Respiratory rate detection using decomposition of ECG

1. A method to determine a respiratory rate of a user using a wireless sensor device that includes at least one electrode, a processor, a transmitter and a memory device coupled to the processor, the memory device stores executable instructions that, when executed by the processor, causes the processor to perform operations, comprising:

measuring an electrocardiogram (ECG) segment of the user via the at least one electrode;

decomposing the ECG segment into a plurality of Intrinsic Mode Functions (IMFs) using Ensemble Empirical Mode Decomposition (EEMD);

evaluating the plurality of IMFs to choose one of the plurality of IMFs based on a respiratory band power of the plurality of IMFs;

evaluating spectral content of each of the plurality of IMFs by obtaining a Power Spectral Density (PSD) for each of the plurality of IMFs;

obtaining the respiratory band power for each of the plurality of IMFs;

choosing one of the plurality of IMFs that contributes maximal percentage of a total respiratory band power of the plurality of IMFs;

providing the one of the plurality of IMFs as a surrogate respiration waveform; and

determining the respiratory rate using the one of the plurality of functions and a time-domain based peak detection algorithm.

2. The method of claim 1 , wherein the determining further comprises:

determining the respiratory rate using breath-to-breath peak detection in time-domain and respiratory frequency peak in the PSD of the surrogate respiration waveform.

3. The method of claim 2 , wherein the breath-to-breath peak detection in time-domain includes an algorithm that identifies peak events by computing a first derivative of the surrogate respiration waveform and finds zero crossing events with positive to negative sign changes.

4. The method of claim 1 , wherein the PSD is obtained using a Welch periodogram.

5. The method of claim 1 , wherein the respiratory band power obtained for each of the plurality of IMFs is an area under the PSD between frequencies 0.11 Hz to 0.45 Hz.

6. The method of claim 1 , further comprising:

processing the ECG segment using both a low pass digital filter with a cutoff frequency (F c ) of 10 Hz and a down sampling with a sampling frequency (F s ) of 25 Hz.

7. The method of claim 1 , wherein the ECG segment is measured at a 125 Hz sampling frequency (F s ) and at a length of 40 seconds.

8. The method of claim 7 , further comprising:

shifting the ECG segment every 5 seconds to provide a 35 second overlap for determining the respiratory rate.

9. A non-transitory computer-readable medium storing executable instructions that, in response to execution, cause a wireless sensor device that includes at least one electrode, a processor, a transmitter and a memory device coupled to the processor, to perform operations to determine a respiratory rate of a user using an electrocardiogram (ECG) segment of the user comprising:

measuring an electrocardiogram (ECG) segment of the user via the at least one electrode;

decomposing the ECG segment into a plurality of Intrinsic Mode Functions (IMFs) using Ensemble Empirical Mode Decomposition (EEMD);

evaluating the plurality of IMFs to choose one of the plurality of IMFs based on a respiratory band power of the plurality of IMFs;

evaluating spectral content of each of the plurality of IMFs by obtaining a Power Spectral Density (PSD) for each of the plurality of IMFs;

obtaining the respiratory band power for each of the plurality of IMFs;

choosing one of the plurality of IMFs that contributes maximal percentage of a total respiratory band power of the plurality of IMFs;

providing the one of the plurality of IMFs as a surrogate respiration waveform; and

determining the respiratory rate using the one of the plurality of functions and a time-domain based peak detection algorithm.

10. The non-transitory computer-readable medium of claim 9 , wherein the determining further comprises:

determining the respiratory rate using breath-to-breath peak detection in time-domain and respiratory frequency peak in the PSD of the surrogate respiration waveform.

11. The non-transitory computer-readable medium of claim 10 , wherein the breath-to-breath peak detection in time-domain includes an algorithm that identifies peak events by computing a first derivative of the surrogate respiration waveform and finds zero crossing events with positive to negative sign changes.

12. The non-transitory computer-readable medium of claim 9 , wherein the PSD is obtained using a Welch periodogram.

13. The non-transitory computer-readable medium of claim 9 , wherein the respiratory band power obtained for each of the plurality of IMFs is an area under the PSD between frequencies 0.11 Hz to 0.45 Hz.

14. The non-transitory computer-readable medium of claim 9 , further comprising:

processing the ECG segment using both a low pass digital filter with cutoff frequency (F c ) of 10 Hz and a down sampling with a sampling frequency (F s ) of 25 Hz.

15. The non-transitory computer-readable medium of claim 9 , wherein the ECG segment is measured at a 125 Hz sampling frequency (F s ) and at a length of 40 seconds.

16. The non-transitory computer-readable medium of claim 15 , further comprising:

shifting the ECG segment every 5 seconds to provide a 35 second overlap for determining the respiratory rate.