Multisensor cardiac function monitoring and analytics systems

1. An apparatus for monitoring cardiac function, the apparatus comprising:

(a) a plurality of Integrated Cardio Respiratory (ICR) acoustic sensors configured to be positioned on the chest of a patient;

(b) a processor coupled to the plurality of ICR acoustic sensors; and

(c) a non-transitory memory storing instructions executable by the processor;

(d) wherein said instructions, when executed by the processor, perform steps comprising:

(i) receiving a phonocardiogram (PCG) acoustic signal from the plurality of ICR acoustic sensors;

(ii) segmenting the PCG acoustic signal to locate one or more cardiac events in the PCG acoustic signal;

(iii) extracting one or more of temporal and amplitude characteristics from the PCG acoustic signal;

(iv) classifying the extracted characteristics into one or more subgroups with a classifier; and

(v) computing ejection fraction (EF) based on the classified characteristics.

2. The apparatus of claim 1 , wherein segmenting the PCG acoustic signal comprises:

detecting heart sounds within the PCG acoustic signal;

identifying the heart sounds based on predefined criteria;

labeling heart sounds as S1 and S2 based on an interval between successive events; and

decomposing the PCG signal into individual cardiac cycles.

3. The apparatus of claim 1 , further comprising:

(e) a plurality of ECG sensors configured to be positioned on the chest of a patient and coupled to the processor;

(f) wherein said instructions, when executed by the processor, further perform steps comprising:

(i) receiving an ECG sensor signal from the plurality of ECG sensors;

(ii) processing the ECG sensor signal to calculate an R wave onset in the ECG sensor signal;

(iii) wherein the R wave onset is used enable timing and identification of PCG acoustic signatures within the cardiac cycle.

4. The apparatus of claim 3 , wherein identification of PCG acoustic signatures comprises identification of S1 and S2 events within the cardiac cycle.

5. The apparatus of claim 4 , wherein the extracted temporal characteristics comprise one or more of:

electromechanical activation time (EMAT), total electromechanical systolic interval (QS 2 ), S1 duration, S2 duration, mitral cessation-to-opening time (MCOT), left ventricular ejection time (LVET), and left-ventricular systolic time (LVST).

6. The apparatus of claim 4 :

wherein the PCG signal is analyzed in within an envelope segment containing two consecutive cardiac cycles; and

wherein the extracted amplitude characteristics comprise one or more of: the root-mean-square (RMS) of the PCG signal envelope segment normalized by RMS of the PCG signal of the entire cardiac cycle; the peak amplitude of the PCG signal segment, normalized by variance of the PCG signal of the entire cardiac cycle; and the peak amplitude of envelope segment, normalized by the envelope mean value for the entire cardiac cycle.

7. The apparatus of claim 3 , wherein processing the ECG sensor signal to calculate an R wave onset in the signal comprises:

(a) band-pass filtering the ECG sensor signal;

(b) multiplying the filtered signal by its derivative;

(c) computing an envelope of the multiplied signal;

(d) identifying R waves in the computed envelope;

(e) identifying corresponding peaks in the filtered signal; and

(f) determining an R wave onset in filtered signal.

8. The apparatus of claim 1 , further comprising:

extracting one or more frequency characteristics from the PCG signal by performing Phonocardiogram Carrier Signal Analysis (PCSA).

9. The apparatus of claim 8 , wherein PCSA comprises AM-FM decomposition of S1 and S2 events within the PCG signal to yield a carrier signal with unit amplitude.

10. The apparatus of claim 1 , wherein classifying the extracted characteristics comprises applying a Neural Network (NN) across a plurality of patients to generate a global classifier, and assigning patients to a set of subgroups based on a computed EF value, where each such subgroup has its own NN sub-classifier.

11. The apparatus of claim 1 , wherein the plurality of ICR acoustic sensors disposed within an ICR sensor support configured to support the ICR acoustic sensors on the patient at locations based on typical auscultatory sites.

12. The apparatus of claim 1 , wherein the instructions are further configured for:

preprocessing the PCG acoustic signal using Short-Time Spectral Amplitude Log Minimum Mean Square Error (STSA-log-MMSE) noise suppression;

wherein timing of the cardiac cycle based the acquired R wave onset is used to determine regions of acoustic inactivity as an input to STSA-log-MMSE.

13. A method for monitoring cardiac function, the method comprising:

receiving a phonocardiogram (PCG) acoustic signal from the plurality of ICR acoustic sensors positioned on the chest of a patient;

segmenting the PCG acoustic signal to locate one or more cardiac events in the PCG acoustic signal;

extracting one or more of temporal and amplitude characteristics from the PCG acoustic signal;

classifying the extracted characteristics into one or more subgroups;

computing ejection fraction (EF) based on the classified characteristics; and

outputting the computed EF for display;

wherein said method is performed by a processor executing instructions stored on a non-transitory memory.

14. The method of claim 13 , wherein segmenting the PCG acoustic signal comprises:

detecting heart sounds within the PCG acoustic signal;

identifying the heart sounds based on predefined criteria;

labeling heart sounds as S1 and S2 based on an interval between successive events; and

decomposing the PCG signal into individual cardiac cycles.

15. The method of claim 13 , further comprising:

receiving an ECG sensor signal, concurrently with reception of the PCG acoustic signal, from the plurality of ECG sensors positioned on the chest of the patient

processing the ECG sensor signal to calculate an R wave onset in the ECG sensor signal; and

wherein the R wave onset is used enable timing and identification of PCG acoustic signatures within the cardiac cycle.

16. The method of claim 15 , wherein identification of PCG acoustic signatures comprises identification of S1 and S2 events within the cardiac cycle.

17. The method of claim 16 , wherein the extracted temporal characteristics comprise one or more of:

electromechanical activation time (EMAT), total electromechanical systolic interval (QS 2 ), S1 duration, S2 duration, mitral cessation-to-opening time (MCOT), left ventricular ejection time (LVET), and left-ventricular systolic time (LVST).

18. The method of claim 16 :

wherein the PCG signal is analyzed in within an envelope segment containing two consecutive cardiac cycles; and

wherein the extracted amplitude characteristics comprise one or more of: the root-mean-square (RMS) of the PCG signal envelope segment normalized by RMS of the PCG signal of the entire cardiac cycle; the peak amplitude of the PCG signal segment, normalized by variance of the PCG signal of the entire cardiac cycle; and the peak amplitude of envelope segment, normalized by the envelope mean value for the entire cardiac cycle.

19. The method of claim 13 , further comprising:

extracting one or more frequency characteristics from the PCG signal by performing Phonocardiogram Carrier Signal Analysis (PCSA).

20. The method of claim 19 , wherein PCSA comprises AM-FM decomposition of S1 and S2 events within the PCG signal to yield a carrier signal with unit amplitude.

21. The method of claim 15 , wherein processing the ECG sensor signal to calculate an R wave onset in the signal comprises:

band-pass filtering the ECG sensor signal;

multiplying the filtered signal by its derivative;

computing an envelope of the multiplied signal;

identifying R waves in the computed envelope;

identifying corresponding peaks in the filtered signal; and

determining an R wave onset in filtered signal.

22. The method of claim 13 , wherein classifying the extracted characteristics comprises applying a Neural Network (NN) across a plurality of patients to generate a global classifier, and assigning patients to a set of subgroups based on a computed EF value, where each such subgroup has its own NN sub-classifier.

23. The method of claim 13 , wherein the instructions are further configured for:

preprocessing the PCG acoustic signal using Short-Time Spectral Amplitude Log Minimum Mean Square Error (STSA-log-MMSE) noise suppression;

wherein timing of the cardiac cycle based the acquired R wave onset is used to determine regions of acoustic inactivity as an input to STSA-log-MMSE.