Low-complexity sensor displacement tolerant pulse oximetry based heart rate measurement
Methods for heart rate measurement based on pulse oximetry are provided that can tolerate some degree of relative displacement of a photoplethysmograph (PPG) heart rate monitor device. In some methods, artifact compensation based on a reference signal is performed on the PPG signal data to remove artifacts in the signal that may be caused, for example, by changes in ambient light and/or motion of a person wearing the monitor device. The reference signal used for artifact compensation may be generated using an LED of a complementary wavelength to that of the LED used to generate the PPG signal, or by driving an LED at a lower current than the current applied to generate the PPG signal.
1. A method performed by a photoplethysmograph (PPG) heart rate monitor device, the method comprising:
receiving an accelerometer signal from a motion sensor of the heart rate monitor device;
capturing a PPG light signal, in which the PPG light signal is:
filtered when the accelerometer signal is below a threshold; or
motion compensated when the accelerometer signal is above the threshold;
detecting a pulse in a portion of the filtered or motion compensated PPG light signal, including detecting a start point of the pulse;
determining a summit point and an end point of the pulse based on the start point of the pulse, in which determining the summit point includes:
determining a first maximum point within a first portion of the pulse;
determining a second maximum point after the first maximum point;
determining a direct linear path difference between the first maximum point and the second maximum point;
determining an indirect path difference between the first maximum point and the second maximum point; and
selecting the summit point as either the first maximum point or the second maximum point based on a comparison of the direct linear path difference and the indirect path difference;
estimating a pulse rate based on the start point, the summit point, and the end point; and
outputting the estimated pulse rate to another device.
2. The method of claim 1 , wherein the selecting the summit point includes:
selecting the first local maximum point as the summit point when the direct linear path difference does not exceed the indirect path difference by a predetermined factor; and
selecting the second local maximum point as the summit point; when the direct linear path difference exceeds the indirect path difference by the predetermined factor.
3. A method performed by a photoplethysmograph (PPG) heart rate monitor device, the method comprising:
receiving an accelerometer signal from a motion sensor of the heart rate monitor device;
capturing a PPG light signal, in which the PPG light signal is:
filtered, when the accelerometer signal is below a threshold; or
motion compensated, when the accelerometer signal is above the threshold;
detecting a pulse in a portion of the filtered or motion compensated PPG light signal, including detecting a start point of the pulse;
determining a summit point and an end point of the pulse based on the start point of the pulse, wherein determining the end point includes:
determining a first local minimum point after the summit point of the pulse;
determining a second minimum point after the first minimum point;
determining a first path between the summit point and the first minimum point;
determining a second path between the summit point and the second minimum point; and
selecting the end point as either the first minimum point or the second minimum point based on a comparison of the first path and the second path;
estimating a pulse rate based on the start point, the summit point, and the end point; and
outputting the estimated pulse rate to another device.
4. The method of claim 3 , wherein the selecting the end point includes:
determining a first normalized difference between a first length and a second length of the first minimum point, wherein:
the first length is measured between the start point and the summit point; and
the second length is measured between the summit point and the first minimum point;
determining a second normalized difference between a first length and a second length of the second minimum point, wherein:
the first length is measured between the start point and the summit point; and
the second length is measured between the summit point and the second minimum point; and
comparing the first normalized difference with the second normalized difference.
5. The method of claim 4 , wherein the selecting the end point includes:
selecting the first minimum point as the end point when the second normalized difference is greater than a predetermined multiple of the first normalized difference; and
selecting the second minimum point as the end point when the second normalized difference is less than the predetermined multiple of the first normalized difference.