may provide valuable information on the muscle state, e.g. for optimizing the training of athletes or adjusting the physical load of industrial workers. Furthermore, recognizing the signs of muscle fatigue can help prevent the risk of injuries related to physical activity in sports or industrial situations.
[0008] Muscle activity is commonly captured by electromyography (EMG) [1], i.e., measuring the electrical activity of muscles. EMG techniques can be invasive, where electrodes are inserted directly into the muscle body, or superficial, where surface electrodes are attached to the skin. The latter techniques have many drawbacks, limiting their use to controlled environments such as laboratories. Disposable electrodes must be used for hygiene reasons, and conductive gel must be used to ensure good coupling with the skin and to reduce impedance changes during sweating. In addition, positioning and use of the electrodes requires technical know-how. This places severe limitations on the practicality of such out-of-laboratory measurements.
[0009] A method called mechanomyography (MMG), which involves the measurement of minute vibrations induced during muscle contraction, overcomes many of the limitations imposed by EMG, such as less impedance changes due to contact with the skin as it is not affected by sweating, a better signal-to-noise ratio (SNR) and less sensitivity to the location of the muscle being analyzed. The concept of "listening to the sounds" produced by muscles on the skin's surface dates back to the early 1800s [2]. This method is characterized by the capture and interpretation of mechanical vibration activity during muscle contraction.
[0010] The oscillatory activity is produced by transverse vibrations of muscle fibers occurring at the muscle's resonant frequency [3]. Low-frequency (2-250 Hz) MMG signals are acquired by accelerometers, microphones, piezoelectric sensors or laser means. Scientific studies show that the analysis of MMG signals makes it possible to investigate many features of muscle function, such as neuromuscular fatigue [4], the efficiency of anesthesia [5] or specific neuromuscular syndromes, such as Parkinson's disease [6].
[0011] Regarding terminology, there are different names to describe mechanomyogram [7]: acoustic or acoustic mechanomyogram (AMG, PMG) and vibromyogram (VMG). Usually, mechanomyogram uses pressure sensors, microphones or piezoelectric transducers, while vibromyogram uses mostly accelerometers. The preferred device of the present invention uses MEMS capacitance accelerometers, but can be extended to other sensors mentioned.
[0012] The instruments used to capture movements and postures generally use inertial units, consisting of a three-axis accelerometer, a three-axis gyrometer, and a three-axis magnetometer. Common designations of inertial units are acronyms derived from English: Inertial Reference System (IRS), Inertial Navigation System (INS), or Inertial Measurement Unit (IMU). This last designation applies only to the sensor block (accelerometer, gyrometer, magnetometer) without a computational unit. These systems are found in a large number of connected objects, especially due to MEMS technology, which has allowed the miniaturization of the components. Furthermore, developments in battery and wireless communication technologies have made these systems extremely compact with a great deal of autonomy. However, the use of inertial units for biomechanical analysis outside the laboratory faces a major obstacle, as the magnetic field generates distortions and causes the magnetometer measurements to drift [8]. This drift is reflected in the movement data, which then becomes unusable. Another drawback of this type of system is the prior calibration of the sensors to align them with the part of the body to be analyzed.
[0013] The instruments used to capture the movements and poses may include markerless optical systems, in particular with the emergence of so-called depth cameras. They overcome the ambiguities inherent in monocular systems (ambiguities due to partial, automatic occultation and planar projection) by providing a depth image directly, making it possible to estimate the pose of the person. Another advantage of this type of camera is that it provides 3D information of the scene from a single viewpoint [9]. Depth-limited cameras have a range of less than 5 m. This limitation can be overcome by using several cameras, but this requires limited calibration or a very controlled environment.
[0014] Therefore, understanding muscle activity correlated with whole-body movements is important in situations with high biomechanical stress outside of controlled environments. Here, we review examples of musculoskeletal disorders (MSDs) in industry. Indeed, epidemiological studies have shown that MSDs result from exposure to the following biomechanical factors: - Posture constraints - Mobilized effort and dynamic force -Static muscle work -Repeated movements over a long period of activity.
[0015] These biomechanical stresses are assessed using three conditions: the intensity of the stress, the frequency of exposure to these stresses, and the duration of exposure. Furthermore, certain environmental factors can exacerbate these biomechanical factors, namely mechanical pressure, shocks and impacts, vibration, and thermal environments.
[0016] The techniques used in ergonomic studies to analyze the biomechanical activity of subjects and their exposure to the above risk factors are mainly based on expert (ergonomist) observation and photo or video analysis. Moreover, due to the subjective nature of such analysis, it is currently not possible to track subjects in real time over very long periods of time. Recently, we have observed the introduction of motion capture instruments such as inertial units or depth cameras attached to the body parts of the subjects being analyzed. However, these techniques are limited in nature to movements and postures and ignore the measurement of muscle activity induced by any movement of the body.
[0017] The combination of EMG electrodes, MMG sensors and IMU sensors has been the subject of several patents for various applications.
[0018] Thus, document US20130317648A1
[10] relates to a sleeve integrating an array of EMG electrodes and an IMU unit for movement recognition intended to control a mechanical or robotic system.
[0019] Document US20170312576A1
[11] also discusses EMG sensors and IMU units for sports training or therapy.
[0020] Thalmic Labs documents US20150169074
[12] and US20140240103
[13] deal with a connection strip containing EMG electrodes, MMG accelerometers, and an IMU unit for motion recognition and control of connected objects, e.g. connected glasses.
[0021] Document WO2015 / 063520 A1
[14] fuses motion data acquired by an IMU unit with biomechanical data from MMG and EMG sensors for use in patient rehabilitation, posture and gesture classification, and even fetal health monitoring.
[0022] Document US20110196262 A1
[15] provides a comprehensive methodology for processing data from MEMS accelerometers (single or dual axis) to quantify absolute muscle effort in real time. Application areas include the analysis of sports activities and patient rehabilitation.
[0023] Document US 10292647 B1 discloses a wearable device comprising a contraction sensor and a motion sensor, which transmits signals to a processor that analyzes the signals. The contraction signals determine whether the user's muscles are contracting or relaxing. The contraction and movement data, together with a video of the user performing the movements, are transmitted to and displayed on a smart device screen. The simultaneous viewing of the video and the sensor data provides the user with immediate feedback on the timing of the core contraction and body movements in an exercise, training or therapeutic movement, allowing the user to correct and improve the coordination of the core contraction and body movements and improve movement performance to obtain better results.
[0024] However, these systems are unable to enable the analysis of the vibration behavior of a subject's muscles with the aim of producing indicators of muscle fatigue from a portable and autonomous device.
[0025] In particular, new methods and means need to be proposed for the purpose of reliable and accurate evaluation of workstations or physical assistive devices. These methods must be digitized for large-scale deployment and take into account a key parameter that is poorly characterized to date: muscle activity.
[0026] The aim of the present invention is to remedy these shortcomings by proposing a new method for identifying biomechanical risks in uncontrolled environments, based on the objective measurement of a subject's biomechanical factors and the degree and frequency of his exposure to these same factors. It is also an aim of the present invention to automate this method through extremely small instruments and to communicate the biomechanical data wirelessly over long acquisition periods.
[0027] This object is achieved by a system for analyzing the biomechanical activity of a subject and an analysis of exposure to biomechanical risk factors during physical activity,
[0028] a. means for collecting vibration signals attached to one or more first body parts of a subject, the measurements of which reflect local muscle activity; b. means for collecting signals representative of the subject's movements, the measurements reflecting the orientation and movement of one or more second body parts in two or three dimensions; c. means for processing these signals to extract therefrom an indicator representative of the magnitude of biomechanical stress; According to the present invention, the analysis system further comprises: d. means for detecting drift of the vibration signal relative to a reference frame of the vibration behavior of said muscle in the context of physical activity; e. A means for predicting the physiological rest period required for the subject's muscles to regain the reference oscillatory behavior.
[0029] The present invention also relates to a method for analyzing the biomechanical activity of a subject undergoing physical exercise, as well as the exposure to one or more biomechanical risk factors, such as: - Posture constraints - Mobilized effort and dynamic force -Static muscle work -Repeated movements over a long period of activity.
[0030] The analysis method includes using a measurement system focused on one or more body parts of interest in an uncontrolled environment, the measurement system being deployed in an uncontrolled environment; a. collecting vibration signals with vibration sensors attached to one or more body parts of a subject, the measurements reflecting local muscle activity; b. collecting signals representative of the subject's movements, the measurements reflecting the orientation and movement of one or more second body parts in two or three dimensions; c. Processing these signals to extract indicators representative of the intensity of the biomechanical stress, which processing implements a data merging technique.
[0031] According to the present invention, the method further comprises: d. Detecting drift of the vibration signal relative to a frame of reference of the vibration behavior of the muscle(s) in the context of physical activity; and e. predicting the physiological rest time required for the muscles of the subject to regain their normal oscillatory behavior.
[0032] Integrating MMG and IMU data could bring another dimension to the analysis of human activity by building a bridge between dynamic movements and muscle activation. Take the example of patient rehabilitation: the recovery of joint motion is based only on the range of motion the patient generates in a split second. Muscle activation information could provide important information about the recovery process from a physiological and biomechanical point of view. However, as mentioned above, those muscle activation tests are performed only by EMG and in a controlled environment for a very short time.
[0033] The use of the technique of integrating movement and muscle activity data, when applied to physical ergonomics, will lead to the development of algorithms that use IMU data to automatically segment human movements and relate the physiological effects via MMG. It will then be possible to characterize the mechanical efficiency of professional gestures and the mechanisms of loss of this efficiency in relation to fatigue levels.
[0034] In certain embodiments of the present invention, the identification method further comprises collecting by optical collection means a context or scene in which the subject is evolving, and processing this context or scene to generate postural and gestural information of the subject in association with the vibration behavior signal.
[0035] The pose information may be used to construct a classification of the movement in a predetermined frame of reference.
[0036] The identification method of the present invention may also include a cross-analysis of movement and muscle activity signals to provide information regarding the subject's performance and health status during physical activity in the subject of work or sporting activity.
[0037] It may also include fusing the motion data and muscle activity data to recommend a personalized arrangement of physiological rest that returns the subject's musculature to a resting state in metabolic and mechanical sensation after exertion.
[0038] The means for collecting movement signals may advantageously comprise an Inertial Measurement Unit (IMU) using MEMS technology.
[0039] The inertial unit IMU may be integrated together with muscle activity sensor means to obtain co-localised measurements on body parts of the subject.
[0040] The inertial unit IMU may be a 6-axis type measuring instrument with linear acceleration (3 axes), rotation (3 axes). The means for collecting the movement signals may also include a magnetometer (3 axes) that determines the orientation of the body part with respect to the magnetic north of the earth.
[0041] The means for detecting muscle activity may advantageously include a MechanoMyoGraphic (MMG) accelerometer arranged to generate a mechanomyogram signal, ideally a high performance capacitive MEMS accelerometer derived from seismic imaging
[16] .
[0042] The analysis system of the invention may implement a number of measurement nodes rigidly attached to a body part of a subject, each of which includes means of communication with a receiving station implementing a communication protocol of the Bluetooth® Low Energy (BLE) type. [Brief description of the drawings]
[0043] The invention will be better understood with reference to the following drawings: [Figure 1] FIG. 1 illustrates the principle according to the present invention to generate information regarding the degree of exposure to biomechanical risk factors. [Diagram 2] FIG. 2 is a diagram showing the configuration of a measurement node implemented in the method of the present invention. [Diagram 3] FIG. 3 is a block diagram illustrating the operations performed by the method of the present invention. [Figure 4] FIG. 4 is a diagram illustrating the data recovered by the analytical method of the present invention. [Diagram 5]Figure 5 shows another type of result generated by the device, where muscle effort and its drift-over time are measured in real time during a person's activity, and the necessary rest time is calculated to avoid injury and prevent MSD risks (related to Baillargeon
[17] ). [Figure 6] Figure 6 shows how the physiological rest period is calculated from the processed data of the device. The upper curve shows the evolution of the RMS amplitude obtained from the filtered mechanomyogram MMG signal, reflecting the intensity of the biomechanical pressure. The lower curve shows the mean power frequency MPF obtained from the processed mechanomyogram MMG signal in the frequency domain. This curve reflects the muscle fiber recruitment schedule. [Figure 7] 7 shows an application of the invention towards the ergonomic evaluation of work situations and equipment used by industrial operators. In particular, the method is used in accordance with the invention for the ergonomic analysis of an exoskeleton supporting a portable electrical device. [Figure 8] Figure 8 shows representative signals collected from an industrial operator during the activity shown in Figure 7. The upper curve shows the angular amplitude in flexion / extension of the right shoulder and the lower curve shows the muscle vibration of the right biceps brachii associated with the movement produced. [Figure 9] FIG. 9 shows the results of an impact analysis of an exoskeleton supporting a jackhammer on the operator's right shoulder. [Figure 10] 10 shows another application of the invention for the field of sports: a runner is equipped with a measuring system according to the invention on the right vastus muscle. [Figure 11] Figure 11 shows representative signals collected during repetitive strides by a runner. The upper curve shows the right hip angle amplitude in flexion / extension, while the lower curve shows the right vastus muscle vibration associated with the produced movement. [Figure 12] Figure 12 shows the time-frequency of strides in a runner from the raw signal in Figure 11. The x-axis shows time, and each vertical line shows the stride period and its frequency characteristics, i.e., the energy distribution of the signal in the frequency band of interest. [Figure 13] FIG. 13 illustrates an example implementation of a method for obtaining high quality mechanomyogram (MMG) signals for use in the present invention. [Figure 14] FIG. 14 illustrates a particular exemplary embodiment of the steps for extracting muscle activation parameters. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0044] The principle 1 is illustrated in Fig. 1, where a preferred use of the device integrates the IMU sensor and the muscle activity sensor in the same box or measurement node 1. However, these two sensors may be separated. Furthermore, the invention also includes the integration of components directly within the garment, for example in the form of a compression band 2 that serves to hold the sensor on the body part and thus gives a good mechanical coupling for detecting muscle vibrations and limiting measurement artifacts.
[0045] Furthermore, the method of the present invention is also extended to measuring movement and pose using markerless optical systems such as a depth camera 3. In environments where the movement space is limited and uncontrolled, this may prove to be an interesting alternative to an IMU or provide additional redundancy in measurements to verify body pose and movement, and elements of the environment (obstacles, objects, etc.) as the subject evolves.
[0046] The embedded signal processing electronics allows to perform certain computational operations in order to facilitate wireless communication to an external receiver 4 (smartphone type). This receiver can perform complex operations on the data and / or communicate via a mobile data network to a computer / server 5. Cross-analysis of the exercise data and the muscle activity data can provide important information on the performance and health status of a person in their daily activities. In particular, the fusion of the exercise data and the muscle activity data allows to individually set a physiological rest so that the muscle tissue returns to a resting state in a metabolic and mechanical sense after an effort.
[0047] We now describe the measurement system implemented in a practical embodiment of the invention. To detect movement and pose, a distinction is made between instruments for detecting movement and pose. Three categories are identified: commercially available IMU sensors, MEMS sensitive elements for integration into connected devices (clothing or objects) and depth cameras.
[0048] The MTw® Awinda® motion tracker from XSENS
[18] may be selected as a fully integrated IMU sensor for 3D motion imaging. The specifications of this sensor are summarized in Table 1.
[0049] [Table 1]
[0050] A MEMS technology inertial unit IMU is integrated together with muscle activity sensors to obtain coexisting measurements on a given body part. This unit may be 6-axis, measuring (3 axes) and rotation (3 axes). It is also possible to associate a magnetometer (3 axes) to determine the orientation of the body part relative to the magnetic north of the Earth. Components from Invensense were selected for their performance and cost. Their characteristics are summarized in Table 2.
[0051] [Table 2]
[0052] Microsoft(R) Kinect(R) is a low-cost system consisting of a color camera (RGB), an infrared camera, and an infrared projector. This system is used by Plantard[9] to capture the movements and postures of industrial operators. The features of Kinect(R) V1 and V2 are shown in Table 3.
[0053] [Table 3]
[0054] For muscle activity measurements, the constant component of the triaxial accelerometer is eliminated to retain only the variability due to micro-vibrations at the skin surface. The sensor requires a very low noise floor of less than 100 μg / √Hz. Accelerometers used in seismic exploration to capture these phenomena are suitable for measuring mechanomyogram signals. Two preferred components for the present invention, the ADXL354 and its digital equivalent, the ADXL355
[19] , have been selected, and their performance is summarized in Table 4.
[0055] [Table 4]
[0056] The preferred use of the device integrates IMU and MMG sensors in the same box. However, these two sensors may be separated, one system for measuring posture and another for measuring muscle activity. Furthermore, it is also possible to provide for the integration of components directly within the connected garment (e.g. a compression band that helps to hold the sensor in place on the body part). Furthermore, the inventive system is extended to posture and movement measurements using markerless optical systems such as depth cameras. This may prove to be an interesting alternative to IMU sensors in situations where the workstation is part of a well-defined environment or may provide information about the scene context and redundancies in measurements to validate body posture and movement.
[0057] The electronics embedded in the sensor make it possible to perform certain computational operations in order to facilitate wireless communication to an external receiver 4 (e.g. a smartphone or a data collector). This receiver then performs complex operations on the data (synchronization, segmentation, processing) and communicates the analysis results to a computer 5, a smartphone or the cloud.
[0058] The results generated relate, for example, to the level of exposure to certain biomechanical risk factors (posture, intensity of muscular activity) and the monitoring of these factors during physical activity. Referring to Fig. 4, an alert could be issued in case of exposure to extreme conditions from a biomechanical point of view.
[0059] -Too strong. Intense muscle activity obtained via mechanomyogram signals -Too fast. High speed movement achieved by speed and acceleration of body parts. -Too far. Obtained via joint angle -Time is too long. Data part allows calculation of exposure time.
[0060] An embodiment of the present invention provides a second level of analysis after processing of the data: the movement and muscle activity data are processed on a PC by algorithms that allow the calculation of the biomechanical performance of the subject's gestures and their physiological and biomechanical characterization. Another result provided by integrating the data is the physiological rest period during which the vibration phenomena to which it refers are restored, thus avoiding exposure to accidents and the risk of occupational diseases.
[0061] Some examples of the production of essential technical bricks are described. Thus, for an MMG sensor, one may provide one of the following:
[0062] -MEMS accelerometers that meet the performance requirements for measuring MMG signals.
[0063] - Signal processing tools to extract MMG parameters related to muscle force. The motion sensor may be integrated -MEMS inertial unit 9D for joint integration with MMG sensors.
[0064] - Inertial data fusion algorithm (using Kalman filter) to correct sensor drift and its sensitivity to leakage fields.
[0065] -Quick setup and calibration method that is robust over time to minimize disruption to operators and production.
[0066] This calibration must precisely synchronize all deployed sensors to ensure a reliable mobile signal.
[0067] The data collector (receiver) may be integrated, - Function of receiving, time of synchronization of movements and MMG data using an external clock (smartphone clock, timestamp sent by PC, etc.).
[0068] - Data storage in source files for post-processing on PC, - Algorithm that automatically divides a sequence of movements.
[0069] Fusion of inertial and mechanomyogram data may include:
[0070] - Calculate the level of exposure to risk factors during physical activity and monitor these factors during physical activity.
[0071] - Effort model that converts mechanomyogram signals into force signals using calibrated effort data for each muscle group of interest.
[0072] -Algorithms for calculating biomechanical performance from source file data, including synchronized movement and mechanomyogram sensor data.
[0073] -Implements machine learning algorithms to automatically recognize operator gestures and their effects on the musculoskeletal system.
[0074] - Calculating the physiological rest period so that the subject's muscles recover the reference oscillatory behavior.
[0075] We now describe the signal processing tools implemented in the method of the invention. To access the motion and attitude parameters, it is necessary to use orientation measurement fusion algorithms. Typically, these algorithms rely on Kalman filters, as presented in
[20] .
[0076] To condition the mechanomyogram signal, under laboratory conditions, researchers generally oversample the signal at a frequency of the order of 1 kHz or 2 kHz, while the characteristic frequency of the MMG signal is below 250 Hz. With a view to the deployment of fields with wireless data communication, a compromise between the amount of data to be transmitted and sampling was found by setting the sampling frequency at 500 Hz according to the Nyquist criterion. The raw signal from the acceleration sensor is digitized and then conditioned. The digitized MMG signal has two components: a static component (DC) and a dynamic component (AC). The DC component is not useful for the assessment of muscle activity and therefore needs to be filtered. Moreover, body movements are low-frequency components that contaminate the muscle activity information. In fact, the cutoff frequency of the high-pass filter is included in the band between 2 Hz and 50 Hz, with 20 Hz being preferred, in order to clean the parasitic components mentioned above.
[0077] Applying a low-pass filter cuts out high-frequency noise and limits the band of interest to frequencies characteristic of muscle microcontractions. A cutoff between 70Hz and 250Hz, and especially between 200Hz and 250Hz, is ideal for the analysis of MMU signals. A preferred value of 250Hz has been established.
[0078] The use of Butterworth low-pass digital filters or Savitzky-Golay filters is common in today's technology.
[0079] For filtering operations, a 5-pole Butterworth filter is ideal because it has constant gain across the passband, despite its low roll-off compared to Chebyshev or elliptical filters. Additionally, the ADXL355 digital accelerometer offers programmable low-pass and high-pass filters to select the frequency band of interest.
[0080] The processing of mechanomyogram signals, as well as the corresponding electromyogram-dependent methods that have been developed similarly, may be classified into four groups: time and frequency methods (the most traditional methods), then time and time-scale methods (the more recent methods). The choice of an appropriate processing method is then crucial for the objective analysis of MMG signals. Indeed, during isometric contractions (muscle contractions without changes in muscle length), the signal can be assumed to be stationary (i.e., its statistical properties are invariant over time) and traditional signal processing methods based on the Fourier transform are applicable. However, during movements with variable dynamics, the muscle can change its length or recruit more motor units, thus giving rise to so-called non-stationary signals. For this type of muscle activity, the use of time frequencies or time scales becomes necessary. Some parameters extracted from
[21] are presented below, with particular attention to their uses and limitations.
[0081] Once the mechanomyogram signal has been segmented and properly conditioned, the parameters of interest can finally be extracted. The MMG signal has three components (MMGX, MMGY and MMGZ) and represents the accelerations injected by the vibration of muscle fibers along the three spatial directions (X, Y, Z). The "total" acceleration signal is calculated by the following operations:
[0082]
number
[0083] The RMS (root mean square) amplitude of the "sum" MMG signal then allows obtaining information about the force generated by the muscle. The RMS amplitude varies with the variations in tension of the muscle fibers and increases with the level of muscle contraction. It is the most used parameter in the temporal analysis of the MMG signal and is given by the following formula:
[0084]
number
[0085] The observation window N is equal to the characteristic period of the movement sequence divided by 2. This characteristic period is defined by the period of the movement to be learned, for example a step in walking, a stride in running, or the period of handling an object. For static postures, a 1 second window allows to establish the RMS amplitude that contains sufficient attributes regarding the physiological and biomechanical behavior of the underlying muscles.
[0086] However, the sensitivity of this parameter to physiological tremors and other mechanical artifacts requires additional methods of analysis.
[0087] Analysis of the power spectral density (PSD) of the MMG signal allows observing the variations in frequency content in order to infer information about muscle fatigue. The standard tool for this type of analysis is the Fast Fourier Transform (FFT), which moves from the time domain to the frequency domain. Parametric methods models using the autoregressive method (AR) are able to estimate the PSD of the MMG signal without using an apodization window and therefore provide better results. The most common methods are Yule - Walker and Burg. The preferred PSD method in the present invention is that of Yule-Walke. Once the PSD is estimated, the mean frequency (MPF frequency of the mean power) may be determined by the following formula:
[0088]
number
[0089] Let PSD be g2 / Hz, power spectral density of MMG signal be fs, and sampling frequency be Hz. MPF is an important indicator for investigating changes in muscle condition and detecting characteristics of fatigue signs.
[0090] During activities with dynamic changes, muscles may change their length, recruit more motor units, adapt the frequency of stimulation, imparting a non-stationary behavior to the MMG signal. Therefore, a time-frequency approach is necessary to split the signal in the time domain before performing the frequency analysis. A compromise between ease of execution on a microcontroller and saving battery (for wireless communication) is the local Fourier transform (STFT, short for short-time Fourier transform), which allows a window to "slide" over the time signal and quickly obtain the PSD.
[0091]
number
[0092] where x(t) is the MMG signal, h(t-τ) is the sliding window, and τ is a parameter that allows the information to be always spectrally analyzed.
[0093] The drawback of such methods is the selection of the appropriate data range, which may result in resolution deficiencies in the frequency domain.One of the innovative features of the present invention is the use of the orientation measurement of the IMU sensor to split the MMG signal in an appropriate way.
[0094] Other time-frequency methods such as the wavelet transform (WT for short) and the Wigner-Ville transform (WVT) are frequently used in laboratories or other controlled environments to analyze MMG signals.
[0095] More recent methods called time-scale have experienced greater success by scientists in processing MMG signals. One method used, particularly in McLeod's invention
[15] , is wavelet packet analysis (WPA). It differs from other time-scale methods due to the multiscale decomposition of the starting signal, which is divided into low-frequency coefficients (approximation level) and high-frequency coefficients (detail level). These coefficients then form "wavelet packets". A specific toolbox is applied in MATLAB(R) software.
[0096] This method is very efficient in the analysis of MMG signals, but it requires heavy post-processing and does not allow automatic estimation of muscle fatigue in real time. Moreover, it requires complex computational operations, which are not compatible with integration within connected objects, and requires wireless communication in uncontrolled environments for several hours. Nevertheless, the present invention allows a multi-resolution analysis of partially post-processed raw MMG data based on the Maximum Overlap Discrete Wavelet Transform (MODWT). This technique allows the extraction of movement artifacts from muscle signals with higher accuracy than traditional filtering techniques.
[0097] The method steps according to the present invention will now be described.
[0098] Measurement nodes 1, whose architecture is shown in Figure 2, are placed on one or more body parts of a person to estimate the level of biomechanical stress during the person's activities. Each node is rigidly attached to the body part by an elastic band, slightly compressing the sensor on the skin and thus providing a good mechanical coupling to detect muscle coupling and limiting measurement artifacts.
[0099] Depending on the complexity of the task and the number of muscles used, the user may equip himself with one or more nodes located on the abdomen of each muscle to be analyzed. Each node may communicate information to a receiving station (e.g., a smartphone) and / or receive information. The communication protocol chosen for the present invention is Bluetooth Low Energy (BLE). The advantage of BLE is that it reduces energy consumption and that slave devices are "discoverable" by the master device while minimizing consumption. Likewise, slave devices may remain connected to the master device and exchange data periodically. In the case of BLE, there is no limit to the number of peripherals supported by the same master, in contrast to the conventional Bluetooth, which is limited to seven peripherals. The typical total throughput of BLE is theoretically 1 Mbps, but in practice remains at 250 kbps, shared between all slave nodes. A feature of the method according to the present invention is that by having a large number of sensors communicate, it allows the analysis of various biomechanical risk factors simultaneously, supporting a range of use of 8 hours. The data is also stored in a micro SD type memory. The measurement node supports wired communication via a USB port.
[0100] The receiving station can activate the sensor and associate the body position to the sensor through Near Field Communication (NFC). Detection of the sensor's position on the body can adjust certain signal acquisition parameters, such as specific filter templates. Once the sensor is installed and marked, acquisition can be started with a simple command at the receiving station. The raw motion data from the IMU sensors are then processed and fused by the microcontroller of the localization node to transmit the orientation and position information of each part and joint to the receiving station. The internal processing of the motion data allows the output signal to be sub-sampled in order to optimize the battery consumption and the amount of information transmitted. The sampling of the IMU output data is typically between 50Hz and 120Hz. The receiving station then performs calculation operations such as counting and detection of excessive motion amplitudes. Alerts regarding repetitive postures and movements can be sent to an external computer or generated by the smartphone.
[0101] To obtain redundancy of motion data, or in the absence of IMU sensors, the use of depth cameras or other optics without markers allows access to the motion data required for biomechanical analysis and also allows the collection of contextual elements of the scene.
[0102] The MMG sensors integrated into the device, in addition to generating results regarding the movements and postures imposed, provide indicators regarding the forces developed by muscles and muscle fatigue, as well as the distribution of stress on all analyzed body parts.
[0103] The cross-analysis of the movements and the mechanomyogram signals allows the objective quantification of the level of biomechanical stress on the subject. Figure 3 shows a block diagram illustrating all the operations and analysis according to the method of the invention. In this mode of operation, the IMU sensors characterize certain parameters such as the number of technical gestures per minute, their speed and the angles of the joints. Furthermore, the vibration signals from the MMG sensors are divided into 1 second intervals to calculate the RMS amplitude level and the mean frequency of the MPF power spectrum. It is generally accepted in the art that the RMS amplitude of the MMG signal reflects the level of force deployed by the muscle. However, the analysis of the MPF allows highlighting changes in terms of muscle activation (muscle fiber recruitment plan, muscle fatigue, etc.) and can give a fatigue indicator. An example of the recovery of biomechanical risk factors is shown in Figure 4. Furthermore, the drift of these biomechanical risk factors over time is analyzed by linear regression (Pearson method) to calculate linear regression coefficients. This indicator is useful for estimating the level of physical strain over time and predicting the physiological rest period required for the target muscle to regain a reference mechanical state (according to Baillargeon
[17] , see Figure 5).
[0104] Let us take the example of a person performing repetitive activities moving the waist in a particular operating mode related to the prevention of risks related to biomechanical stress. The method of the invention, as well as the associated measurement system, may be used to calculate the rest time required for the muscles of the lumbar region to return to a reference state corresponding to a low level of physical strain as shown in FIG. 6. The level of muscle activity of the muscles is characterized by the RMS amplitude 61 obtained from the filtered mechanomyogram MMG signal. This is related to the mean power frequency MPF 62, the variations of which indicate the changes in the recruitment state of the muscle fibers. A maximum strain threshold 63 is first determined, for example obtained at 30% of the maximum value of the filtered mechanomyogram signal. Other techniques for determining this strain threshold may be envisaged, for example by taking the maximum voluntary force in the muscle group of interest (MVF). A minimum strain threshold 64 is then determined, for example by taking 10% of the filtered mechanomyogram signal. The objective is to calculate the rest time to bring the level of physical strain of the lumbar region muscles closer to the minimum threshold.
[0105] The average values of RMS amplitude and mean power frequency MPF are calculated for the target portion 65. This calculation is repeated for each active portion and linear interpolation is performed for the average RMS amplitude and mean MPF 66. Note that the RMS profile increases during activity, indicating increasingly demanding activity on the lumbar muscles and a decrease in the MPF, indicating the recruitment of a greater number of muscle fibers.
[0106] To calculate the physiological rest time for the muscle to return to the reference state, the technique used in the present invention uses the previously calculated linear regression coefficients to determine a linear decrease profile of the mean RMS amplitude relative to the minimum physical strain threshold. Conversely, a linear growth profile is determined so that the MPF returns to the reference state called fatigue-free.
[0107] Thus, by comparison with the reference state, the fluctuations or drifts of the fatigue index (MPF) and the force index (RMS amplitude) can be determined, indicating changes in the muscle fiber recruitment plan and stress intensity, respectively. An alert can be sent to the operator to warn him that he is exposed to a zone of high physical and biological strain.
[0108] In a particular operating mode related to physical ergonomics, the method of the invention produces an ergonomic assessment at a workstation or equipment item as well as the associated measurement system. Take the example of a worker illustrated in FIG. 7, wearing an exoskeleton 71 to facilitate the wearing of a portable electrical appliance 72. The movements and vibrational movements are measured at the right shoulder and right arm by the measurement system 1 to determine whether the exoskeleton is beneficial to the skeletal health and safety of the worker (posture correction, distribution of the carried load at arm's length). A typical example of the collected data is shown in FIG. 8. We can clearly see periods of strong muscular activity 81, uncomfortable postures 82 that cause excessive static efforts, and dynamic forces 83 during rapid movements. From this information, it is possible to provide a fully automated and objective ergonomic assessment by comparing the situation before and after wearing the exoskeleton. Indeed, changes in the arm joint flexion angles can cause a decrease or increase in the RMS amplitude of the MMG signal (an indicator of the deployment forces), reflecting different joint recruitment depending on whether the operator is wearing the exoskeleton or not. Statistical analysis of the differences in means between the no-exoskeleton and the exoskeleton situations allows the generation of a map of biomechanical risk, shown in Figure 9.
[0109] Another innovative mode of operation of the present invention uses the motion data to segment the physical activity into different states (maintaining a position, recognizing periodic gestures, etc.). This segmentation allows to determine the optimal acquisition parameters (acquisition window, filter template, and sampling) of the MMG sensor to analyze the impact on the muscles of a gesture or a series of clear gestures. This method is used in particular for the segmentation of the squat activity according to Woodward in
[22] . The microcontroller of the measurement node calculates the RMS amplitude level to detect the muscle stress level and provides an estimate of the PSD of each acquisition. The data extracted from the MMG sensor are retransmitted to the receiving station to determine a time-frequency representation of the activity. This method allows to correlate a gesture or a very precise series of gestures with the vibration signature of the muscle activity. The segmentation can be done manually retroactively or automatically by the receiving station for gesture recognition methods and machine learning techniques.
[0110] This mode of operation may find application in the sports field, where an athlete strives to develop his performance by perfecting technical gestures. In the example of FIG. 10, a runner 101 is equipped with the measurement system 1 to analyze the muscle fatigue caused by the repeated gesture of striding. The IMU and MMG signals of a typical joint, the right hip joint, are presented in FIG. 11, where the stride is characterized by a pattern 111. The duration of this pattern (less than 1 second) allows the determination of a window of acquisitions for the post-processing of the MMG signal. The measurement nodes are processed to estimate the PSD, which is then processed by the receiving station providing the STFT representation of FIG. 12. The power spectral density PSD is estimated by the Burg or Yule-Walker algorithm and can be observed in a frequency band between 20 Hz and 250 Hz. Each vertical line represents an acquisition with its frequency signature, i.e. the distribution of the signal energy over the frequency band of interest. It is therefore possible to observe the vibration signature of part 111 that characterizes striding, to observe the drift of this signature over time, and in particular the MPF, to prevent the runner from tiring and to optimize his training. Indeed, the variation of the MPF over time reflects the change in the way muscle fibers are recruited, in particular by the ratio between the high-frequency and low-frequency components of the MMG signal. High frequencies are caused by fast fibers and low frequencies by slow fibers. A decrease in the ratio between high and low frequencies (and therefore a decrease in the MPF) over a long period of time can objectify peripheral fatigue. Thus, as fatigue accumulates, the fast fibers tend to be blocked and then partially compensated by the activation of the slow fibers. Conversely, phases of short, intense contractions recruit faster muscle fibers and therefore increase the MPF.
[0111] A specific example of a method for obtaining a high quality mechanomyogram for use in the present invention will now be described with reference to FIG.
[0112] Mechanomyogram (MMG) signals obtained by a three-axis accelerometer are capable of detecting changes in muscle activity behavior due to fatigue and effort intensity.
[0113] Filtering motion artifacts is a real challenge, since the characteristics related to the activity of the subject change. Indeed, to analyze the quadriceps during a typical walk, the leg movements have a frequency of around 1 Hz, reaching up to 4 Hz during high-speed running. Also, the impact transmitted to the foot by the impact then travels along the leg with a spectrum containing components up to 20 Hz, and is then attenuated by the abdominal-lumbar belt.
[0114] Therefore, the movement signal needs to be analyzed in order to select the processing operations to be applied to the mechanomyogram signal beforehand. The movement signal from the inertial unit allows the identification of static postures, sudden gestures or pauses between different sequences of movements, thus isolating "sequences" in the mechanomyogram signal. The identification of these sequences is the result of a physical activity segmentation operation that facilitates the selection of the signal processing operations to be applied to the mechanomyogram signal (MMG). Then, the processing of the movement signal results in the segmentation of the activity, see FIG. 14. The processing of the MMG signal then results in the extraction of the RMS amplitude and the mean power frequency MPF.
[0115] Using video can be practical to improve the quality of the segmentation when done post-processing and manually. Nevertheless, it is possible to consider automating the motion segmentation for real-time analysis by training algorithms through the use of inertial units and data from video.
[0116] In the static posture example, a digital 3-pole high-pass Butterworth filter with a cutoff at 0.5 Hz is a good candidate for processing the MMG signal.
[0117] Other types of filters may be used, for example Chebyshev or elliptical filters with a steeper slope in the rejected band to impair ripples in the transmitted and / or rejected bands.
[0118] For slow motion, cycles, and no shocks (i.e., ground impacts), the spectrum of the acceleration signal can be calculated to identify a narrow frequency band that can be filtered with a digital 5-pole Butterworth passband filter.
[0119] In the last case, where there is a movement of variable dynamics and which may cause an impact, the acceleration signal of the IMU sensor can be used in an adaptive filtering process, for example using the LMS (Least Mean Square) algorithm. Another technique employing Multiresolution Analysis (MRA), by applying the 7-level Daubechies wavelet "db6", allows the reconstruction of the mechanomyogram signal deprived of the components linked to the movement.
[0120] Applying a low-pass filter also cuts out high-frequency noise. A cutoff between 200 Hz and 250 Hz is ideal for the analysis of mechanomyogram signals. Using a Butterworth low-pass digital filter or a Savitzky-Golay filter is common practice in today's technology.
[0121] Of course, the invention is not limited to the examples described above and many other embodiments may be envisaged without departing from the scope of the invention.
[0122] References 1. MBI Reaz, MS Hussain, and F. Mohd-Yasin. Techniques of EMG signal analysis: Detection, processing, classification and applications. Biological Procedures Online, 8(1):11 35, Mar 2006. 2. WH Wollaston. On the duration of muscular action. Philos Trans R Soc London, pages 1-5, 1810. 3. F. V. Brozovich and G. H. Pollack. Muscle contraction generates discrete sound bursts. Biophysical Journal, 41(1):35-40, 1983 4. N. Babault et al. “Neuromuscular fatigue development during maximal concentric and isometric knee extensions.” In: Journal of Applied Physiology 100.3 (Mar. 2006), pp. 780-785. 5. G. Trager et al. “Comparison of phonomyography, kinemyography and mechanomyography for neuromuscular monitoring.” In: Canadian Journal of Anesthesia / Journal canadien d’anesthesie 53.2 (Feb. 2006), pp. 130-135. 6. J. Marusiak et al. “Electromyography and mechanomyography of elbow agonists and antagonists in Parkinson’s disease.” In: Muscle & Nerve 40.2 (Aug. 2009), pp. 240-248. 7. A. Islam et al. “Mechanomyography Sensors for Muscle Assessment: a Brief Review.” In: Journal of Physical Therapy Science 24.12 (2012), pp. 1359-1365. 8. X. Robert-Lachaine, H. Mecheri, C. Larue, and A. Plamondon. Effect of local magnetic field disturbances on inertial measurement units accuracy. Applied Ergonomics, 63:123-132, Sep 2017. 9. P. Plantard. “Objectivation et standardisation des evaluations ergonomiques des postes de travail a partir de donnees Kinect” [Objectification and standardization of ergonomic assessments of workstations based on Kinect data.] PhD thesis. July 2016. 10. Biosleeve human-machine interface. May 2013. URL: https: / / patents.google.com / patent / US20130317648A1 / en. 11. Wearable Physiological Sensor System for Training and Therapeutic Purposes. Apr. 2017. URL: https: / / patents.google.com / patent / US20170312576A1 / en. 12. Systems, articles, and methods for gesture identification in wearable electromyography devices. Dec. 2014. URL: https: / / patents.google.com / patent / US20150169074. 13. Methods and devices for combining muscle activity sensor signals and inertial sensor signals for gesture-based control. Feb. 2014. URL: https: / / patents.google.com / patent / US20140240103. 14. Biomechanical activity monitoring. May 2015. URL: https: / / patentscope.wipo.int / search / en / detail.jsf?docid=W02015063520. 15. Real-time assessment of absolute muscle effort during open and closed chain activities. Feb. 2010. URL: https: / / patents.google.com / patent / US20110196262A1 / en. 16. J. Laine and D. Mougenot. “Benefits of MEMS Based Seismic Accelerometers for Oil Exploration.” In: TRANSDUCERS 2007 - 2007 International Solid-State Sensors, Actuators and Microsystems Conference. IEEE, 2007, pp. 1473-1477. 17. M. Baillargeon and L. Patry. “Work-related upper limb musculoskeletal disorders: definitions, functional anatomy, pathophysiological mechanisms and risk factors” [Work-related upper limb musculoskeletal disorders: definitions, functional anatomy, pathophysiological mechanisms and risk factors] In: Regional Health and Social Services Authority of Montreal-Centre. Public Health Department (2003). URL: http: / / www.bdsp.ehesp.fr / Base / 508934 / . 18. MTw Awinda - Products - Xsens 3D motion tracking. URL: https: / / www.xsens.com / products / mtw-awinda / . 19. ADXL355 Datasheet and Product Info | Analog Devices. URL: https: / / www.analog.com / en / products / adxl355.html. 20. AM Sabatini. “Quaternion-based strap-down integration method for applications of inertial sensing to gait analysis.” In: Medical & Biological Engineering & Computing 43.1 (Feb. 2005), pp. 94-101. 21. M. Ibitoye et al. “Mechanomyographic Parameter Extraction Methods: An Appraisal for Clinical Applications.” In: Sensors 14.12 (Dec. 2014), pp. 22940-22970. 22. R. B. Woodward et al. “Segmenting Mechanomyography Measures of Muscle Activity Phases Using Inertial Data.” In: Scientific Reports 9.1 (Dec. 2019), p. 5569
Claims
1. 1. A system for analyzing a subject's biomechanical activity and exposure to biomechanical risk factors in the context of physical activity, comprising: a means attached to one or more first body parts of the subject for collecting vibration signals, the measurements of which reflect local muscle activity; and b) means for collecting signals representative of the subject's movements, the measurements reflecting orientation and movements of one or more second body parts in two or three dimensions; c means for processing those signals and extracting therefrom indicators representative of the intensity of the biomechanical stress and its frequency, and (d) means for detecting drift of said vibration signal with respect to a reference frame of a vibration behavior of said muscle in the context of a physical activity, The indicators thus extracted include RMS amplitude and mean power frequency MPF; and means for calculating a linear regression coefficient by analyzing the drift of the RMS amplitude and the mean power frequency MPF averaged for each active portion of a plurality of active portions.
2. 2. The analysis system of claim 1, further comprising means for comparing the RMS amplitude and / or the mean power frequency MPF extracted from a predetermined portion of an activity before and after use of a physical assist device, the RMS amplitude and / or the mean power frequency MPF being implemented in the context of using a physical assist device.
3. 2. The analysis system according to claim 1, characterized in that it is carried out in the context of an ergonomic evaluation of a workstation and further comprises means for comparing the RMS amplitude and / or the mean power frequency MPF extracted from a predetermined portion of activity before and after a change in the workstation.
4. 4. Analysis system according to claim 1, characterized in that the means for collecting the movement signals comprise an inertial unit IMU (Inertial Measurement Unit).
5. 5. Analysis system according to claim 4, characterized in that the inertial unit IMU is integrated with muscle activity sensor means in order to obtain co-localized measurements on a body part of the subject.
6. 6. The analysis system according to claim 4, wherein the inertial unit IMU is a 6-axis type that measures linear acceleration (3 axes) and rotation (3 axes).
7. 7. Analysis system according to any one of claims 1 to 6, characterized in that the means for collecting the movement signals further comprises a magnetometer (3-axis) for determining the orientation of the body part relative to the magnetic north of the earth.
8. 8. The analysis system according to claim 1, wherein the means for detecting muscle activity comprises an MMG (Mechanomyogram) accelerator arranged to generate a Mechanomyogram signal.
9. 9. An analysis system according to any one of claims 1 to 8, comprising a plurality of measurement nodes rigidly attached to a body part of a subject.
10. 10. The analysis system of claim 9, wherein the measurement node includes communication means for communicating with the receiving station.
11. 11. The analysis system according to claim 10, wherein the communication means implements a wireless communication protocol.
12. 1. A method for achieving ergonomic evaluation of a physical assist device provided to a workstation or operator, comprising: a) attaching a vibration sensor to one or more first body parts of a subject to collect vibration signals, the measurements of which reflect local muscle activity; b collecting signals representative of the subject's movements, the measurements reflecting orientation and movements of one or more second body parts in two or three dimensions; c) processing the signals and extracting therefrom an indicator representative of the intensity of the biomechanical stress, this signal processing producing a frequency of biological activity; The collecting and processing steps (a), (b), and (c) are completed (i) before changing the workstation or donning the physical assist device, and (ii) after changing the workstation or donning the physical assist device, respectively; and The method further comprises the step of comparing the indicators of intensity and frequency of the biomechanical stress extracted (i) before changing the workstation or wearing the physical assist device and (ii) after changing the workstation or wearing the physical assist device.
13. A method for analyzing a subject's biomechanical activity and exposure to biomechanical risk factors in the context of physical activity, implemented in an analysis system according to one of claims 1 to 11, comprising: a) attaching to one or more first body parts of the subject for collecting vibration signals, the measurements of which reflect local muscle activity; b collecting signals representative of the subject's movements, the measurements reflecting orientation and movements of one or more second body parts in two or three dimensions; c processing the signals and extracting therefrom an indicator representative of the intensity of biomechanical stress, this signal processing producing a frequency signature of biological activity; d. detecting a drift of said vibration signal with respect to a reference frame of the behavior of said vibration signal of said muscle in said context of physical activity; e predicting a physiological rest period required for the muscles of the subject to recover to their reference oscillatory behavior; The indicators thus extracted include RMS amplitude and average power frequency; The method, wherein the predicting step includes calculating a linear regression line by analyzing the RMS amplitude and the average power frequency averaged over each portion of a plurality of portions of activity, and using the linear regression line to determine a linear decay profile of the RMS amplitude toward a predetermined minimum threshold of physical strain.
14. 14. The method of claim 13, further comprising the step of collecting by optical collection means a context or scene in which the subject is developing, and processing this context or scene to generate information about the subject's posture and gestures that correlate with the vibration behavior signal.
15. 15. The method of claim 14, further comprising estimating the muscle state and its drift over time by segmenting the subject's activity into specific and / or repetitive tasks or tasks and correlating them with muscle activity signals.
16. 16. The method according to any one of claims 12 to 15, further comprising the step of generating an individual biomechanical risk reference frame for a given subject from the set of biomechanical characteristics of muscle activity obtained for this person.
17. The method of claim 16, wherein the step of generating an individual biomechanical risk frame of reference implements "machine learning" techniques.
18. 18. The method according to any one of claims 12 to 17, further comprising the step of cross-analysing the movement signals and the muscle activity signals to extract information relating to the subject's performance and health during the repetitive muscle activity.
19. 18. The method of any one of claims 12 to 17, further comprising processing the movement data and muscle activity data to recommend a personalized placement of a physiological rest, such that after an effort, the subject's muscle tissue returns to a metabolic and mechanical resting state.
Citation Information
Patent Citations
Action condition estimating device, action condition learning device, action condition estimation method, the action condition learning method, and program
JP2011182824A
method and system for quantifying an intention of movement of a user
US20100292617A1
Monitoring and treating pain with epidermal electronics
US20170136264A1
Signal processing for decoding intended movements from electromyographic signals
WO2018026842A1
