Processing of electrophysiological signals
In an embodiment, PhotoPlethysmoGraphy (PPG) signals are processed by detecting peaks and valleys in the PPG signal, segmenting the PPG signal to provide a time series of PPG waveforms located between two subsequent valleys in the PPG signal, applying to the waveforms in the time series pattern recognition with respect to a reference PPG waveform pattern produced based on a mathematical model of the PPG signal by assigning to the waveforms in the time series a recognition score. A resulting PPG signal is produced by retaining the waveforms in the time series having an assigned recognition score reaching a recognition threshold, and discarding the waveforms in the time series having an assigned recognition score failing to reach the recognition threshold.
1. A system comprising:
a photoplethysmography (PPG) sensing apparatus configured to receive a PPG signal; and
a PPG signal processor coupled to the PPG sensing apparatus, the PPG signal processor configured to:
detect peaks and valleys in the PPG signal;
segment the PPG signal to provide a first time series of PPG waveforms located between two subsequent valleys in the PPG signal;
apply pattern recognition to the first time series with respect to a reference PPG waveform pattern produced based on a mathematical model of the PPG signal by assigning a recognition score to the PPG waveforms in the first time series;
produce the reference PPG waveform pattern with a reaction-diffusion model, with diastolic and systolic phases of a heart coupled with the reaction and diffusion properties of the reaction-diffusion model;
apply the pattern recognition by being configured to apply cross-correlation analysis between the PPG waveforms in the first time series and the reference PPG waveform pattern; and
produce a resulting PPG signal by retaining the PPG waveforms in the first time series having a recognition score higher than or equal to a recognition threshold, and discarding the PPG waveforms in the first time series having a recognition score lower than the recognition threshold.
2. The system of claim 1 , wherein the PPG sensing apparatus is further configured to receive an electrocardiogramal waveforms and wherein the PPG signal processor is further configured to process the ECG signal waveforms.
3. The system of claim 2 , wherein the PPG signal processor is configured to process the ECG signal waveforms by:
calculating a first derivative of the resulting PPG signal;
performing cross-correlation of the ECG signal waveforms and the first derivative of the resulting PPG signal by assigning to the ECG signal waveforms cross-correlation scores with the first derivative of the resulting PPG signal;
comparing with a validation threshold the cross-correlation scores of the ECG signal waveforms; and
validating as valid ECG signal waveforms the ECG signal waveforms having cross-correlation scores higher than or equal to the validation threshold.
4. The system of claim 3 , wherein the PPG signal processor is further configured to process the ECG signal waveforms by:
performing cross-correlation of the ECG signal waveforms and an ECG reference waveform by assigning to the ECG signal waveforms second cross-correlation scores with the ECG reference waveform;
comparing with a second validation threshold the second cross-correlation scores of the ECG signal waveforms; and
validating as valid ECG signal waveforms the ECG signal waveforms having both the cross-correlation scores higher than or equal to the validation threshold and the second cross-correlation scores higher or equal to the second validation threshold.
5. The system of claim 1 , further comprising a bandpass filter configured to filter the PPG signal received by the PPG sensing apparatus.
6. A system comprising:
a processor; and
a non-transitory computer readable medium coupled to the processor, the non-transitory computer readable medium comprising instructions that, when executed on the processor, cause the processor to:
detect peaks and valleys in a photoplethysmography (PPG) signal;
segment the PPG signal to provide a first time series of PPG waveforms located between two subsequent valleys in the PPG signal;
apply pattern recognition to the first time series with respect to a reference PPG waveform pattern produced based on a mathematical model of the PPG signal by assigning a recognition score to the PPG waveforms in the first time series;
produce the reference PPG waveform pattern with a reaction-diffusion model, with diastolic and systolic phases of a heart coupled with the reaction and diffusion properties of the reaction-diffusion model, and wherein the instructions to apply the pattern recognition comprises instructions to apply cross-correlation analysis between the PPG waveforms in the first time series and the reference PPG waveform pattern;
retain the PPG waveforms in the first time series having a recognition score higher or equal to a recognition threshold; and
discard the PPG waveforms in the first time series having a recognition score lower than the recognition threshold to produce a resulting PPG signal.
7. The system of claim 6 , wherein the non-transitory computer readable medium comprises further instructions that, when executed on the processor, cause the processor to:
calculate a first derivative of the resulting PPG signal;
perform cross-correlation of ECG signal waveforms and the first derivative of the resulting PPG signal by assigning to the ECG signal waveforms cross-correlation scores with the first derivative of the resulting PPG signal;
compare with a validation threshold the cross-correlation scores of the ECG signal waveforms; and
validate as valid ECG signal waveforms the ECG signal waveforms having cross-correlation scores higher or equal to the validation threshold.
8. The system of claim 6 , further comprising a bandpass filter configured to filter the PPG signal.
9. The system of claim 8 , wherein the bandpass filter comprises a joint low-pass and high-pass filter.
10. The system of claim 6 , wherein the instructions to detect the peaks and valleys comprises instructions to calculate first and second derivatives of the PPG signal.
11. The system of claim 6 , wherein the instructions further comprise instructions to normalize the PPG signal to a unitary range prior to the segmenting.
12. The system of claim 6 , wherein the instructions to produce the reference PPG waveform pattern comprise instructions to produce via a self-adaptive nonlinear oscillator or with a neural network, and wherein the instructions further comprise instructions to rescale the reference PPG waveform pattern over time to facilitate applying the pattern recognition to time-comparable waveforms.
13. A system comprising:
a processor; and
a non-transitory computer readable medium coupled to the processor, the non-transitory computer readable medium comprising instructions that, when executed on the processor, cause the processor to:
receive a photoplethysmography (PPG) signal collected by a PPG sensing apparatus;
detect peaks and valleys in the PPG signal;
segment the PPG signal to provide a first time series of PPG waveforms located between two subsequent valleys in the PPG signal;
produce a reference PPG waveform pattern based on a mathematical model of the PPG signal, the reference PPG waveform pattern being produced with a reaction-diffusion model, with diastolic and systolic phases of a heart coupled with the reaction and diffusion properties of the reaction-diffusion model;
apply pattern recognition to the first time series with respect to the reference PPG waveform pattern by assigning a recognition score to the PPG waveforms in the first time series, wherein applying the pattern recognition comprises applying cross-correlation analysis between the PPG waveforms in the first time series and the reference PPG waveform pattern; and
produce a resulting PPG signal by retaining the PPG waveforms in the first time series having a recognition score higher or equal to a recognition threshold, and discarding the PPG waveforms in the first time series having a recognition score lower than the recognition threshold.
14. The system of claim 13 , further comprising a bandpass filter configured to filter the PPG signal collected by the PPG sensing apparatus.
15. The system of claim 14 , wherein the bandpass filter comprises a joint low-pass and high-pass filter.
16. The system of claim 13 , wherein the instructions to detect the peaks and valleys comprises the instructions to calculate first and second derivatives of the PPG signal.
17. The system of claim 13 , wherein the instructions further comprise instructions to normalize the PPG signal to a unitary range prior to the segmenting.
18. The system of claim 13 , wherein the instructions to produce the reference PPG waveform pattern comprise instructions to produce via a self-adaptive nonlinear oscillator.
19. The system of claim 13 , wherein the instructions to produce the reference PPG waveform pattern comprise instructions to produce with a neural network.
20. The system of claim 13 , wherein the instructions further comprise instructions to rescale the reference PPG waveform pattern over time to facilitate applying pattern recognition to time-comparable waveforms.