Fall detection algorithm utilizing a three-axis accelerometer
First Claim
Patent Images
1. A medical device comprising:
- a physiological sensor for sensing a signal in a patient;
a three-axis accelerometer for generating a 3D accelerometer signal; and
a processor configured to;
compute a reference vector corresponding to an upright patient position based on the 3D accelerometer signal;
determine a current patient position vector based, at least in part, on the 3D accelerometer signal sampled at a low sample rate;
determine that a difference between the current patient position vector and the reference vector exceeds a difference threshold;
in response to the determination that the difference exceeds the difference threshold, determine an acceleration magnitude based, at least in part, on the 3D accelerometer signal sampled at a high sample rate being higher than the low sample rate; and
detect that the patient has fallen based, at least in part, on the patient position and the acceleration magnitude.
1 Assignment
0 Petitions
Accused Products
Abstract
A three-axis accelerometer signal is used in a medical device for detecting that a patient has fallen. The accelerometer signal is sampled at a low sampling rate for determining a patient position. The accelerometer signal is sampled at a high sampling rate for determining an acceleration magnitude in response to determining the patient position. The patient position and acceleration magnitude are used to detect that a patient has fallen.
-
Citations
9 Claims
-
1. A medical device comprising:
-
a physiological sensor for sensing a signal in a patient; a three-axis accelerometer for generating a 3D accelerometer signal; and a processor configured to; compute a reference vector corresponding to an upright patient position based on the 3D accelerometer signal; determine a current patient position vector based, at least in part, on the 3D accelerometer signal sampled at a low sample rate; determine that a difference between the current patient position vector and the reference vector exceeds a difference threshold; in response to the determination that the difference exceeds the difference threshold, determine an acceleration magnitude based, at least in part, on the 3D accelerometer signal sampled at a high sample rate being higher than the low sample rate; and detect that the patient has fallen based, at least in part, on the patient position and the acceleration magnitude. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9)
-
Specification