Google Patentsopen access

Biomechanical activity monitoring — Imperial Innovations Limited (US10335080B2)

ABSTRACT

Abstract

A wearable sensor apparatus comprises a motion sensor configured to sense two or three dimensional movement and orientation of the sensor and a vibration sensor configured to sense acoustic vibrations. The apparatus includes means for attaching the motion sensor and the vibration sensor to a body. The sensor apparatus enables long term monitoring of mechanomyographic muscle activity in combination with body motion for a number of applications.

Description

The present invention relates to apparatus and methods for monitoring and/or analysing biomechanical activity in, for example, the human or animal body.

Monitoring human biomechanical activity is an important function in a diverse range of technical applications, including in both clinical and non-clinical environments. In the clinical environment, these functions may include health monitoring, such as for diagnosis, therapeutic intervention, rehabilitation, well-being and foetal monitoring etc. In the non-clinical environment, these functions may include providing human-machine interfaces, robot control, haptic systems and systems for use in sports training etc.

When providing combined muscle activity and motion recording, prior art systems have focused on monitoring muscle activity using electromyography (EMG) sensors, i.e. monitoring electrical activity produced by skeletal muscles. This technique can have significant disadvantages and has limited use outside well-controlled environments such as laboratories and clinical care establishments. Such electrical recording systems generally require single-use sensors (e.g. for clinical safety and hygiene reasons) and the use of adhesives and electrically conductive gel for attachment of the sensors to the body to ensure adequate electrical contact with the human subject. This imposes limitations on the practicality and ease of use of muscle activity sensing, on the environment of use, and on the duration of use. Thus detailed data acquired with such sensors may only be obtained for short periods of time, for a limited range of the subject's motion.

It is an object of the present invention to provide an alternative technique for monitoring and/or analysing biomechanical activity which reduces or mitigates some or all of these disadvantages.

According to one aspect, the present invention provides a wearable sensor apparatus comprising:

a motion sensor configured to sense two or three dimensional movement and orientation of the sensor;

a vibration sensor configured to sense acoustic vibrations; and

means for attaching the motion sensor and the vibration sensor to a body.

The acoustic vibrations may be, or may be represented by, bioacoustic signals. The acoustic vibrations may be bioacoustic vibrations. The motion sensor may comprise an inertial measurement unit. The vibration sensor may be configured to sense skeletal muscle vibrations. The inertial measurement unit may comprise one or more of an accelerometer, a gyroscope, and a magnetometer. The inertial measurement unit may be configured to sense rotation of the sensor body around at least one axis in space. The inertial measurement unit may be capable of sensing translational motion in three perpendicular axes and/or rotation about three perpendicular axes. The vibration sensor may comprise an acoustic pressure sensor. The vibration sensor may comprise one or more of an accelerometer, a microphone, and a piezoelectric transducer. The vibration sensor may comprise a volumetric chamber closed at one end by a flexible membrane, and a pressure transducer coupled to the chamber distal from the flexible membrane. The apparatus may comprise a barometer. The barometer may be configured to sense an ambient pressure. The vibration sensor and the barometer may be provided by a single sensor.

The apparatus may further comprise a data logging device coupled to receive motion signals from the motion sensor and acoustic signals, such as muscle vibration signals or mechanomyographic (MMG) muscle signals, from the vibration sensor, and to store said signals as a function of time. The apparatus may further comprise a classification processor configured to receive motion signals from the motion sensor and to receive the muscle signals (or mechanomyographic muscle signals) from the vibration sensor, and to classify a pattern of movement, or a posture of, at least one part of a body on which the wearable sensor apparatus is attached. The classification processor may be configured to use both the motion signals and the muscle vibration signals (or mechanomyographic muscle signals) to determine simultaneous patterns of movement, or postures of, multiple articulating parts of the body on which the wearable sensor apparatus is attached. The muscle vibration signals may be mechanomyographic muscle signals. The classification processor may be configured to classify a pattern of movement, or a posture of, at least one part of a body on which the wearable sensor apparatus is attached based on the mechanomyographic muscle signals.

The classification processor may be configured to separate the signals from the vibration sensor into windowed data. The classification processor may be configured to perform cluster analysis on the windowed data to determine a correlation between the signals from the vibration sensor and a type of activity.

The cluster analysis may comprise determining clusters of the windowed data and comparing one or more properties of the clusters with one or more threshold values. A property of a cluster may be an averaged value of a property of the windowed data within the cluster. The one or more properties comprise one or more of: gyroscopic magnitude, peak gyroscopic magnitude, ambient pressure, a cadence of a user (which may be a number of steps taken in a unit of time) and orientation of the motion sensor.

The apparatus may comprise a classification processor configured to receive motion signals from the motion sensor, to receive an ambient pressure signal from the barometer and to receive signals from the vibration sensor, and to classify a pattern of movement, or a posture of, at least one part of a body on which the wearable sensor apparatus is attached based on the received signals.

The apparatus may further comprise a classification processor configured to receive motion signals from the motion sensor and to receive muscle vibration signals, such as mechanomyographic muscle signals, from the vibration sensor, and to classify a pattern of movement, or a posture of, at least one part of a body on which the wearable sensor apparatus is attached, based on said motion signals, and to identify muscular activity used during said pattern of movement or posture. The classification processor may be further configured to determine whether or not the identified muscular activity conforms to a predetermined pattern consistent with the classified pattern of movement.

The apparatus may further comprise a classification processor configured to receive motion signals from the motion sensor and to receive acoustic signals from the vibration sensor, and to determine when the acoustic signals correspond to foetal movement.

The apparatus may further include an interface module configured to provide output control signals for a computer processor based on the output of the classification processor. The apparatus may further include an interface module configured to provide output control signals for a motive apparatus based on the output of the classification processor. The apparatus may further include a said motive apparatus. The motive apparatus may comprise a prosthesis or a robotic device.

According to another aspect, the invention provides a method of classifying a pattern of movement, or a posture of, at least one part of a human or animal subject comprising the steps of:

obtaining motion signals from a motion sensor attached to the subject, the motion signals being indicative of sensed movement of the subject in two or three dimensions; simultaneously obtaining vibration signals from a vibration sensor attached to the subject, the vibration signals being indicative of sensed skeletal muscle vibrations or other acoustic output from the subject; and using the motion signals and the sensed vibration signals, such as skeletal muscle vibration signals, to classify a pattern of movement, or a posture, of at least one part of the subject.

Other acoustic output from the subject may comprise acoustic output due to foetal movements. The method may further include using the motion signals and the sensed vibration signals to classify a pattern of movement, or a posture of, multiple articulating body parts of the subject. The method may further include using the motion signals and the sensed vibration signals to control movement of a computer processor or a motive apparatus. The vibration signals may be skeletal muscle vibration signals.

According to another aspect, the invention provides a method of assisting a human subject in rehabilitation comprising:

obtaining motion signals from a motion sensor attached to the subject, the motion signals being indicative of sensed movement of the subject in two or three dimensions; simultaneously obtaining vibration signals from a vibration sensor attached to the subject, the vibration signals being indicative of sensed skeletal muscle vibrations or other acoustic output from the subject, such as a heart rate of, or breathing from, the subject; and classifying a pattern of movement, or a posture, of at least one part of the subject using at least the motion signals; determining whether the pattern of movement, or posture, of the at least one part of the subject conforms to a predetermined pattern of the sensed skeletal muscle vibration signals; and providing an audible or visual feedback to the user dependent on whether the pattern of movement, or posture, of the at least one part of the subject is consistent with the predetermined pattern of the sensed skeletal muscle vibration signals.

According to another aspect, the invention provides a method of monitoring a maternal body for foetal movement comprising:

obtaining motion signals from a motion sensor attached to the maternal body, the motion signals being indicative of sensed movement of the subject in two or three dimensions; simultaneously obtaining acoustic vibration signals from one or more acoustic vibration sensors attached to the maternal body; and using at least the motion signals from the motion sensor to determine periods of maternal activity to attenuate or exclude maternal acoustic vibration signals from the vibration signals to thereby detect acoustic vibrations associated with foetal movement.

The acoustic vibration signals may be bioacoustic signals. The method may further comprise using acoustic vibration signals obtained from the acoustic vibration sensor to determine an orientation of a foetus. The method may further comprise using acoustic vibration signals obtained from the acoustic vibration sensor to generate a heart rate or a breathing rate from the maternal and/or foetal body.

Embodiments of the present invention will now be described by way of example and with reference to the accompanying drawings in which:

FIG. 1 shows a schematic diagram of a wearable activity sensor apparatus;

FIG. 2 shows a perspective view of a wearable activity sensor implemented in a knee brace;

FIG. 3 shows a schematic diagram of a wearable activity sensor apparatus including signal processing and analysis processing modules;

FIG. 4 shows x, y and z accelerations of a subject during a ten step walk and classification thereof;

FIG. 5 shows a subject wearing two activity sensors deployed on upper arm and forearm with real time feedback display;

FIG. 6 shows a graph of mechanomyography (MMG) data obtained during isometric contraction of the forearm with fist clenches;

FIG. 7 shows sensor data collected during a sitting to standing task from (a) electromyogram sensing; (b) accelerometer MMG sensing; (c) microphone MMG sensing; and (d) magnetic angular rate and gravity (MARG) sensing;

FIG. 8 shows both motion sensor data and mechanomyographic sensor data responsive to a subject walking;

FIG. 9 shows a graph of motion sensor classified stationary and moving periods overlaid on MMG data to further support activity classification;

FIG. 10 shows wearable activity sensors provided on both upper and lower leg of a subject together with sensor data received therefrom during standing and a knee-lift task;

FIG. 11 shows (a) acoustic sensor output, and (b) a gel-based vibration sensor output corresponding to foetal movement in a pregnant subject;

FIG. 12 shows a wearable activity monitor for detecting foetal movements;

FIG. 13 shows a flow chart of a method for classifying a subject's movement, including foetal movement within the subject;

FIG. 14 shows another flow chart of a method for classifying a gait, posture or movement of a subject;

FIG. 15 shows acoustic sensor output and maternal sensation data corresponding to foetal movement in pregnant subjects;

FIG. 16 shows (a) gyroscope and (b) corresponding muscle response data for a subject during a walking task; and

FIG. 17 illustrates (a) gyroscopic data, (b) accelerometer data with calculated magnitude, (c) unprocessed (raw) MMG data, (d) filtered MMG data, and (e) filtered and processed MMG for a subject during a walking task.

Mechanomyographic (MMG) muscle sensing exploits a low frequency vibration emitted by skeletal muscle the measurement of which does not require electrodes, gel or direct skin contact, unlike electromyographic sensing. This offers potential for much more efficient implementation in everyday use. Combining mechanomyographic muscle sensing with two- or three-dimensional movement sensing has been found to provide a considerable advance in human activity monitoring by combining human dynamics and muscle activity information. Bioacoustic sensors, such as MMG sensors, can readily be provided at relatively low cost, and can be packaged in a lightweight wearable package which is easy to attach to the subject's body without the use of complicated procedures for ensuring good electrical contact with the subject's skin.

In a general aspect, the techniques described here combine the use of inertial measurement units (IMUs) and acoustic sensors, e.g. bioacoustic sensors, such as mechanomyographic (MMG) sensors for muscle activity sensing or foetal movement monitoring. An arrangement described herein in relation to sensing a mechanomyographic muscle signal may also be used to sense bioacoustic signals or muscle vibration signals more generally.

FIG. 1 shows a schematic diagram of a wearable activity sensor 1 comprising a support structure 2 , a motion sensor 3 and a <figure-callout id="4" la

The present invention relates to apparatus and methods for monitoring and/or analysing biomechanical activity in, for example, the human or animal body.

Monitoring human biomechanical activity is an important function in a diverse range of technical applications, including in both clinical and non-clinical environments. In the clinical environment, these functions may include health monitoring, such as for diagnosis, therapeutic intervention, rehabilitation, well-being and foetal monitoring etc. In the non-clinical environment, these functions may include providing human-machine interfaces, robot control, haptic systems and systems for use in sports training etc.

When providing combined muscle activity and motion recording, prior art systems have focused on monitoring muscle activity using electromyography (EMG) sensors, i.e. monitoring electrical activity produced by skeletal muscles. This technique can have significant disadvantages and has limited use outside well-controlled environments such as laboratories and clinical care establishments. Such electrical recording systems generally require single-use sensors (e.g. for clinical safety and hygiene reasons) and the use of adhesives and electrically conductive gel for attachment of the sensors to the body to ensure adequate electrical contact with the human subject. This imposes limitations on the practicality and ease of use of muscle activity sensing, on the environment of use, and on the duration of use. Thus detailed data acquired with such sensors may only be obtained for short periods of time, for a limited range of the subject&#39;s motion.

It is an object of the present invention to provide an alternative technique for monitoring and/or analysing biomechanical activity which reduces or mitigates some or all of these disadvantages.

According to one aspect, the present invention provides a wearable sensor apparatus comprising:

a motion sensor configured to sense two or three dimensional movement and orientation of the sensor;

a vibration sensor configured to sense acoustic vibrations; and

means for attaching the motion sensor and the vibration sensor to a body.

The acoustic vibrations may be, or may be represented by, bioacoustic signals. The acoustic vibrations may be bioacoustic vibrations. The motion sensor may comprise an inertial measurement unit. The vibration sensor may be configured to sense skeletal muscle vibrations. The inertial measurement unit may comprise one or more of an accelerometer, a gyroscope, and a magnetometer. The inertial measurement unit may be configured to sense rotation of the sensor body around at least one axis in space. The inertial measurement unit may be capable of sensing translational motion in three perpendicular axes and/or rotation about three perpendicular axes. The vibration sensor may comprise an acoustic pressure sensor. The vibration sensor may comprise one or more of an accelerometer, a microphone, and a piezoelectric transducer. The vibration sensor may comprise a volumetric chamber closed at one end by a flexible membrane, and a pressure transducer coupled to the chamber distal from the flexible membrane. The apparatus may comprise a barometer. The barometer may be configured to sense an ambient pressure. The vibration sensor and the barometer may be provided by a single sensor.

The apparatus may further comprise a data logging device coupled to receive motion signals from the motion sensor and acoustic signals, such as muscle vibration signals or mechanomyographic (MMG) muscle signals, from the vibration sensor, and to store said signals as a function of time. The apparatus may further comprise a classification processor configured to receive motion signals from the motion sensor and to receive the muscle signals (or mechanomyographic muscle signals) from the vibration sensor, and to classify a pattern of movement, or a posture of, at least one part of a body on which the wearable sensor apparatus is attached. The classification processor may be configured to use both the motion signals and the muscle vibration signals (or mechanomyographic muscle signals) to determine simultaneous patterns of movement, or postures of, multiple articulating parts of the body on which the wearable sensor apparatus is attached. The muscle vibration signals may be mechanomyographic muscle signals. The classification processor may be configured to classify a pattern of movement, or a posture of, at least one part of a body on which the wearable sensor apparatus is attached based on the mechanomyographic muscle signals.

The classification processor may be configured to separate the signals from the vibration sensor into windowed data. The classification processor may be configured to perform cluster analysis on the windowed data to determine a correlation between the signals from the vibration sensor and a type of activity.

The cluster analysis may comprise determining clusters of the windowed data and comparing one or more properties of the clusters with one or more threshold values. A property of a cluster may be an averaged value of a property of the windowed data within the cluster. The one or more properties comprise one or more of: gyroscopic magnitude, peak gyroscopic magnitude, ambient pressure, a cadence of a user (which may be a number of steps taken in a unit of time) and orientation of the motion sensor.

The apparatus may comprise a classification processor configured to receive motion signals from the motion sensor, to receive an ambient pressure signal from the barometer and to receive signals from the vibration sensor, and to classify a pattern of movement, or a posture of, at least one part of a body on which the wearable sensor apparatus is attached based on the received signals.

The apparatus may further comprise a classification processor configured to receive motion signals from the motion sensor and to receive muscle vibration signals, such as mechanomyographic muscle signals, from the vibration sensor, and to classify a pattern of movement, or a posture of, at least one part of a body on which the wearable sensor apparatus is attached, based on said motion signals, and to identify muscular activity used during said pattern of movement or posture. The classification processor may be further configured to determine whether or not the identified muscular activity conforms to a predetermined pattern consistent with the classified pattern of movement.

The apparatus may further comprise a classification processor configured to receive motion signals from the motion sensor and to receive acoustic signals from the vibration sensor, and to determine when the acoustic signals correspond to foetal movement.

The apparatus may further include an interface module configured to provide output control signals for a computer processor based on the output of the classification processor. The apparatus may further include an interface module configured to provide output control signals for a motive apparatus based on the output of the classification processor. The apparatus may further include a said motive apparatus. The motive apparatus may comprise a prosthesis or a robotic device.

According to another aspect, the invention provides a method of classifying a pattern of movement, or a posture of, at least one part of a human or animal subject comprising the steps of:

obtaining motion signals from a motion sensor attached to the subject, the motion signals being indicative of sensed movement of the subject in two or three dimensions; simultaneously obtaining vibration signals from a vibration sensor attached to the subject, the vibration signals being indicative of sensed skeletal muscle vibrations or other acoustic output from the subject; and using the motion signals and the sensed vibration signals, such as skeletal muscle vibration signals, to classify a pattern of movement, or a posture, of at least one part of the subject.

Other acoustic output from the subject may comprise acoustic output due to foetal movements. The method may further include using the motion signals and the sensed vibration signals to classify a pattern of movement, or a posture of, multiple articulating body parts of the subject. The method may further include using the motion signals and the sensed vibration signals to control movement of a computer processor or a motive apparatus. The vibration signals may be skeletal muscle vibration signals.

According to another aspect, the invention provides a method of assisting a human subject in rehabilitation comprising:

obtaining motion signals from a motion sensor attached to the subject, the motion signals being indicative of sensed movement of the subject in two or three dimensions; simultaneously obtaining vibration signals from a vibration sensor attached to the subject, the vibration signals being indicative of sensed skeletal muscle vibrations or other acoustic output from the subject, such as a heart rate of, or breathing from, the subject; and classifying a pattern of movement, or a posture, of at least one part of the subject using at least the motion signals; determining whether the pattern of movement, or posture, of the at least one part of the subject conforms to a predetermined pattern of the sensed skeletal muscle vibration signals; and providing an audible or visual feedback to the user dependent on whether the pattern of movement, or posture, of the at least one part of the subject is consistent with the predetermined pattern of the sensed skeletal muscle vibration signals.

According to another aspect, the invention provides a method of monitoring a maternal body for foetal movement comprising:

obtaining motion signals from a motion sensor attached to the maternal body, the motion signals being indicative of sensed movement of the subject in two or three dimensions; simultaneously obtaining acoustic vibration signals from one or more acoustic vibration sensors attached to the maternal body; and using at least the motion signals from the motion sensor to determine periods of maternal activity to attenuate or exclude maternal acoustic vibration signals from the vibration signals to thereby detect acoustic vibrations associated with foetal movement.

The acoustic vibration signals may be bioacoustic signals. The method may further comprise using acoustic vibration signals obtained from the acoustic vibration sensor to determine an orientation of a foetus. The method may further comprise using acoustic vibration signals obtained from the acoustic vibration sensor to generate a heart rate or a breathing rate from the maternal and/or foetal body.

Embodiments of the present invention will now be described by way of example and with reference to the accompanying drawings in which:

FIG. 1 shows a schematic diagram of a wearable activity sensor apparatus;

FIG. 2 shows a perspective view of a wearable activity sensor implemented in a knee brace;

FIG. 3 shows a schematic diagram of a wearable activity sensor apparatus including signal processing and analysis processing modules;

FIG. 4 shows x, y and z accelerations of a subject during a ten step walk and classification thereof;

FIG. 5 shows a subject wearing two activity sensors deployed on upper arm and forearm with real time feedback display;

FIG. 6 shows a graph of mechanomyography (MMG) data obtained during isometric contraction of the forearm with fist clenches;

FIG. 7 shows sensor data collected during a sitting to standing task from (a) electromyogram sensing; (b) accelerometer MMG sensing; (c) microphone MMG sensing; and (d) magnetic angular rate and gravity (MARG) sensing;

FIG. 8 shows both motion sensor data and mechanomyographic sensor data responsive to a subject walking;

FIG. 9 shows a graph of motion sensor classified stationary and moving periods overlaid on MMG data to further support activity classification;

FIG. 10 shows wearable activity sensors provided on both upper and lower leg of a subject together with sensor data received therefrom during standing and a knee-lift task;

FIG. 11 shows (a) acoustic sensor output, and (b) a gel-based vibration sensor output corresponding to foetal movement in a pregnant subject;

FIG. 12 shows a wearable activity monitor for detecting foetal movements;

FIG. 13 shows a flow chart of a method for classifying a subject&#39;s movement, including foetal movement within the subject;

FIG. 14 shows another flow chart of a method for classifying a gait, posture or movement of a subject;

FIG. 15 shows acoustic sensor output and maternal sensation data corresponding to foetal movement in pregnant subjects;

FIG. 16 shows (a) gyroscope and (b) corresponding muscle response data for a subject during a walking task; and

FIG. 17 illustrates (a) gyroscopic data, (b) accelerometer data with calculated magnitude, (c) unprocessed (raw) MMG data, (d) filtered MMG data, and (e) filtered and processed MMG for a subject during a walking task.

Mechanomyographic (MMG) muscle sensing exploits a low frequency vibration emitted by skeletal muscle the measurement of which does not require electrodes, gel or direct skin contact, unlike electromyographic sensing. This offers potential for much more efficient implementation in everyday use. Combining mechanomyographic muscle sensing with two- or three-dimensional movement sensing has been found to provide a considerable advance in human activity monitoring by combining human dynamics and muscle activity information. Bioacoustic sensors, such as MMG sensors, can readily be provided at relatively low cost, and can be packaged in a lightweight wearable package which is easy to attach to the subject&#39;s body without the use of complicated procedures for ensuring good electrical contact with the subject&#39;s skin.

In a general aspect, the techniques described here combine the use of inertial measurement units (IMUs) and acoustic sensors, e.g. bioacoustic sensors, such as mechanomyographic (MMG) sensors for muscle activity sensing or foetal movement monitoring. An arrangement described herein in relation to sensing a mechanomyographic muscle signal may also be used to sense bioacoustic signals or muscle vibration signals more generally.

FIG. 1 shows a schematic diagram of a wearable activity sensor 1 comprising a support structure 2 , a motion sensor 3 and a muscle vibration sensor 4 . The support structure 2 may be any suitable structure for attachment to the human or animal body at an appropriate location. In one example, shown in FIG. 2 , the support structure is a flexible, stretchable knee brace 20 with two motion sensors 3 sewn into pouches 21 with a muscle vibration sensor 4 having its membrane presented against the subject&#39;s skin on the inside surface of the support structure 20 . Use of a flexible support structure 2 such as a knee brace 20 or a tubular bandage allows generally unhindered movement of the subject and suitability for extended periods of wear, e.g. over a whole day or several days. Many other forms of wearable activity sensor are possible, for example integrated into articles of clothing, and for attachment to any suitable part of the body such as arms, legs, hands, trunk etc. The activity sensors can be positioned on the body at any suitable location from which limb or body motion can be sensed, and from which vibrations from any target muscle groups can be detected. In another example, shown in FIG. 10 and discussed in more detail later, activity sensors may be provided on both sides of the subject&#39;s knee, i.e. on the upper leg and on the lower leg. Such an arrangement provides motion signals specific to both upper and lower parts of the leg and muscle vibration signals from both upper and lower leg. Such an arrangement is particularly useful in monitoring activity relating to walking, standing, sitting and lying down, gait and posture analysis, etc, as will be discussed further below.

The motion sensor 3 preferably comprises one or more Inertial Measurement Units (IMUs) which may consist of tri-axis gyroscopes and accelerometers and Magnetic Angular Rate and Gravity (MARG) sensor arrays that also include tri-axis magnetometers. In a general aspect, however, any motion sensor capable of sensing two- or three-dimensional translational movement of the sensor body in space and rotation of the sensor body around at least one axis in space can be used. Preferably, the motion sensor should be capable of sensing translational motion in three perpendicular axes (forward/backward, up/down, left/right) and rotation about three perpendicular axes (pitch, yaw and roll). However, it will be understood that certain types of motion tracking may not require all six degrees of freedom. MARG sensors are preferred due to their low cost, small size, light weight and accuracy.

The mechanomyographic muscle vibration sensor preferably comprises a pressure sensor. MMG muscle vibrations are low frequency vibrations emitted by skeletal muscle, believed to be the mechanical activity of muscle generated by lateral oscillations of muscle fibres. The MMG vibration sensor collects signals indicative of mechanical attributes of contracting muscles, such as fatigue. Unlike EMG sensors, MMG sensors do not require gel or direct skin contact and can be reused and readily applied by an unskilled user, for example, because the precision of placement required for an MMG sensor may be less stringent than for an EMG sensor. The collection time of measurements using MMG sensors can be greater than that achieved using EMG sensors. The MMG vibration sensor may comprise one or more accelerometers, microphones, piezoelectric transducers, hydrophones or laser distance sensors. A preferred example is a pressure sensor in the form of a microphone sensor which provides very accurate results in detecting muscle activity and benefits by being little affected by motion noise. It is also easily integrated into a flexible support structure 2 as discussed above.

The MMG vibration sensor preferably provides a volumetric chamber with a membrane stretched over an opening in a housing of the chamber. A difference in volume of the volumetric chamber whenever the membrane is deformed by muscle vibration is detected using a microphone. When the membrane is placed over a subject&#39;s muscle, lateral contractions produce a physical change in the muscle&#39;s form, which in turn changes the membrane position and/or profile and creates a pressure change within the chamber. The microphone signals of particular interest lie between 1 and 256 Hz and are therefore preferably sampled at between 0.5 and 1 kHz. The signal may be boosted using an operational amplifier-based preamplifier which increases the power of the signal by a factor of approximately 21 times, specifically in the range of 1 Hz to 1 kHz which is sufficient for an MMG signal that has a dominant frequency in the range 25±2.5 Hz.

With further reference to FIG. 1 , the wearable activity sensor 1 is coupled to a microprocessor 5 configured to receive motion signals from the motion sensor 3 via communication link 6 a and to receive muscle vibration signals from the vibration sensor 4 via communication link 6 b . The communication links 6 a , 6 b can be wired links or wireless links. The microprocessor 5 may be coupled to a local memory 7 for logging data from the motion sensor 3 and the vibration sensor 4 . Thus, in a general aspect, the activity sensor may include a data logging device coupled to receive motion signals from the motion sensor 3 and mechanomyographic muscle signals from the vibration sensor 4 , and store these signals as a function of time. These functions can be provided by the microprocessor and memory 7 .

MMG data and motion data (e.g. from a MARG sensor) can preferably be collected simultaneously by microprocessor 5 . Sampling rates may be the same or different for the two sensor types. An exemplary sampling rate from a MARG sensor suitable for gait analysis is 16 Hz. The microprocessor 5 may have multiple channels to collect data from multiple motion sensors/multiple vibration sensors at once or multiple processors may be deployed, each dedicated to one or more sensor or sensor groups. The microprocessor(s) may also receive signals from other types of sensor if required, and/or add markers in the data for other sensed events from other sensor types. Other sensed events could include, for example, physiological phenomena detected from ECG data.

In one arrangement, the microprocessor 5 and/or the memory 7 are mounted on the flexible support structure 2 as an integral part of the activity sensor 1 , together with a suitable power supply. In another arrangement, the communication links

6 a , 6 b could comprise a short range wireless link such as Bluetooth or other near-field communication channel, and the microprocessor 5 and memory 7 could be located on a separate device. In one preferred example, the separate device could be a mobile telephone, smart phone or other personal computing device. Wired links using, for example, USB interface, are also possible.

In an alternative arrangement, the muscle vibration sensor 4 may be replaced with any other acoustic sensor, or bioacoustic sensor. In such arrangements, references to MMG signals may be replaced with corresponding references to acoustic signals. For example, other bioacoustic vibration analysis may be performed instead of muscle activity analysis. A bioacoustic sensor can collect signals indicative of foetal movements, as well as MMG signals. In a bioacoustic sensor constructed in a similar manner to the MMG sensor described above, vibration signals caused by foetal movements lead to a pressure change in the chamber when the membrane is placed on the abdomen of an expectant mother.

The activity sensor 1 has a wide range of applications of which selected ones are discussed below.

Rehabilitation

Using the activity sensor 1 , a patient&#39;s activities can be continuously monitored to allow new intervention treatments. Data can be logged from both motion tracking and physiological sensing for extended periods of time. Logging muscle activity provides vital cues in relation to the health and progress of a patient in response to rehabilitation. More precise data representative of a subject&#39;s natural movement and muscle activity over extended periods of time (e.g. an entire day or even a week) has the potential to provide a generational leap in patient monitoring and lead to an entire range of new rehabilitative treatments for conditions such as stroke and neuromuscular disorders.

Applications include pre- and post-knee surgery rehabilitation monitoring, posture observation, fall detection, prosthetic control and manipulation, and general human activity analysis, predominantly in geriatrics and paediatrics. Combining muscle activity and motion data opens new horizons in anthropometrics by increasing knowledge of conditions with the addition of two different forms of biomechanical information. Motion and muscle activity give important information separately, but when synergistically combined provide better information for medical treatment and clinical intervention.

Gait/Movement/Posture Classification

Data collected over a prolonged period of time, both inertial and muscular, contains a lot of information and features that are specific to an individual subject. However, without proper analysis, these characteristics can be hard to distinguish from noise and irrelevant data. The activity sensor may therefore include a classification processor in order to analyse the collected data and extract the information and features that are specific to the subject. The classification processor is configured to receive the motion signals from the motion sensor 3 and may also receive the mechanomyographic muscle signals from the vibration sensor 4 , and to classify a pattern of movement, or a posture, of at least one part of the body. Typically, this would be the part of the body to which the wearable sensor is attached, but in some circumstances it may be possible to determine movement, or a posture of, other parts of the body more remote from the wearable sensor. The classification processor may be configured to distinguish between standing, sitting, reclining and walking activities and various different postures. Thus, the classification processor may be configured to identify, from a library or database of known activity and/or posture types, one or more activities and/or postures corresponding to the received signals. The classification processor may be configured with algorithms adapted to detect signature signals indicative of predetermined activities and/or postures.

In one algorithm, applied to a knee brace sensor apparatus of FIG. 2 mounted on the upper leg above the knee, sensing the direction of gravity may be used. During a standing posture, the x plane is pointing towards the ground, which gives a reading around −1±0.1 g, and the other planes (y and z) give a reading of 0±0.1 g. When sitting, the y plane is now pointing towards the ground, which will give a reading of around −1±0.1 g and the other planes (x and z) read 0±0.1 g. Muscular activity data can then be correlated with the particular standing posture.

The classification of walking can be detected by, for example, combining accelerometer data from each plane to determine magnitude using equation 1, where i is the current sample and n is the total number of samples; x, y and z represent the accelerations from each respective plane:

magnitude=Σ i=1 n ( x i 2 +y i 2 +z i 2 ) 0.5   (1)

A threshold is determined per subject by a controlled walking task of five steps in a straight line. The data is retrieved and the threshold is determined offline by trial and error until the computed value also determines the subject to have walked five steps. Stationary states are determined to be whenever the magnitude is beneath the threshold, as per equation 2.

stationary=magnitude&lt;threshold  (2)

Any period above the threshold is deemed active. Data is then split into windows of one second and the average of each window is used to determine what state the subject is in by checking the stationary periods against the averaged windows. One second windows allow each step to be recognised, as the subject&#39;s leg is stationary between strides, therefore the calculation doubles as a pedometer to determine number of steps taken. If the data window is deemed to be ‘active’ then the gait segment is classified as walking. However, if the data window states the subject is stationary, the calculation then determines which plane gravity is in to determine standing or sitting. In the event the subject is lying on his or her side or gravity is in a plane unrecognisable to the algorithm, the subject&#39;s state may be put into an ‘other’ category which can otherwise be classified differently.

FIG. 4 shows analysis of a ten step walk correctly determining active periods using the threshold technique described above. Solid line 41 represents x-axis acceleration, dot- dash line 42 represents y-axis acceleration, and dashed line 43 represents z-axis acceleration. The combined acceleration data corresponding to magnitude is represented by line 44 . The predefined threshold has been set at 0.3 g, and a walking motion may be classified when the combined data exceeds 0.3 g. This classification is indicated by solid line 45 with the standing posture indicated at +1 ( portions 45 a ) and walking motion indicated at 0 ( portions 45 b ). The method was determined to be approximately 90% accurate during walking calibration tests over 15 tests.

Although the arrangement described above only requires the motion sensor data to classify a generic pattern of movement, or posture, of at least one part of the body (e.g. walking, standing, sitting), the muscle vibration sensing data may be used in conjunction therewith to further refine the classification process.

For example, the muscle vibration sensing data may be used to detect and classify other movements of the body separate or distinct from the major patterns of movement classified using the motion sensor data as further discussed below.

Thus, in a general aspect, the classification processor may be configured to use both the motion signals and mechanomyographic muscle signals to determine simultaneous patterns of movement, or postures of, multiple articulating parts of the body on which the wearable sensor apparatus is attached.

In another arrangement, the wearable activity sensor apparatus may be configured as a belt. The belt-based sensors may be configured to monitor balance, posture and key muscle activity; facilitate adoption of new movement patterns; and engage patients with direct feedback on their progress in ameliorating low back pain. The apparatus may be configured to monitor and facilitate user activation of key postural muscles associated with back dysfunction, principally lower abdominal and gluteal muscles, and activity monitors that will track balance and posture. By continuously measuring posture and providing real time biofeedback, the wearable sensor apparatus will provide the ability to facilitate active postural correction.

FIG. 3 shows a schematic diagram of an activity monitoring system configured to provide analysis and feedback in relation to muscle usage correlated with subject motion.

A wearable activity sensor 30 comprises a support structure 31 in the form of a belt, which supports a motion sensor 3 and a muscle vibration sensor 4 such as previously described in connection with FIG. 1 . The sensors

33 , 34 are coupled to a processor including a movement analysis module 32 and a muscle activity analysis module 33 . The movement analysis module 32 may be configured to perform such functions as identification of body and/or limb movements, and identification of postures, as previously described. The muscle activity analysis module 33 may be configured to identify individual muscle activities and identify muscles or muscle groups involved in sensed vibration events. The motion and muscle activity data from modules

32 , 33 are combined in a sensor data fusion module 34 so that individual movement and/or posture ‘events’ can be aligned in time. A correlation process 35 may be configured to correlate the various movement and/or posture events with muscle activity events.

The correlated data from correlation process 35 may then be used to provide one or more possible feedback processes 36 .

In one example, the feedback process 36 could be provision of a visual display of motion and related muscle activity, e.g. charted over time, for review by a clinician or the subject. The feedback process could be provision of motion and related muscle activity monitored over periods of time, e.g. successive days, to indicate whether muscle activity for predetermined postures and/or activity classifications changes over time. Such feedback process could be used to indicate an improvement, or deterioration, in activity or posture attributes over time.

In another example, the feedback process 36 could provide real time analysis directly to the subject wearing the sensor, by visual or audible feedback. For example, the processor may comprise a classification processor which classifies a pattern of movement, or a posture, of at least one part of the body of the subject, based on the motion signals, and identifies the muscular activity associated with that pattern of movement or posture, and determines whether the sensed muscular activity conforms to a predetermined pattern that is consistent with the classified pattern of movement. This predetermined pattern may represent an ideal or optimum muscular activity for a given sensed motion or posture. In the event that the sensed muscular activity does not conform to an ideal or optimum pattern of muscular activity for the sensed motion or posture, the feedback process 36 may be configured to provide a real time alert to the subject wearing the activity sensor. The real time alert may be any indication useful to assist the subject, such as a warning of poor posture or poor movement execution, an instruction to improve posture, an instruction to alter position, an instruction to engage in a specified pattern of movement or similar.

FIG. 14 illustrates a flow chart of another algorithm 1400 for classifying a gait, posture or movement of a subject. The algorithm 1400 combines motion and muscle activity in an unsupervised classification algorithm. The algorithm 1400 enables eight commonly performed activities to be identified: walking, running, ascending stairs, descending stairs, ascending in an elevator, descending in an elevator, standing, and lying down. A ninth “activity” containing noise and other unclassified activities data is also categorized.

The algorithm may be performed by a classification processor that is configured to receive:

motion signals from a motion sensor, ambient pressure signals from a barometer; and bioacoustic signals from a vibration sensor.

The classification processor is configured to classify, based on the received signals, a pattern of movement, or a posture of, at least one part of a body on which the sensors are attached.

The algorithm 1400 processes data in two stages

1402 , 1404 . In a gross group clustering stage 1402 , a data or time window is split into one of three groups, each of which relates to a different class of activities; stationary activities 1424 (such as standing, lying down, and elevator), dynamic activities 1426 (such as walking, running, and noise), and dynamic-altitude activities 1428 (such as travelling on stairs). In a subsequent activity classification stage 1404 , windows from each cluster group are further classified into one of nine more specific activities.

At the start of the gross group clustering stage 1402 , data from each sensor type (bioacoustic and accelerometer) is windowed 1430 . The window size may be selected as a number of data samples or a period of time. For example, a window size of 200 samples (four seconds of data at 50 Hz) may be used for inertial data. Such a window may be resized in order to obtain the same window size for the other sensors sampled at a different rate. For example, a four sample window may be used for a barometer operating at 1 Hz and a 4000 sample window may be used for a MMG operating at 1 kHz. It has been found that windows with a 50% overlap produce suitably good results at 50 Hz for the inertial data.

Inertial data can, optionally, be smoothed using a moving average in order to reduce the effect of transient noise on the output of the algorithm. An example moving average is:

ys n

=

1

2 ⁢ N

+ 1

⁢

(

y

(

n + N

)

+

y

(

n + N - 1

)

+ … +

y

(

n - N

)

)

,

where ys n is the smoothed value for the n th data point, N is the number of neighbouring data points on either side of ys n , and 2N+1 is the span, such as 15 data points. Windowing 1430 can also be performed after smoothing has taken place.

Features of the windowed data are determined after the optional smoothing step. In this example, four features (mean, standard deviation, power, and covariance) are determined for each of the three axes of the accelerometer (x, y, and z) resulting in twelve parameters per window. The features can be calculated using the equations below.

μ

=

1

N

⁢

∑

n

=

1

N

⁢

⁢

y

n

σ

=

1

N

-

1

⁢

∑

n

=

1

N

⁢

⁢

(

y

n

-

μ

)

2

power

=

∑

n

=

1

N

⁢

⁢

y

n

2

N

COV

⁡

(

a

,

b

)

=

∑

n

=

1

N

⁢

⁢

(

a

n

-

μ

a

</mrow

CLAIMS

Claims ( 27 )

The invention claimed is:

1. A wearable sensor apparatus comprising:

a motion sensor configured to sense two or three dimensional movement and orientation of the motion sensor;

a vibration sensor configured to sense acoustic vibrations, wherein the vibration sensor comprises an acoustic pressure sensor;

means for attaching the motion sensor and the vibration sensor to a body; and

a classification processor configured to receive motion signals from the motion sensor and to receive muscle vibration signals from the vibration sensor, and to classify a pattern of movement, or a posture of, at least one part of a body on which the wearable sensor apparatus is attached, based on said motion signals, and to identify muscular activity used during said pattern of movement or posture based on the sensed acoustic vibrations.

2. The apparatus of claim 1 in which the motion sensor comprises an inertial measurement unit.

3. The apparatus of claim 1 in which the vibration sensor is configured to sense skeletal muscle vibrations.

4. The apparatus of claim 2 in which the inertial measurement unit comprises one or more of an accelerometer, a gyroscope, and a magnetometer.

5. The apparatus of claim 2 in which the inertial measurement unit is configured to sense rotation of the motion sensor body around at least one axis in space.

6. The apparatus of claim 1 in which the vibration sensor comprises a volumetric chamber closed at one end by a flexible membrane, and a pressure transducer coupled to the chamber distal from the flexible membrane.

7. The apparatus of claim 1 comprising a barometer configured to sense an ambient pressure.

8. The apparatus of claim 1 further comprising a data logging device coupled to receive motion signals from the motion sensor and muscle vibration signals from the vibration sensor, and to store said signals as a function of time.

9. The apparatus of claim 1 wherein the signals from the vibration sensor are mechanomyographic muscle signals and the classification processor is configured to classify a pattern of movement, or a posture of, at least one part of a body on which the wearable sensor apparatus is attached based on the mechanomyographic muscle signals.

10. The apparatus of claim 1 in which the classification processor is configured to use both the motion signals and the muscle vibration signals to determine simultaneous patterns of movement, or postures of, multiple articulating parts of the body on which the wearable sensor apparatus is attached.

11. The apparatus of claim 1 wherein the classification processor is configured to:

separate the signals from the vibration sensor into windowed data;

perform cluster analysis on the windowed data to determine a correlation between the signals from the vibration sensor and a type of activity.

12. The apparatus of claim 11 wherein the cluster analysis comprises determining clusters of the windowed data and comparing one or more properties of the clusters with a corresponding threshold value.

13. The apparatus of claim 12 wherein the one or more properties comprise one or more of: gyroscopic magnitude, peak gyroscopic magnitude, ambient pressure, a cadence of a user and orientation of the motion sensor.

14. The apparatus of claim 7 , wherein the classification processor configured to receive motion signals from the motion sensor, to receive an ambient pressure signal from the barometer and to receive signals from the vibration sensor, and to classify a pattern of movement, or a posture of, at least one part of a body on which the wearable sensor apparatus is attached based on the received signals.

15. The apparatus of claim 8 wherein the signals from the muscle vibration signals are mechanomyographic muscle signals.

16. The apparatus of claim 1 in which the classification processor is further configured to determine whether or not the identified muscular activity conforms to a predetermined pattern consistent with the classified pattern of movement.

17. The apparatus of claim 1 further comprising:

a classification processor configured to receive motion signals from the motion sensor and to receive acoustic signals from the vibration sensor, and to determine when the signals correspond to foetal movement.

18. The apparatus of claim 1 further including an interface module configured to provide output control signals for a computer processor based on the output of the classification processor.

19. The apparatus of claim 1 further including an interface module configured to provide output control signals for a motive apparatus based on the output of the classification processor.

20. The apparatus of claim 19 further including a said motive apparatus.

21. The apparatus of claim 20 in which the motive apparatus comprises a prosthesis or a robotic device.

22. The apparatus of claim 10 wherein the signals from the muscle vibration signals are mechanomyographic muscle signals.

23. The apparatus of claim 1 wherein the signals from the muscle vibration signals are mechanomyographic muscle signals.

24. The apparatus of claim 10 further including an interface module configured to provide output control signals for a computer processor based on the output of the classification processor.

25. The apparatus of claim 10 further including an interface module configured to provide output control signals for a motive apparatus based on the output of the classification processor.

26. The apparatus of claim 25 further including said motive apparatus.

27. The apparatus of claim 26 in which the motive apparatus comprises a prosthesis or a robotic device.

US15/034,165

2013-11-04

2014-11-04

Biomechanical activity monitoring

Active

2035-09-15

US10335080B2

( en )

Applications Claiming Priority (3)

Application Number

Priority Date

Filing Date

Title

GB1319434.5A

GB2519987B

( en )

2013-11-04

2013-11-04

Biomechanical activity monitoring

GB1319434.5

2013-11-04

PCT/GB2014/053276

WO2015063520A1

( en )

2013-11-04

2014-11-04

Biomechanical activity monitoring

Related Parent Applications (1)

Application Number

Title

Priority Date

Filing Date

PCT/GB2014/053276

A-371-Of-International

WO2015063520A1

( en )

2013-11-04

2014-11-04

Biomechanical activity monitoring

Related Child Applications (1)

Application Number

Title

Priority Date

Filing Date

US16/459,349

Continuation-In-Part

US20190320944A1

( en )

2013-11-04

2019-07-01

Biomechanical activity monitoring

Publications (2)

Publication Number

Publication Date

US20160262687A1

US20160262687A1 ( en )

2016-09-15

US10335080B2

true

US10335080B2 ( en )

2019-07-02

Family

ID=49767613

Family Applications (1)

Application Number

Title

Priority Date

Filing Date

US15/034,165

Active

2035-09-15

US10335080B2

( en )

2013-11-04

2014-11-04

Biomechanical activity monitoring

Country Status (7)

Country

Link

US

( 1 )

US10335080B2

( en )

EP

( 1 )

EP3065628B1

( en )

CN

( 1 )

CN106061369A

( en )

ES

( 1 )

ES2940664T3

( en )

GB

( 1 )

GB2519987B

( en )

PL

( 1 )

PL3065628T3

( en )

WO

( 1 )

WO2015063520A1

( en )

Cited By (2)

* Cited by examiner, † Cited by third party

Publication number

Priority date

Publication date

Assignee

Title

US20200118334A1

( en )

*

2018-01-25

2020-04-16

Ctrl-Labs Corporation

Techniques for anonymizing neuromuscular signal data

US12426804B1

( en )

*

2025-02-24

2025-09-30

King Saud University

System and method for alleviating lower back pain

Families Citing this family (149)

* Cited by examiner, † Cited by third party

Publication number

Priority date

Publication date

Assignee

Title

US20100201512A1

( en )

2006-01-09

2010-08-12

Harold Dan Stirling

Apparatus, systems, and methods for evaluating body movements

GB201317746D0

( en )

2013-10-08

2013-11-20

Smith &amp; Nephew

PH indicator

US20190212311A1

( en )

2013-01-11

2019-07-11

Smith &amp; Nephew Plc

Ph and moisture indicator devices and formulations

US10188309B2

( en )

2013-11-27

2019-01-29

North Inc.

Systems, articles, and methods for electromyography sensors

US20150124566A1

( en )

2013-10-04

2015-05-07

Thalmic Labs Inc.

Systems, articles and methods for wearable electronic devices employing contact sensors

US12504816B2

( en )

2013-08-16

2025-12-23

Meta Platforms Technologies, Llc

Wearable devices and associated band structures for sensing neuromuscular signals using sensor pairs in respective pods with communicative pathways to a common processor

US11921471B2

( en )

2013-08-16

2024-03-05

Meta Platforms Technologies, Llc

Systems, articles, and methods for wearable devices having secondary power sources in links of a band for providing secondary power in addition to a primary power source

US10042422B2

( en )

2013-11-12

2018-08-07

Thalmic Labs Inc.

Systems, articles, and methods for capacitive electromyography sensors

US9880632B2

( en )

2014-06-19

2018-01-30

Thalmic Labs Inc.

Systems, devices, and methods for gesture identification

HK1203120A2

( en )

*

2014-08-26

2015-10-16

Well Being Digital Limited

A gait monitor and a method of monitoring the gait of a person

US11534104B2

( en )

2014-10-29

2022-12-27

Bloom Technologies NV

Systems and methods for contraction monitoring and labor detection

US10456074B2

( en )

*

2014-10-29

2019-10-29

Bloom Technologies NV

Method and device for contraction monitoring

WO2016077489A1

( en )

*

2014-11-11

2016-05-19

Innovaura Corporation

Heart rate monitor

US10716517B1

( en )

*

2014-11-26

2020-07-21

Cerner Innovation, Inc.

Biomechanics abnormality identification

US9713430B2

( en )

*

2015-03-16

2017-07-25

Nuvo Group Ltd.

Acoustic sensors for abdominal fetal cardiac activity detection

WO2016168610A1

( en )

2015-04-15

2016-10-20

Nike, Inc.

Activity monitoring device with assessment of exercise intensity

JP6545516B2

( en )

*

2015-04-24

2019-07-17

株式会社オプティãƒ

Behavior analysis server, behavior analysis method and program for behavior analysis server

US11039782B2

( en )

2015-05-27

2021-06-22

Georgia Tech Research Corporation

Wearable technologies for joint health assessment

EP3304256B1

( en )

2015-05-29

2021-09-01

Nike Innovate C.V.

Activity monitoring device with assessment of exercise intensity

US20160374566A1

( en )

*

2015-06-23

2016-12-29

Microsoft Technology Licensing, Llc

Sample-count-based sensor data calculations

US11185275B2

( en )

2015-09-02

2021-11-30

Khalifa Univeristy of Science and Technology

Low cost fetal phonocardiogram

CN105361889B

( en )

*

2015-09-06

2017-12-29

江苏矽望电子科技有限公司

A kind of detection method of the optimal movement of the foetus counting position based on acceleration transducer

US20210196154A1

( en )

*

2015-09-21

2021-07-01

Figur8, Inc.

Body part consistency pattern generation using motion analysis

US10317875B2

( en )

*

2015-09-30

2019-06-11

Bj Services, Llc

Pump integrity detection, monitoring and alarm generation

CN105411594A

( en )

*

2015-11-02

2016-03-23

广州阿路比电子科技有限公司

Novel sensor used for collecting human motion signals

US10055948B2

( en )

*

2015-11-30

2018-08-21

Nike, Inc.

Apparel with ultrasonic position sensing and haptic feedback for activities

US11089146B2

( en )

*

2015-12-28

2021-08-10

The Mitre Corporation

Systems and methods for rehabilitative motion sensing

US10004461B2

( en )

*

2015-12-30

2018-06-26

Automotive Research &amp; Testing Center

Physiological signal processing system and method for filtering noise generated by the same

RU168584U1

( en )

*

2016-01-11

2017-02-09

Алексей Викторович Горбунов

DEVICE FOR REGISTRATION OF HUMAN MOTOR ACTIVITY

KR101759621B1

( en )

*

2016-02-22

2017-07-21

주식회사 지파워

Behavior analysis system and analysis device using sensing of movement pattern and sound pattern using the same

CN105726048A

( en )

*

2016-05-05

2016-07-06

郑州大学第一附属医院

Functional exercise monitoring device for common orthopedic disease

US20180284758A1

( en )

2016-05-09

2018-10-04

StrongForce IoT Portfolio 2016, LLC

Methods and systems for industrial internet of things data collection for equipment analysis in an upstream oil and gas environment

US10983507B2

( en )

2016-05-09

2021-04-20

Strong Force Iot Portfolio 2016, Llc

Method for data collection and frequency analysis with self-organization functionality

US11774944B2

( en )

2016-05-09

2023-10-03

Strong Force Iot Portfolio 2016, Llc

Methods and systems for the industrial internet of things

US11327475B2

( en )

2016-05-09

2022-05-10

Strong Force Iot Portfolio 2016, Llc

Methods and systems for intelligent collection and analysis of vehicle data

CA3023772A1

( en )

2016-05-13

2017-11-16

Smith &amp; Nephew Plc

Sensor enabled wound monitoring and therapy apparatus

EP3257437A1

( en )

2016-06-13

2017-12-20

Friedrich-Alexander-Universität Erlangen-Nürnberg

Method and system for analyzing human gait

US11237546B2

( en )

2016-06-15

2022-02-01

Strong Force loT Portfolio 2016, LLC

Method and system of modifying a data collection trajectory for vehicles

US20170360378A1

( en )

*

2016-06-16

2017-12-21

Ghassan S. Kassab

Systems and methods for identifying fetal movements in an audio/visual data feed and using the same to assess fetal well-being

US11216069B2

( en )

2018-05-08

2022-01-04

Facebook Technologies, Llc

Systems and methods for improved speech recognition using neuromuscular information

US12554325B2

( en )

2016-07-25

2026-02-17

Meta Platforms Technologies, Llc

Methods and apparatuses for low latency body state prediction based on neuromuscular data

EP3487402B1

( en )

2016-07-25

2021-05-05

Facebook Technologies, LLC

Methods and apparatus for inferring user intent based on neuromuscular signals

WO2018022658A1

( en )

2016-07-25

2018-02-01

Ctrl-Labs Corporation

Adaptive system for deriving control signals from measurements of neuromuscular activity

EP3487595A4

( en )

2016-07-25

2019-12-25

CTRL-Labs Corporation

SYSTEM AND METHOD FOR MEASURING MOVEMENTS OF ARTICULATED RIGID BODIES

US11331045B1

( en )

2018-01-25

2022-05-17

Facebook Technologies, Llc

Systems and methods for mitigating neuromuscular signal artifacts

WO2018022602A1

( en )

*

2016-07-25

2018-02-01

Ctrl-Labs Corporation

Methods and apparatus for predicting musculo-skeletal position information using wearable autonomous sensors

US20180028109A1

( en )

*

2016-07-27

2018-02-01

Andrew TESNOW

System and method for a wearable knee injury prevention

CN106308760A

( en )

*

2016-08-15

2017-01-11

中国科学院深圳先进技术研究院

Remote monitoring device, remote monitoring system and wearable equipment for remote monitoring

CN106491115A

( en )

*

2016-10-28

2017-03-15

歌尔科技有限公司

A kind of method and apparatus for realizing phonetic incepting and heart rate detection

CN106344023B

( en )

*

2016-11-10

2020-02-11

重庆邮电大学

Unsteady state respiratory wave detection device based on atmospheric pressure and acceleration

CN106491116A

( en )

*

2016-11-11

2017-03-15

深圳市响音科技有限公司

A kind of passive type fetal rhythm detecting system and detection method

CN106821308A

( en )

*

2016-11-15

2017-06-13

深圳大学

A method and device for evaluating contraction activity under stimulation of skeletal muscle

CN107126302B

( en )

*

2017-02-15

2020-05-22

上海术理智能科技有限公司

Upper and lower limb movement simulation processing method

US11793461B2

( en )

2017-03-07

2023-10-24

Motionize Israel Ltd.

Football smart footwear with automatic personal and team performance statistics extraction

WO2018162732A1

( en )

2017-03-09

2018-09-13

Smith &amp; Nephew Plc

Apparatus and method for imaging blood in a target region of tissue

JP7091356B2

( en )

2017-03-09

2022-06-27

スミス アンド ネフュー ピーエルシー

Devices, devices, and methods for determining skin perfusion pressure

WO2018162736A1

( en )

2017-03-09

2018-09-13

Smith &amp; Nephew Plc

Wound dressing, patch member and method of sensing one or more wound parameters

CN110650713B

( en )

2017-04-11

2022-06-24

史密夫及内修公开有限公司

Component positioning and stress relief for sensor-enabled wound dressings

CN110870021B

( en )

2017-05-15

2024-12-20

史密夫及内修公开有限公司

Negative pressure wound therapy system using Euler video zoom

CN110650681B

( en )

2017-05-15

2023-08-15

布鲁姆技术公司

System and method for monitoring fetal health

JP7272962B2

( en )

2017-05-15

2023-05-12

スミス アンド ネフュー ピーエルシー

wound analyzer

CN108937907A

( en )

*

2017-05-26

2018-12-07

北京小米移动软件有限公司

The acquisition method and device of heart rate

US11633153B2

( en )

2017-06-23

2023-04-25

Smith &amp; Nephew Plc

Positioning of sensors for sensor enabled wound monitoring or therapy

WO2019008497A1

( en )

*

2017-07-04

2019-01-10

Moradi Ali

System and method for muscle movements detection

WO2019016759A1

( en )

2017-07-19

2019-01-24

Bloom Technologies NV

Monitoring uterine activity and assessing pre-term birth risk

GB201804502D0

( en )

2018-03-21

2018-05-02

Smith &amp; Nephew

Biocompatible encapsulation and component stress relief for sensor enabled negative pressure wound therapy dressings

GB201809007D0

( en )

2018-06-01

2018-07-18

Smith &amp; Nephew

Restriction of sensor-monitored region for sensor-enabled wound dressings

KR102830583B1

( en )

2017-08-02

2025-07-07

스트롱 포스 아이오티 포트폴리오 2016, 엘엘씨

Methods and systems for detection in industrial Internet of Things data collection environments with large data sets

US10921801B2

( en )

2017-08-02

2021-02-16

Strong Force loT Portfolio 2016, LLC

Data collection systems and methods for updating sensed parameter groups based on pattern recognition

CA3072006A1

( en )

2017-08-10

2019-02-14

Smith &amp; Nephew Plc

Positioning of sensors for sensor enabled wound monitoring or therapy

GB201804971D0

( en )

2018-03-28

2018-05-09

Smith &amp; Nephew

Electrostatic discharge protection for sensors in wound therapy

WO2019048624A1

( en )

2017-09-10

2019-03-14

Smith &amp; Nephew Plc

Systems and methods for inspection of encapsulation and components in sensor equipped wound dressings

GB201718870D0

( en )

2017-11-15

2017-12-27

Smith &amp; Nephew Inc

Sensor enabled wound therapy dressings and systems

GB201718859D0

( en )

2017-11-15

2017-12-27

Smith &amp; Nephew

Sensor positioning for sensor enabled wound therapy dressings and systems

EP3687380A1

( en )

2017-09-27

2020-08-05

Smith &amp; Nephew plc

Ph sensing for sensor enabled negative pressure wound monitoring and therapy apparatuses

WO2019072531A1

( en )

2017-09-28

2019-04-18

Smith &amp; Nephew Plc

Neurostimulation and monitoring using sensor enabled wound monitoring and therapy apparatus

EP3697297A4

( en )

2017-10-19

2020-12-16

Facebook Technologies, Inc.

SYSTEMS AND METHODS FOR IDENTIFYING BIOLOGICAL STRUCTURES ASSOCIATED WITH NEUROMUSCULAR SOURCE SIGNALS

US20200210689A1

( en )

*

2017-10-20

2020-07-02

Mindfio Limited

A system and a method for analyzing a behavior or an activity of an object

EP3709943B1

( en )

2017-11-15

2024-07-24

Smith &amp; Nephew PLC

Integrated sensor enabled wound monitoring and/or therapy dressings and systems

TWI669681B

( en )

*

2017-11-30

2019-08-21

財團法人資訊工業策進會

Electronic calculatiing apparatus, system, and method for providing body posture health information

US11961494B1

( en )

2019-03-29

2024-04-16

Meta Platforms Technologies, Llc

Electromagnetic interference reduction in extended reality environments

US10937414B2

( en )

2018-05-08

2021-03-02

Facebook Technologies, Llc

Systems and methods for text input using neuromuscular information

US11493993B2

( en )

2019-09-04

2022-11-08

Meta Platforms Technologies, Llc

Systems, methods, and interfaces for performing inputs based on neuromuscular control

EP3743892B1

( en )

2018-01-25

2025-03-05

Meta Platforms Technologies, LLC

Visualization of reconstructed handstate information

EP3743901A4

( en )

*

2018-01-25

2021-03-31

Facebook Technologies, Inc.

REAL-TIME PROCESSING OF HAND REPRESENTATION MODEL ESTIMATES

US12579768B2

( en )

2018-01-25

2026-03-17

Meta Platforms Technologies, Llc

Wearable electronic devices, extended reality systems including neuromuscular sensors, and methods for generating text from speech input and modifying the generated text based on neuromuscular data

US11907423B2

( en )

2019-11-25

2024-02-20

Meta Platforms Technologies, Llc

Systems and methods for contextualized interactions with an environment

CN111902077B

( en )

2018-01-25

2023-08-04

元平台技术有限公司

Calibration techniques for hand state representation modeling using neuromuscular signals

US11150730B1

( en )

2019-04-30

2021-10-19

Facebook Technologies, Llc

Devices, systems, and methods for controlling computing devices via neuromuscular signals of users

WO2019147958A1

( en )

2018-01-25

2019-08-01

Ctrl-Labs Corporation

User-controlled tuning of handstate representation model parameters

US10817795B2

( en )

2018-01-25

2020-10-27

Facebook Technologies, Llc

Handstate reconstruction based on multiple inputs

US11481030B2

( en )

2019-03-29

2022-10-25

Meta Platforms Technologies, Llc

Methods and apparatus for gesture detection and classification

US10592001B2

( en )

2018-05-08

2020-03-17

Facebook Technologies, Llc

Systems and methods for improved speech recognition using neuromuscular information

WO2019226259A1

( en )

*

2018-05-25

2019-11-28

Ctrl-Labs Corporation

Methods and apparatus for providing sub-muscular control

CN112261907A

( en )

2018-05-29

2021-01-22

脸谱科技有限责任公司

Noise reduction shielding technology in surface electromyogram signal measurement and related system and method

CN112585600A

( en )

2018-06-14

2021-03-30

脸谱科技有限责任公司

User identification and authentication using neuromuscular signatures

EP3593959B1

( en )

*

2018-07-10

2022-09-28

Tata Consultancy Services Limited

Method and system for online non-intrusive fatigue- state detection in a robotic co-working environment

US11045137B2

( en )

2018-07-19

2021-06-29

Facebook Technologies, Llc

Methods and apparatus for improved signal robustness for a wearable neuromuscular recording device

US12167895B2

( en )

*

2018-08-02

2024-12-17

The Johns Hopkins University

Safety feature for use with robotically manipulated endoscopes and other tools in otolaryngology and neurosurgery

EP3836836B1

( en )

2018-08-13

2024-03-20

Meta Platforms Technologies, LLC

Real-time spike detection and identification

GB201814011D0

( en )

2018-08-29

2018-10-10

Smith &amp; Nephew

Componet positioning and encapsulation for sensor enabled wound dressings

US11273357B2

( en )

2018-08-30

2022-03-15

International Business Machines Corporation

Interactive exercise experience

EP3843617B1

( en )

2018-08-31

2023-10-04

Facebook Technologies, LLC.

Camera-guided interpretation of neuromuscular signals

US20200077906A1

( en )

*

2018-09-07

2020-03-12

Augusta University Research Institute, Inc.

Method and System for Monitoring Brain Function and Intracranial Pressure

WO2020053290A1

( en )

2018-09-12

2020-03-19

Smith &amp; Nephew Plc

Device, apparatus and method of determining skin perfusion pressure

EP3853698A4

( en )

2018-09-20

2021-11-17

Facebook Technologies, LLC

NEUROMUSCULAR TEXT, WRITING AND DRAWING ENTRY IN AUGMENTED REALITY SYSTEMS

US10921764B2

( en )

2018-09-26

2021-02-16

Facebook Technologies, Llc

Neuromuscular control of physical objects in an environment

WO2020064937A1

( en )

2018-09-28

2020-04-02

T.J.Smith And Nephew,Limited

Optical fibers for optically sensing through wound dressings

EP3860527B1

( en )

2018-10-05

2024-07-10

Meta Platforms Technologies, LLC

Use of neuromuscular signals to provide enhanced interactions with physical objects in an augmented reality environment

GB201816838D0

( en )

2018-10-16

2018-11-28

Smith &amp; Nephew

Systems and method for applying biocompatible encapsulation to sensor enabled wound monitoring and therapy dressings

CN113423341B

( en )

2018-11-27

2024-12-03

元平台技术有限公司

Method and apparatus for automatic calibration of wearable electrode sensor systems

FR3089319A1

( en )

*

2018-12-04

2020-06-05

Orange

Method for evaluating the bodily activity of a user

GB201820927D0

( en )

2018-12-21

2019-02-06

Smith &amp; Nephew

Wound therapy systems and methods with supercapacitors

GB201901242D0

( en )

2019-01-30

2019-03-20

Smith &amp; Nephew

Optical sensing systems and methods for sensing enabled wound dressings and systems

JP7528097B2

( en )

2019-01-30

2024-08-05

スミス アンド ネフュー ピーエルシー

Sensor-integrated covering material and system

US10905383B2

( en )

2019-02-28

2021-02-02

Facebook Technologies, Llc

Methods and apparatus for unsupervised one-shot machine learning for classification of human gestures and estimation of applied forces

GB2597148B

( en )

2019-03-18

2022-12-21

Smith &amp; Nephew

Design rules for sensor integrated substrates

<span itemprop="p

Related documents

• Generating augmented reality experiences utilizing physical objects to … — Meta Platforms Technologies, Llc (US20260037214A1)#653960
• Virtual reality system using multiple force arrays for a solver — DeepMotion, Inc. (US10657696B2)#653961
• Tracking position of device inside-out for virtual reality interactivity — Sony Interactive Entertainment America Llc (US10535153B2)#653962
• Augmented reality system with spatialized audio tied to user manipulated … — Magic Leap, Inc. (US11231770B2)#653963
• Big data-based autonomous flight drone system and autonomous flight method … — Rgblab Co., Ltd. (US20210390867A1)#653964