Transtelephonic monitoring of multi-channel ECG waveforms

• US 6,377,843 B1
• Filed: 03/03/2000
• Issued: 04/23/2002
• Est. Priority Date: 03/03/2000
• Status: Expired due to Term

• Current Assignee: Medtronic USA Incorporated (Medtronic Plc)
• Original Assignee: Paceart Associates L.P.
• Inventors: Marchesini, Anthony, Naydenov, Nartzis
• Primary Examiner(s): Bockelman, Mark
• Application Number:
  US09/518,546
• Time in Patent Office: 781 Days
• Field of Search: 128/903-904, 600/508-509
• US Class Current: 600/509
• CPC Class Codes: A61B 5/0006 ECG or EEG signals
• CPC Class Codes: A61B 5/30 Input circuits therefor
• CPC Class Codes: A61B 5/308 for electrocardiography [ECG]
• CPC Class Codes: A61B 5/7257 using Fourier transforms
• CPC Class Codes: Y10S 128/904 Telephone telemetry

Abstract

Remote monitoring of biological signals such as ECG waveforms, in which an FM encoded version of an ECG waveform is transmitted from a patient'"'"'s home to a central station where it is analyzed to estimate local frequencies at many more time points than the zero crossovers of the FM signal. A plurality of ECG waveforms is encoded in a composite FM signal any one point of which can represent at the same time a number of ECG waveforms. The composite waveform is analyzed at the central station to extract and reconstruct the individual waveforms by separation in the frequency domain.

Claims

1. A method of remotely monitoring at a central station cardiac conditions existing at a remote station, comprising:

deriving three or more ECG waveforms at a local station and frequency modulating a respective carrier in accordance with each of said waveforms to thereby produce three or more respective FM signals;

combining the FM signals into a composite FM signal containing concurrent information from the three or more ECG waveforms;

transmitting the composite FM signal to a central station;

processing the composite FM signal received at the central station both in the time domain and in the frequency domain to reconstruct the three or more individual ECG waveforms.

2. A method as in claim 1 in which said processing comprises estimating local frequencies at portions of the composite FM signal that are substantially closer to each other than zero crossovers of the composite FM signal.

3. A method as in claim 2 in which said processing comprises estimating local phase differences at portions of the FM signal that are substantially closer than said zero crossovers and using the local phase differences to estimate said local frequencies.

4. A method as in claim 3 in which the estimating of local phase differences comprises combining digital samples of the composite FM signal with a phase shifted version of the digital samples.

5. A method as in claim 4 in which the comparing comprises comparing the samples with a Hilbert-transformed version of the samples.

6. A method as in claim 5 in which the processing comprises digitizing the composite FM signal into digital samples in the time domain, converting the time domain samples into frequency domain samples and filtering and separating the frequency domain samples into groups related to the respective ECG waveforms, and converting a Hilbert-transformed version of the frequency domain samples into the time domain.

7. A method of reconstructing at a central station an ECG waveform derived at a remote station, comprising the steps of:

deriving an ECG waveform at a remote station;

frequency modulating a carrier in accordance with the ECG waveform to generate an FM signal related to the ECG waveform and having zero crossovers;

transmitting the FM signal to a central station;

processing the FM signal at the central station to estimate local frequencies thereof at times occurring substantially more frequently than the zero crossovers of the FM signal; and

using the estimated local frequencies to reconstruct the ECG waveform at the central station.

8. A method as in claim 7 in which the local frequencies are estimated at a rate at least equal to the carrier frequency.

9. A method as in claim 7 in which the local frequencies are estimated at least at hundreds of time slots between successive zero crossovers.

10. A method as in claim 7 in which the processing comprises processing a digitized version of the FM signal in both the time domain and the frequency domain.

11. A method as in claim 7 in which said processing comprises comparing a version of the FM signal with a phase shifted version thereof to estimate local phase differences for successive samples of the FM signal, and using the estimated phase differences to reconstruct the FM signal.

12. A method of reconstructing an ECG waveform from an FM signal comprising a carrier that is frequency modulated with the ECG waveform, comprising:

digitizing the FM signal into first time domain digital values representing local amplitudes of the FM signal;

converting the time domain digital values into frequency domain digital values representing content of the FM signal at respective frequencies within a band of interest for said ECG waveform;

generating a phase-shifted version of the frequency domain digital values;

converting the frequency domain digital values and the phase-shifted version thereof to respective second and third time-domain digital values;

using said second and third time domain digital values to calculate local frequencies at locations in said FM signal corresponding to at least twice the carrier frequency; and

using said local frequencies to reconstruct said ECG waveform.

13. A method as in claim 12 in which the converting to the frequency domain comprises using an FFT process.

14. A method as in claim 12 in which the generating of a phase-shifted version comprises using a Hilbert transform in digital form.

15. A method as in claim 12 including filtering out frequencies outside said band of interest for the ECG waveform.

16. A method of reconstructing a plurality of ECG waveforms from a composite FM signal formed by combining a plurality of carriers FM encoded with respective ones of the ECG waveforms such that a time slot of the composing signal contains information from each of a plurality the ECG waveforms, said composite waveforms having zero crossovers, comprising the steps of:

digitizing the composite signal;

processing the composite signal in the digital domain to estimate local frequencies thereof at times occurring substantially more frequently that said zero crossovers; and

using the estimated local frequencies to reconstruct said ECG waveforms.

17. A method as in claim 16 in which said processing comprises processing the composite signal both in the time domain and in the frequency domain.

18. A method as in claim 17 in which said processing comprises filtering a frequency domain version of the composite signals and separating out portions thereof corresponding to frequency bands of the respective ECG waveforms.

19. A method as in claim 18 in which said processing comprises comparing a version of said composite signal with a phase-shifted version thereof to estimate local phase shifts, and using the estimated local phase shifts in estimating said local frequencies.

20. A method as in claim 19 in which said comparing comprises comparing a version of the composite signal with a Hilbert-transformed version thereof.