Device and method for removal of artifacts in physiological measurements

1. A method for generating an electronic signal representing physiological measurements with artifacts in the physiological measurements removed, the method comprising:

obtaining, using a physiological measurement sensor integrated into a wearable device, an electronic physiological signal of a user;

obtaining, using an accelerometer integrated into the wearable device, corresponding electronic motion data representative of motion of the user;

determining, using a processor coupled to the physiological measurement sensor and the accelerometer, two or more motion cycles in the electronic motion data;

constructing, using the processor, an electronic noise reference based on segments of the electronic physiological signal corresponding to the two or more motion cycles respectively; and

generating, using the processor, the electronic signal representing the physiological measurements with the artifacts in the physiological measurements removed by filtering the electronic physiological signal with the electronic noise reference;

wherein constructing, using the processor, the electronic noise reference comprises:

normalizing, using the processor, the segments of the physiological signal in time;

obtaining, using the processor, an average of the normalized segments of the physiological signal as a phase-locked artifact component;

re-scaling, using the processor, the phase-locked artifact component to generate rescaled versions of the phase-locked artifact component corresponding to the respective segments of the electronic physiological signal; and

concatenating, using the processor, the respective re-scaled versions of the phase-locked artifact component.

2. The method as claimed in claim 1 , wherein determining, using the processor, two or more motion cycles in the motion data comprises bandpass filtering, using the processor, the motion data to generate filtered motion data.

3. The method as claimed in claim 2 , wherein determining, using the processor, two or more motion cycles in the motion data further comprises at least one of: a) performing differentiation, using the processor, on the filtered motion signal; b) determining, using the processor, peaks or valleys in a derivative of the filtered motion data; and c) determining, using the processor, segments of the motion data between two peaks or valleys as respective motion cycles.

4. The method as claimed in claim 1 , wherein constructing, using the processor, the electronic noise reference based on the segments of the electronic physiological signal comprises enhancing motion artifacts.

5. The method as claimed in claim 1 , wherein constructing, using the processor, the electronic noise reference based on the segments of the electronic physiological signal comprises suppressing a physiological component.

6. The method as claimed in claim 1 , wherein obtaining the average of the normalized segments of the electronic physiological signal comprises enhancing in-phase components between the normalized segments of the electronic physiological signal.

7. The method as claimed in claim 1 , wherein the obtained corresponding motion data comprises single-, dual- or multi-axial motion signals, and wherein determining the two or more motion cycles is based on one or more of the single-, dual- or multi-axial motion signals.

8. The method as claimed in claim 1 , further comprising setting a number of the motion cycles to be determined for the construction of the electronic noise reference, wherein the setting the number is based on user input, and wherein the setting the number comprises determining an intensity of the motion based on the motion data.

9. The method as claimed in claim 1 , wherein filtering, using the processor, the electronic physiological signal with the electronic noise reference comprises applying a least-mean-square (LMS) algorithm, or a recursive-least-square (RLS) algorithm.

10. A device for generating an electronic signal representing physiological measurements with artifacts in the physiological measurements removed, the device comprising:

a physiological measurement sensor configured to obtain an electronic physiological signal of a user;

an accelerometer configured to obtain corresponding motion data representative of motion of the user; and

a processor coupled to the physiological measurement sensor and the accelerometer and configured to determine two or more motion cycles in the motion data, to construct an electronic noise reference based on segments of the electronic physiological signal corresponding to the two or more motion cycles respectively, and to generate the electronic signal representing the physiological measurements with the artifacts in the physiological measurements removed by filtering the electronic physiological signal with the electronic noise reference;

wherein the processor is configured to construct the electronic noise reference by executing the steps of:

normalizing the segments of the physiological signal in time;

obtaining an average of the normalized segments of the physiological signal as a phase-locked artifact component;

re-scaling the phase-locked artifact component to generate rescaled versions of the phase-locked artifact component corresponding to the respective segments of the electronic physiological signal; and

concatenating the respective re-scaled versions of the phase-locked artifact component.

11. The device as claimed in claim 10 , wherein determining two or more motion cycles in the motion data comprises bandpass filtering the motion data to generate filtered motion data.

12. The device as claimed in claim 11 , wherein determining two or more motion cycles in the motion data further comprises at least one of: a) performing differentiation on the filtered electronic motion signal; b) determining peaks or valleys in a derivative of the filtered motion data; and c) determining segments of the motion data between two peaks or valleys as respective motion cycles.

13. The device as claimed in claim 10 , wherein constructing the electronic noise reference based on the segments of the electronic physiological signal comprises enhancing motion artifacts.

14. The device as claimed in claim 10 , wherein constructing the electronic noise reference based on the segments of the electronic physiological signal comprises suppressing a physiological component.

15. The device as claimed in claim 10 , wherein obtaining the average of the normalized segments of the electronic physiological signal comprises enhancing in-phase components between the normalized segments of the electronic physiological signal.

16. The device as claimed in claim 10 , wherein the obtained corresponding motion data comprises single-, dual- or multi-axial motion signals, and wherein determining the two or more motion cycles is based on one or more of the single-, dual- or multi-axial motion signals.

17. The device as claimed in claim 10 , further comprising the processor being configured for setting a number of the motion cycles to be determined for the construction of the electronic noise reference, wherein the setting the number is based on user input, and wherein the setting the number comprises determining an intensity of the motion based on the motion data.

18. The device as claimed in claim 10 , wherein filtering the electronic physiological signal with the noise reference comprises applying a least-mean-square (LMS) algorithm, or a recursive-least-square (RLS) algorithm.