Method and system for extracting cardiac parameters from plethysmographic signals

1. A method for determining cardiac parameters of a subject, comprising:

receiving a signal from a thoracocardiograph (TCG) sensor, the signal being sensitive to positions and/or motions of an anterior chest wall of the subject, and the signal comprising a cardiac component, a respiratory component, and noise and/or artifact components;

receiving one or more electrocardiogram (ECG) signals;

filtering the received TCG signal such that one or more of the respiratory component and the noise and/or artifact components in the received TCG signal are reduced, the filtering comprising one or more of wavelet de-noising, non-linear filtering, and state space filtering;

ensemble averaging the filtered TCG signal such that a respiratory component remaining in the filtered TCG signal is reduced, ensemble members being triggered by occurrence of one or more selected fiducial points determined in the ECG signal; and

extracting parameters of cardiac functioning from the ensemble averaged signal.

2. The method of claim 1 wherein the extracted parameters of cardiac functioning comprise one or more of stroke volume, cardiac output, pre-ejection period, peak ejection rate, time to peak ejection rate, peak ejection rate divided by stroke volume, 50% filling fraction, peak filling rate, E/A ratio, deceleration time, indicators of a presence or absence of myocardial ischemia, and indicators of a severity of myocardial ischemia if present.

3. The method of claim 1 wherein the extracted parameters of cardiac functioning comprise indicators of dyskinetic ventricular wall motion, the method further comprising determining whether or not myocardial ischemia is likely to be present, and if present, determining a severity of ischemia in dependence on the indicators of dyskinetic motion.

4. The method of claim 1 further comprising determining whether or not hypovolemia is likely to be present by comparing current values of at least a stroke volume and cardiac output selected parameters with prior values of at least the stroke volume and cardiac output.

5. The method of claim 1 further comprising determining whether or not congestive heart failure (CHF) is likely to be present, and if present, determining a severity of the CHF in dependence on values of selected parameters.

6. The method of claim 1 wherein the fiducial points in the ECG signal selected for triggering the ensemble averaging comprise one or more of a P wave, a Q wave, a R wave, a S wave, and a T wave.

7. The method of claim 1 further comprising:

determining whether artifacts and/or noise are likely to be present in a signal during a period of time if either a fiducial point in an input signal that is expected to occur during the period cannot be recognized during the period, or a fiducial point occurs during the period at a time that is not expected; and

excluding all signals during the period if artifacts and/or noise are likely to be present in any signal.

8. The method of claim 7 wherein an expected fiducial point is selected from one or more of P, Q, R, S, and T waves in the ECG signal; the extracted parameters of cardiac functioning; and features of ventricular wall motion and a first derivative of the ventricular wall motion including maxima, minima, or zero crossings.

9. The method of claim 7 further comprising determining an expected occurrence time of a selected fiducial point during a period in dependence on a plurality of prior occurrence times of the selected fiducial point.

10. The method of claim 1 , wherein the filtering comprises the wavelet de-noising.

11. The method of claim 1 , wherein the filtering comprises the state space filtering.

12. The method of claim 1 , wherein the filtering comprises the non-linear filtering.