CALCULATION OF MINIMUM GROUND CLEARANCE USING BODY WORN SENSORS

• US 20130060512A1
• Filed: 09/01/2011
• Published: 03/07/2013
• Est. Priority Date: 09/01/2011
• Status: Active Grant

• Current Assignee: Linus Health Europe Limited
• Original Assignee: Intel-GE Care Innovations, LLC (Intel Corporation)
• Inventors: GREENE, Barry R.
• Granted Patent: US 9,524,424 B2
• Time in Patent Office: Days
• US Class Current: 702/141
• CPC Class Codes: A61B 5/1036 Measuring load distribution, e.g. podologic studies
• CPC Class Codes: A61B 5/112 Gait analysis
• CPC Class Codes: A61B 5/1121 Determining geometric values, e.g. centre of rotation or angular range of movement
• CPC Class Codes: A61B 5/1127 using markers
• CPC Class Codes: A61B 5/7285 for synchronising or triggering a physiological measurement or image acquisition with a physiological event or waveform, e.g. an ECG signal A61B6/5205, A61B6/541, A61B8/5207, A61B8/543 take precedence
• CPC Class Codes: G06F 2218/20 by applying autoregressive analysis
• CPC Class Codes: G06V 40/23 Recognition of whole body movements, e.g. for sport training
• CPC Class Codes: G16H 50/50 for simulation or modelling of medical disorders

Abstract

Methods, systems, and apparatus for deriving a relationship between minimum ground clearance (MGC) and inertial sensor data. A regression model may be generated by collecting tri-axial angular velocity and acceleration data from inertial sensors and MGC data from optical motion capture systems during a walking trial. A linear, quadratic, interaction, stepwise interaction, or another regression model may be generated. The regression model may estimate the MGC as a function of one or more parameters measured by or derived from the inertial sensor data. The regression model may be used to calculate an estimate of the MGC based on inertial sensor data collected from one or more individuals.

Claims

1. A method comprising:

calculating, with a regression model, an estimate of a minimum ground clearance parameter of a person associated with motion data, wherein an input to the regression model comprises the motion data, and

wherein the motion data comprises data obtained by one or more inertial sensors mounted on the person'"'"'s body.

2. The method of claim 1, wherein the minimum ground clearance parameter comprises a mean minimum ground clearance, a coefficient of variation of minimum ground clearance, or any combination thereof.

3. The method of claim 1, wherein the one or more inertial sensors comprise an accelerometer and a gyroscope.

4. The method of claim 3, wherein the one or more inertial sensors comprise foot-mounted inertial sensors.

5. The method of claim 3, further comprising calculating a mean of absolute values of vertical angular velocity, a mean of absolute values of vertical acceleration, or any combination thereof from the motion data, wherein calculating the estimate of the minimum ground clearance parameter is based on the mean of absolute values of vertical angular velocity and based on the mean of absolute values of vertical acceleration.

6. The method of claim 3, wherein calculating the estimate of the minimum ground clearance parameter is based on a mean angular velocity at a mid-swing point, a minimum angular velocity, a maximum angular velocity, a coefficient of variation of angular velocity, or any combination thereof.

7. The method of claim 1, wherein the selected regression model is a quadratic regression model, an interaction regression model, or any combination thereof.

8. The method of claim 1, wherein the selected regression model is a linear regression model.

9. The method of claim 8, wherein the slope of the linear regression model is based on the person'"'"'s weight, shank length, age, or any combination thereof.

10. The method of claim 1, further comprising generating a classifier model of falls risk based on the calculated minimum ground clearance parameter.

11. A non-transitory computer-readable medium, the computer-readable medium comprising one or more instructions that, when executed by one or more processors, cause the one or more processors to:

calculate, with a regression model, an estimate of a minimum ground clearance parameter of a person associated with motion data, wherein an input to the regression model comprises the motion data, and

wherein the motion data comprises data obtained by one or more inertial sensors mounted on the person'"'"'s body.

12. The non-transitory computer-readable medium of claim 11, wherein the motion data comprises angular velocity, mean of the absolute values of angular velocity, maximum angular velocity, minimum angular velocity, coefficient of variation of angular velocity, or any combination thereof.

13. The non-transitory computer-readable medium of claim 12, wherein the motion data comprises acceleration, mean of the absolute values of acceleration, maximum acceleration, minimum acceleration, coefficient of variation of acceleration, or any combination thereof.

14. The non-transitory computer-readable medium of claim 11, wherein determining the optimal regression model comprises calculating a Pearson'"'"'s correlation coefficient between the motion data and the minimum ground clearance data.

15. The non-transitory computer-readable medium of claim 11, further comprising receiving a second set of motion data;

and

calculating a second set of minimum ground clearance data based on the second set of motion data and the generated regression model.

16. A method comprising:

receiving motion data comprising one or more of mean angular velocity measured at mid-swing points of a person'"'"'s motion, mean of absolute values of angular velocity of the person'"'"'s motion, and minimum angular velocity of the person'"'"'s motion;

generating a value indicative of the person'"'"'s risk of falling based on the received motion data.

17. The method of claim 16, wherein the received angular velocity data is measured along the person'"'"'s anatomical sagittal plane.

18. The method of claim 17, wherein generating the value further comprises calculating a correlation between the motion data and one or more of a TUG time and a BBS score of the person.