Method and apparatus for estimating the fall risk of a user

1. A non-transitory computer program product, comprising computer program code that, when executed on a computer or processor, causes the computer or processor to determine a fall risk for a user by:

low-pass filtering a plurality of measurements of an acceleration which were measured by an accelerometer of a device of the user, wherein the plurality of measurements include a separate accelerometer signal for each of three axes of the accelerometer, to generate three low-pass filtered accelerometer signals, including a separate low-pass filtered accelerometer signal for each of the three axes;

processing the three low-pass filtered accelerometer signals to determine a vertical component of acceleration;

high-pass filtering the three low-pass filtered accelerometer signals to generate a separate band-pass filtered accelerometer signal for each of the three axes;

computing three standard deviations, one for each of the three band-pass filtered accelerometer signals, for a time window of a predetermined length;

identifying a maximum standard deviation of the three standard deviations;

estimating a change in altitude of the device from an air pressure measured by a sensor of the device;

analyzing the plurality of measurements of the acceleration of the user to determine if the user has performed a sit-to-stand transfer by:

identifying a sub-portion of the vertical component of acceleration as a candidate sit-to-stand movement; and

identifying the sub-portion as including an actual sit-to-stand transfer in response to the candidate sit-to-stand movement occurring during a same predetermined time window as a change in altitude of the sensor;

determining a timing of the sit-to-stand transfer in the sub-portion based on the maximum standard deviation, including a start time and an end time of the sit-to-stand transfer;

identifying a peak vertical acceleration of the user during the sit-to-stand transfer from the plurality of measurements of the acceleration of the user from samples of the sub-portion corresponding to the timing, wherein the identified peak vertical acceleration is scaled using an estimate of gravity obtained from the measurements of the acceleration of the user thereby producing a scaled peak vertical acceleration; and

estimating a fall risk for the user from the scaled peak vertical acceleration, wherein the fall risk is inversely proportional to the scaled peak vertical acceleration.

2. An apparatus for estimating a fall risk for a user, the apparatus comprising:

an accelerometer configured to take a plurality of measurements of acceleration of the user and generate three acceleration signals respectively indicative of an acceleration along each axis of three orthogonal axes;

a low-pass filter configured to low pass filter each of the three acceleration signals, generating three low-pass filtered acceleration signals;

a high-pass filter configured to high pass filter each of the three low-pass filtered acceleration signals, generating three band-pass filtered acceleration signals;

a computer processor configured to:

compute a standard deviation of each of the three band-pass filtered acceleration signals based on samples for a time window of a predetermined length, generating three standard deviation signals respectively for the three band-pass filtered acceleration signals;

analyze the plurality of measurements of the acceleration of the user to determine if the user has performed a sit-to-stand transfer by matching a vertical acceleration signal of the three band-pass filtered acceleration signals with a predetermined pattern representing a vertical acceleration that is expected to occur during the sit-to-stand transfer where a matched filter applies an impulse response that is an approximated vertical acceleration experienced during the sit-to-stand transfer to the vertical acceleration signal;

determine a timing of the sit-to-stand transfer based on a maximum standard deviation of the three standard deviation signals;

identify a peak vertical acceleration of the user during the sit-to-stand transfer from the plurality of measurements of the acceleration of the user based on the determined timing, wherein the identified peak vertical acceleration is scaled using an estimate of gravity obtained from the plurality of measurements of the acceleration of the user thereby producing a scaled peak vertical acceleration; and

estimate a fall risk for the user by computing an inverse of the scaled peak vertical acceleration.

3. A device that is configured to be worn by a user, the device comprising:

the apparatus as claimed in claim 2 .

4. A system, comprising:

a device that is configured to be worn by a user, the device comprising the accelerometer as claimed in claim 2 ; and

a base unit that is configured to communicate with the device, and the base unit comprising the computer processor as claimed in claim 2 .

5. A system, comprising:

a force plate; and

a base unit that comprises the apparatus as claimed in claim 2 , wherein the computer processor is configured to receive measurements of forces from the force plate and to process the plurality of measurements of forces to determine measurements of the acceleration.

6. The apparatus of claim 2 , wherein the computer processor is configured to:

identify a signal of the three low-pass filtered acceleration signals having a highest component of acceleration;

determine an orientation of the accelerometer by determining an angle between an acceleration acting on the accelerometer and an axis for the signal with the highest component of acceleration; and

determine a vertical component from the plurality of measurements of acceleration based on the orientation of the accelerometer.

7. The apparatus of claim 6 , wherein the acceleration acting on the accelerometer is an acceleration due to gravity.

8. The apparatus of claim 6 , wherein the computer processor is configured to determine the timing by:

identifying a first sample in the vertical component with a first value smaller than a first predetermined threshold to determine an onset of the sit-to-stand transfer; and

identifying a second sample after a largest local minimum of the vertical component with a second value greater than a second predetermined threshold to determine an end of the sit-to-stand transfer,

wherein the peak vertical acceleration is identified based on the onset and the end of the sit-to-stand transfer.

9. The apparatus of claim 8 , wherein the onset and the end of the sit-to-stand transfer are utilized to determine a subset of samples that are analyzed to determine the peak vertical acceleration.

10. The apparatus of claim 8 , further comprising:

a sensor configured to measure an altitude or a height of the apparatus above ground.

11. The apparatus of claim 10 , where the sensor is one of an altimeter or air pressure sensor.

12. The apparatus of claim 10 , where the computer processor determines the sit-to-stand transfer is a true sit-to-stand transfer when the sit-to-stand transfer occurs at a same time as a change in the height of the sensor within a predetermined range.

13. The apparatus of claim 12 , where the peak vertical acceleration is determined during the sit-to-stand transfer only if the sit-to-stand transfer is a true sit-to-stand transfer.

14. The apparatus of claim 13 , where the peak vertical acceleration is discarded in response to the sit-to-stand transfer not being a true sit-to-stand transfer.

15. The apparatus of claim 2 , wherein an output of the matched filter is a set of coefficients that indicate a match to the predetermined pattern.

16. The apparatus of claim 15 , wherein each coefficient of the set of coefficients represents a match of a number of consecutive measurement samples covering a time period of a same length as the predetermined pattern.

17. The apparatus of claim 2 , wherein the gravity is obtained from the plurality of measurements of acceleration in a time period before a start of the sit-to-stand transfer only if the three standard deviation signals are less than a threshold value.

18. The apparatus of claim 2 , wherein the gravity is obtained from the plurality of measurements of acceleration in a time period before a start of the sit-to-stand transfer only if a variance of the plurality of measurements of acceleration occurring in the time period is less than a threshold value.

19. The apparatus of claim 2 , wherein the gravity is obtained from the plurality of measurements of acceleration in a time period before a start of the sit-to-stand transfer only if a range of the plurality of measurements of acceleration occurring in the time period is less than a threshold value.