Ambulatory system for measuring and monitoring physical activity and risk of falling and for automatic fall detection

• US 8,206,325 B1
• Filed: 10/12/2008
• Issued: 06/26/2012
• Est. Priority Date: 10/12/2007
• Status: Active Grant

• Current Assignee: BBY Solutions Incorporated (Best Buy Co., Inc.)
• Original Assignee: BioSensics, LLC (Best Buy Co., Inc.)
• Inventors: Najafi, Bijan, Vaziri, Ashkan, Boloori, Ali-Reza
• Primary Examiner(s): Hindenburg, Max
• Assistant Examiner(s): Henson, Devin
• Application Number:
  US12/249,948
• Time in Patent Office: 1,353 Days
• Field of Search: 600/587, 600/595, 702/141, 702/150, 702/152, 702/153
• US Class Current: 600/595
• CPC Class Codes: A61B 2503/08 Elderly
• CPC Class Codes: A61B 2505/07 Home care
• CPC Class Codes: A61B 2505/09 Rehabilitation or training
• CPC Class Codes: A61B 2562/0219 Inertial sensors, e.g. accelerometers, gyroscopes, tilt switches
• CPC Class Codes: A61B 5/0004 characterised by the type of physiological signal transmitted
• CPC Class Codes: A61B 5/0022 Monitoring a patient using a global network, e.g. telephone networks, internet
• CPC Class Codes: A61B 5/11 Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb for measuring pulse A61B5/02 ;A61B5/1038 takes precedence; motion detection to correct for motion artifacts in physiological signals A61B5/721
• CPC Class Codes: A61B 5/1116 Determining posture transitions
• CPC Class Codes: A61B 5/1117 Fall detection
• CPC Class Codes: A61B 5/1118 Determining activity level
• CPC Class Codes: A61B 5/112 Gait analysis
• CPC Class Codes: A61B 5/1123 Discriminating type of movement, e.g. walking or running A61B5/1116, A61B5/112 take precedence
• CPC Class Codes: A61B 5/6823 Trunk, e.g., chest, back, abdomen, hip
• CPC Class Codes: A61B 5/6831 Straps, bands or harnesses
• CPC Class Codes: A61B 5/7203 for noise prevention, reduction or removal
• CPC Class Codes: A61B 5/7225 Details of analog processing, e.g. isolation amplifier, gain or sensitivity adjustment, filtering, baseline or drift compensation A61B5/30, A61B5/316 take precedence; transformers for specific applications H01F38/00; amplifiers H03F
• CPC Class Codes: A61B 5/7264 Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems neural networks per se G06N3/00; expert systems per se G06N5/00
• CPC Class Codes: A61B 5/7275 Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor
• CPC Class Codes: A61B 5/7282 Event detection, e.g. detecting unique waveforms indicative of a medical condition A61B5/4094, A61B5/4818, A61B5/349, A61B5/0823 take precedence
• CPC Class Codes: A61B 5/746 Alarms related to a physiological condition, e.g. details of setting alarm thresholds or avoiding false alarms alarms per se G08B
• CPC Class Codes: G08B 21/043 : detecting an emergency event, e.g. a fall
• CPC Class Codes: G08B 21/0446 : worn on the body to detect changes of posture, e.g. a fall, inclination, acceleration, gait
• CPC Class Codes: G16H 50/20 : for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: G16Z 99/00 : Subject matter not provided for in other main groups of this subclass

Abstract

The present invention relates to a light-weight, small and portable ambulatory sensor for measuring and monitoring a person'"'"'s physical activity. Based on these measurements and computations, the invented system quantifies the subject'"'"'s physical activity, quantifies the subject'"'"'s gait, determines his or her risk of falling, and automatically detects falls. The invention combines the features of portability, high autonomy, and real-time computational capacity. High autonomy is achieved by using only accelerometers, which have low power consumption rates as compared with gyroscope-based systems. Accelerometer measurements, however, contain significant amounts of noise, which must be removed before further analysis. The invention therefore uses novel time-frequency filters to denoise the measurements, and in conjunction with biomechanical models of human movement, perform the requisite computations, which may also be done in real time.

Claims

1. A body movement monitoring system comprising:

a sensor, adapted to be attached to an upper part of a body of a person, the sensor comprising an accelerometer component adapted to generate signals in response to movement of the body, the signals comprising a frontal acceleration signal and a vertical acceleration signal;

one or more processor circuits programmed to;

filter at least a portion of the signals generated by the accelerometer component to obtain a first set of filtered signals;

determine an initial estimated postural transition duration for a first postural transition based on one or more peaks in the first set of filtered signals, and for a first time window;

apply one or more additional filters to a segment of the signals generated by the accelerometer component to obtain a second set of filtered signals, the segment corresponding to only a second time window that is narrower than the first time window and that is selected based at least in part on a timing corresponding to the initial estimated postural transition duration for the first postural transition;

determine a revised estimated postural transition duration that is more accurate than the initial estimated postural transition duration for the first postural transition based on a location of one or more peaks in the second set of filtered signals; and

derive information related to the first postural transition using the revised estimated postural transition duration and the second set of filtered signals, the information including identification of at least one of a time of occurrence, a duration, and a type of the first postural transition.

2. The system of claim 1, wherein said one or more processor circuits are further programmed to:

filter the frontal accelerometer signal of the signals from the accelerometer component with at least one time frequency filter to obtain the first set of filtered signals.

3. The system of claim 2, wherein said one or more additional filters applied to the first set of filtered signals comprises a multi-stage wavelet.

4. The system of claim 3, wherein said one or more processor circuits are further programmed to derive the information by at least:

estimating an inertial acceleration of the person during a time period corresponding to the revised estimated postural transition duration, the estimating further comprising at least one of;

processing the second set of filtered signals to correct for a sagittal trunk tilt angle of the person relative to the sensor during the time period corresponding to the revised estimated postural transition duration;

applying one or more filters to the vertical acceleration signal from the time period corresponding to the revised estimated postural transition duration to obtain a third set of filter signals; and

identifying one or more peaks in the third set of filter signals.

5. The system of claim 4, wherein the first postural transition is classified based on a positive extrema and a negative extrema of the one or more identified peaks in the third set of filter signals.

6. The system of claim 5, wherein the first postural transition is classified as:

a postural transition to standing when positive extrema of the vertical acceleration is followed by the negative extrema, and

a postural transition to sitting when negative extrema of the vertical acceleration is followed by the positive extrema.

7. The system of claim 1, wherein said one or more processor circuits are further programmed to:

identify a potential time period corresponding to a duration of walking of the person from the vertical acceleration signal by removing from consideration a portion of the vertical acceleration signal that is from a time corresponding to the first postural transition, the time based at least in part on the derived information; and

identify, in the potential time period, at least three peaks within a predefined time interval as an indicator of a walking pattern.

8. The system of claim 7, wherein said one or more processor circuits are further programmed to:

remove a portion of the signals attributable to the person'"'"'s trunk oscillations for better identification of the walking pattern.

9. The system of claim 1, wherein the accelerometer component comprises a tri-axial accelerometer.

10. The system of claim 1, wherein the accelerometer component comprises a plurality of mono-axial accelerometers measuring accelerations in a plurality of perpendicular directions.

11. The system of claim 1, further comprising:

one or more data storage systems configured to store at least one of;

signals recorded by said sensor, and

the information derived by said one or more processor circuits.

12. The system of claim 1, further comprising:

a communications system configured to transmit data to a receiving system, the data comprising at least one of;

data reflecting at least a portion of the signals from said accelerometer component; and

at least a portion of the information derived by said at least one processor circuits.

13. A fall detection monitoring system comprising:

a sensor adapted to be attached to an upper part of a body of a person, the sensor comprising a movement detection component adapted to generate signals in response to a movement of the body; and

one or more processor circuits programmed to;

identify one or more peaks in the signals from the movement detection component; and

for at least one of the one or more identified peaks,

process the signals identify a non-fall activity performed by the person in a time period prior to a time period corresponding to the peak;

in response to identifying that the non-fall activity preceding the peak was a first type of non-fall activity, set one or more fall thresholds to a first set of one or more values;

in response to identifying that the non-fall activity preceding the peak was a second type of non-fall activity, set the one or more fall thresholds to a second set of one or more values; and

identify the peak as indicative of a fall event if the peak meets the one or more fall thresholds.

14. The system of claim 13, wherein the movement detection component comprises an accelerometer component.

15. The system of claim 14, wherein the signals comprise a frontal acceleration signal, a vertical acceleration signal, and a lateral acceleration signal, and wherein the one or more processor circuits are further programmed to identify one or more peaks in the signals from the movement detection component by:

determining a magnitude of acceleration in a traversal plane based on the frontal acceleration and lateral acceleration signals; and

detecting one or more peaks within the determined magnitude of acceleration in the traversal plane.

16. The system of claim 13, wherein said first type of non-fall activity comprises walking and said first set of one or more values are associated with either walking on a right step or with walking on a left step.

17. The system of claim 13, wherein said one or more processor circuits are further programmed to:

determine information related to physical activity of the person after a time period corresponding to the fall event; and

classify the fall event as a serious fall event if the determined information indicates that the fall event is followed by a long period of rest or multiple unsuccessful postural transitions in succession.

18. The system of claim 13, wherein said one or more processor circuits are further programmed to:

when the identified non-fall activity is not walking, segment a shock pattern identified based on the one or more peaks;

estimate a width of the shock pattern; and

estimate a vertical speed of the person in a time period prior to a time period corresponding to the shock pattern.

19. The system of claim 13, wherein said one or more processor circuits are further programmed to:

record a history of one or more identified fall events; and

evaluate a risk of falling of the person based on;

the history of the one or more identified fall events within a pre-determined time period and one or more of;

a duration of postural transitions of the person within the pre-determined time period,

a number of identified postural transitions within the pre-determined time period,

a walking performance within the pre-determined time period, and

a longest detected duration of walking within the pre-determined time period.

20. The system of claim 19, wherein the pre-determined time period is a day.

21. A method of monitoring body movement using signals from a sensor adapted to be attached to an upper part of a body of a person, the sensor comprising an accelerometer component adapted to generate signals in response to movement of the body, the signals comprising a frontal acceleration signal and a vertical acceleration signal, the method comprising:

filtering, by one or more processor circuits, at least a portion of the signals generated by the accelerometer component to obtain a first set of filtered signals;

determining, by said one or more processor circuits, an initial estimated postural transition duration for a first postural transition based on one or more peaks in the first set of filtered signals and for a first time window;

applying, by said one or more processor circuits, one or more additional filters to a segment of the signals generated by the accelerometer component to obtain a second set of filtered signals, the segment corresponding to only a second time window that is narrower than the first time window and that is selected based at least in part on a timing corresponding to the initial estimated postural transition duration for the first postural transition;

determining, by said one or more processor circuits, a revised estimated postural transition duration that is more accurate than the initial estimated postural transition duration for the first postural transition based on a location of one or more peaks in the second set of filtered signals; and

deriving, by said one or more processor circuits, information related to the first postural transition using the revised estimated postural transition duration and the second set of filtered signals, the information including identification of at least one of a time of occurrence, a duration, and a type of the first postural transition.

22. A method of detecting fall events using signals from a sensor adapted to be attached to an upper part of a body of a person, the sensor comprising a movement detection component adapted to generate signals in response to a movement of the body, the method comprising:

identifying, by one or more processor circuits, one or more peaks in the signals generated by the movement detection component; and

for at least one of the one or more identified peaks,

processing the signals to identify a non-fall activity performed by the person in a time period prior to a time period corresponding to the peak;

if the non-fall activity preceding the peak was identified as a first type of non-fall activity, setting one or more fall thresholds to a first set of one or more values;

if the non-fall activity preceding the peak was identified as a second type of non-fall activity, setting the one or more fall thresholds to a second set of one or more values; and

identifying the peak as indicative of a fall event if the peak meets the one or more fall thresholds.