Implantable medical device with autosensitivity algorithm for controlling sensing of cardiac signals

• US 6,745,076 B2
• Filed: 11/16/2001
• Issued: 06/01/2004
• Est. Priority Date: 11/16/2001
• Status: Active Grant

• Current Assignee: Medtronic Incorporated (Medtronic Plc)
• Original Assignee: Medtronic Incorporated (Medtronic Plc)
• Inventors: Wohlgemuth, Peter, Westendorp, Henk A., De Bruyn, Harry W. M., Munneke, J. Dave
• Primary Examiner(s): MANUEL, GEORGE C
• Application Number:
  US09/987,777
• Publication Number:
  US 20030097157A1
• Time in Patent Office: 928 Days
• Field of Search: 607/9, 607/11, 607/27, 607/28, 607/29, 607/60, 128/901
• US Class Current: 607/27
• CPC Class Codes: A61N 1/3704 Circuits specially adapted therefor, e.g. for sensitivity control
• CPC Class Codes: A61N 1/3718 Monitoring of or protection against external electromagnetic fields or currents

Abstract

A cardiac pacing system and method incorporate DSP processing and software algorithms for collecting signal amplitude data and noise data, and organizing the data for automatic checking of signal channel gain and signal detection sensitivity. Unfiltered signals are used to obtain values representative of maximum amplitude, which values are stored in a gain histogram, from which determination of the percentage of clipped signals can be made. Gain is adjusted by limiting clipping to a predetermined range of allowed clipping, to optimize use of the DSP range. The signals, both P waves and R waves for a dual chamber system, are also processed by DSP and parameters representing maximum amplitudes are stored in amplitude histograms. At the same time, noise is analyzed for respective windows of time following each ventricular event, and noise amplitude data is stored in noise histograms. After a predetermined waiting period, the signal amplitude and noise histograms are analyzed and compared, and for each channel the sensitivity is adjusted to fall between a calculated noise floor and the lowest bin of the amplitude histogram that contains valid data.

Claims

1. A cardiac pacing system having a pacing device and a lead for connecting patient cardiac signals to said pacing device, said device having at least one DSP channel having DSP means for amplifying and processing said cardiac signals, said DSP means further having sensitivity means for setting the sensitivity level for use in detection of a said cardiac signal, gain control means for adjusting the gain used in said amplifying, and event means for determining when a said cardiac signal is determined to be a valid cardiac event, further comprising:

amplitude sample means for enabling said DSP means to amplify and process cardiac signals periodically at a first sampling interval in the range of 1 minute to 60 minutes;

amplitude means for obtaining and storing signal amplitude data representative of the amplitude of sampled cardiac signals that have been amplified by said DSP means;

noise sample means for enabling said DSP means to sample noise in said DSP channel following each sampled cardiac signal;

noise means for obtaining and storing noise amplitude data representative of sampled noise in said DSP channel; and

check means for comparing said stored signal amplitude data with said stored noise amplitude data periodically at a predetermined check interval greater than at least one hour and for adjusting said sensitivity level in accord with each said comparison.

2. The system as described in claim 1, wherein said gain control means comprises:

clipping means for determining when a sampled signal has been clipped, means for maintaining a running measure of the percentage of sampled signals that have been clipped, and reducing means for reducing gain when said percentage is above a first predetermined clipping percentage.

3. The system as described in claim 2, wherein said gain control means comprises increasing means for increasing gain when said percentage is below a second predetermined clipping percentage.

4. The system as described in claim 3, wherein said gain control means comprises means for setting a clipping level, and gain hysteresis means for setting said first clipping percentage at a predetermined level above said clipping level and for setting said second clipping percentage at a predetermined level below said clipping level.

5. The system as described in claim 2, wherein said gain control means comprises gain histogram means for storing data representative of maximum values of sampled signal amplitudes in a gain histogram having n bins, each said bin containing counts of sampled signals within a respective amplitude range, gain calculation means for calculating gain to be one of n gain levels depending on the contents of said bins, and shift means operative whenever there is a change in gain for shifting the data in said gain histogram to correspond to the change in said gain to a different level.

6. The system as described in claim 5, comprising means for adjusting said DSP sensitivity level to correspond to said calculated gain.

7. The system as described in claim 6, comprising divide means for multiplying the data in each bin of said gain histogram by a predetermined factor in the range of 0.1 to 0.9 after each said check interval to decrease the historical influence of said data.

8. The system as described in claim 5, wherein one of said bins contains a count of signals that were clipped, and said reducing means has means for determining when said count of clipped signals compared to all of the counts in said gain histogram exceeds said first percentage.

9. The system as described in claim 1, wherein said DSP means comprises max signal means for providing digital counts of the maximum amplitude values of said sampled signals, and said amplitude means comprises signal voltage means for converting said digital counts into voltage values and for classifying and storing each said signal voltage value into one of n signal histogram bins.

10. The system as described in claim 9, wherein said DSP means comprises max noise means for providing digital counts of the maximum amplitude values of said sampled noise, and said noise means comprises noise voltage means for converting said noise digital counts into noise voltage values and for classifying and storing each said noise voltage value into one of n noise histogram bins.

11. The system as described in claim 1, comprising check time means for determining when a predetermined check interval in the range of one hour to 1 month has passed, sufficiency means operative at least every check interval for determining if a predetermined sufficient amount of noise amplitude and signal amplitude data has been collected, and activate means for activating said check means whenever such sufficient amount of data has been collected.

12. The system as described in claim 11, wherein said check time means comprises stored interval data representing a check interval in the range of 1 to 14 days, and said sufficiency means comprises stored data representing a count of 10 to 1000 events as a sufficient amount of data.

13. The system as described in claim 11, wherein said amplitude means comprises amplitude histogram means for storing signal amplitude data representative of maximum values of signal amplitudes in a signal amplitude histogram having n bins, each said signal amplitude bin containing counts of sampled signals having representative maximum values within a respective amplitude range, and said noise means comprises noise amplitude means for storing noise data representative of maximum values of noise amplitudes in a noise histogram having n bins, each said noise bin containing counts of sampled noise having representative maximum values within a respective noise amplitude range.

14. The system as described in claim 13, wherein said check means comprises:

means for discarding data in the lowest bins of said signal amplitude histogram that represent counts up to but no more than a first predetermined percentage of all the counts in said signal amplitude histogram, and then determining the lowest signal amplitude from the lowest remaining bin of said signal amplitude histogram;

means for discarding data in the highest bins of said noise histogram that represent counts up to but no more than a second predetermined percentage of all the counts in said noise histogram and then determining a noise floor from the highest remaining bin of said noise histogram;

gap means for determining the gap between said noise floor and said lowest signal amplitude; and

adjust means for adjusting said sensitivity level to a value represented by said noise floor plus a fraction of said gap.

15. The system as described in claim 14, wherein said DSP channel is an R wave channel, and said adjust means comprises means for setting said fraction to about ½

for R wave detection.

16. The system as described in claim 14, wherein said DSP channel is a P wave channel, and said adjust means comprises means for setting said fraction to about ⅓

for P wave detection.

17. The system as described in claim 14, comprising means for setting each of said first and second predetermined percentages at a level less than about 50%.

18. The system as described in claim 14, comprising means for setting each of said first and second predetermined percentages at about 5%.

19. The system as described in claim 14, comprising multiply means for multiplying the counts in each bin of each of said histograms by a fraction less than one following each adjustment of sensitivity level.

20. The system as described in claim 19, wherein said multiply means comprises means for setting said fraction to a value between ½

and ⅞

.

21. An implantable medical device system, said system having signal channel means for receiving and processing medical signals from the patient in whom it is implanted, said channel means having sensitivity means for setting a sensitivity threshold for discriminating between valid signals and noise, said channel means further comprising:

amplitude sample means for sampling a received signal for a first sample time every predetermined first sample interval and for storing data representative of the amplitude of said sampled signal;

noise sample means for sampling noise in said channel for a second sample interval that has a predetermined time relation to said first sample time, and for storing data representative of the sampled noise;

event analysis means for determining a valid signal only when a sampled signal has an amplitude above said sensitivity threshold;

an amplifier circuit having a signal range with an upper limit and adjustable gain;

adjust means operable every second interval for processing said stored amplitude data and said stored noise data, said adjust means having gain adjust means for adjusting said gain as a function of the percentage of said sampled signals having a maximum signal amplitude above said upper limit and sensitivity adjust means for adjusting said sensitivity threshold as a function of said stored amplitude data and said stored noise data; and

data reduction means for reducing said stored noise data and said stored amplitude data before continuing sampling of data in a manner that gives greater weight to data taken during each second interval than data stored in prior second intervals.

22. The system as described in claim 21, wherein said event analysis means comprises DSP circuitry and means for setting a digital threshold corresponding to said adjusted sensitivity threshold.

23. The system as described in claim 21, wherein said amplitude sample means comprises means for programming said first sample interval to a value within the range of 1 hour to 1 day.

24. The system as described in claim 23, wherein said amplitude sample means comprises signal amplitude histogram means for storing data in n bins, each said bin containing counts of sampled signals having amplitudes within a predetermined range.

25. The system as described in claim 21, wherein said noise sample means comprises noise histogram means for storing data in n bins, each said bin containing counts of sampled noise having amplitudes within a predetermined range.

26. The system as described in claim 21, wherein said noise sample means comprises means for setting said second sample interval to follow said first sample interval by a predetermined delay.

27. The system as described in claim 21, wherein said gain circuit is a DSP circuit and has means for adjusting the gain to one of n gain levels.

28. The system as described in claim 27, wherein said gain adjust means comprises gain histogram means for collecting and storing signal amplitude data in a gain histogram having n bins.

29. The system as described in claim 28, wherein said gain histogram means comprises clipping means for counting in the highest bin the sampled signals that have been clipped.

30. The system as described in claim 29, comprising wherein said gain adjust means comprises means for adjusting said gain when the percentage of clipped signals is greater than a predetermined percentage.

31. The system as described in claim 30, comprising bin shift means for shifting the data in said gain histogram each time that said gain is adjusted.

32. The system as described in claim 21, wherein said adjust means comprises means for programming said second interval to have a value within a range of 1 hour to 1 month.

33. The system as described in claim 32, wherein said data reduction means comprises means for reducing said stored amplitude data and said stored noise data following adjustment of sensitivity threshold each said second interval.

34. The system as described in claim 33, wherein said data reduction means comprises means for reducing said stored amplitude data and said stored noise data by a programmable factor within a range that reduces said data by 0 to 50%.

35. The system as described in claim 21, wherein said event analysis means comprises DSP circuitry, and further comprising software for starting an event analysis during a timed analysis window each said first sample time.

36. The system as described in claim 35, wherein said sensitivity adjust means comprises software for comparing said signal amplitude data and said noise data.

37. The system as described in claim 36, wherein said software comprises means for constructing a signal amplitude histogram and a noise histogram, and comparison means for comparing the signal amplitude bin data with the noise bin data.

38. The system as described in claim 37, wherein said software comprises means for determining a gap between the signal bin data and the noise bin data, and for setting the sensitivity threshold as a function of said gap.

39. The system as described in claim 38, wherein said software comprises means for establishing each of said histograms with exponentially widening bins and said amplifier has exponentially increasing gain levels, whereby recalculation of the bin values is easily done following any gain change.

40. A method of receiving and processing signals from a patient'"'"'s heart, the method employing an implantable signal channel, comprising:

setting a first interval of 1 minute to 1 hour, and periodically sampling every said first interval to obtain a heart signal, and storing data relating to each said sampled signal;

periodically sampling, during a time period in a predetermined time relation to each said sampling of a signal, noise present in said channel, and storing data relating to said noise;

periodically checking sensitivity in accord with predetermined check criteria, said checking being done after a plurality of said first intervals, said checking comprising;

removing a first amount of stored signal data according to predetermined signal error criteria to obtain adjusted signal data;

removing a second amount of stored noise data according to predetermined noise error criteria to obtain adjusted noise data; and

calculating and setting a sensitivity threshold for detection of valid signals as a function of said adjusted signal and noise data.

41. The method as described in claim 40, comprising:

calculating a measure of the maximum amplitude of each said sampled signal, and

determining valid signals as being received only when a said sampled signal has an amplitude measure greater than said sensitivity threshold.

42. The method as described in claim 40, comprising setting said first interval at less than 10 minutes.

43. The method as described in claim 40, comprising periodically checking every second interval, where said second interval is a programmed interval in the range of 1 hour to 1 month.

44. The method as described in claim 43, wherein said checking in accord with predetermined criteria includes the step of determining if there is at least a predetermined amount of sampled signal data and sampled noise data.

45. The method as described in claim 40, comprising amplifying said received signals with a gain adjustable to one of n gain levels, and storing data representative of sampled signal amplitudes in a gain histogram having n bins.

46. The method as described in claim 45, comprising counting clipped signals in the highest bin of said gain histogram, and measuring the percentage of events in said highest bin as the percentage of clipped signals.

47. The method as described in claim 46, comprising adjusting said gain lower when said clipping percentage is greater than a first predetermined percentage.

48. The method as described in claim 47, comprising adjusting said gain higher when said clipping percentage is less than a second predetermined minimum percentage.

49. The method as described in claim 40, comprising storing said sampled signal data in a signal histogram, and storing said noise data in a noise histogram.

50. The method as described in claim 49, comprising removing bin signal data from at least the lowest filled bin of said signal histogram if said bin signal data represents less than a predetermined percentage of the total data in said histogram, and removing bin noise data from at least the highest filled bin of said noise histogram if said bin noise data represents less than a predetermined percentage of the total data in said noise histogram.

51. The method as described in claim 50, comprising determining a gap between the highest remaining noise bin and the lowest remaining signal bin, and setting said sensitivity level within said gap.

52. The method as described in claim 51, comprising multiplying the counts in each of said signal and noise histograms by a factor less than 1 following each setting of sensitivity threshold.

53. The method as described in claim 40, comprising sampling R waves from a patient'"'"'s heart and setting a sensitivity threshold for detection of valid R wave events.

54. The method as described in claim 53, comprising sampling P waves from a patient'"'"'s heart and setting a sensitivity threshold for detection of valid P wave events.

55. An implantable cardiac system, having at least one signal processing channel for receiving cardiac signals, said channel having an amplifier circuit with an adjustable gain and having event validation means for validating when a received signal is valid and is not noise, comprising:

gain histogram means for collecting gain data representative of the amplification of sampled ones of received signals and storing said gain data in a gain histogram;

amplitude histogram means for collecting amplitude data representative of the amplitude of said sampled received signals and storing said amplitude data in an amplitude histogram;

noise histogram means for collecting noise data representative of channel noise associated with each sampled signal and storing said noise data in a noise histogram; and

setting the gain of said amplifier circuit and setting a sensitivity threshold for said event validation means on a periodic basis and as a function of the data in said histograms.

56. The system as described in claim 55, comprising sampling said received cardiac signals at an interval within the range of 1 minute to 1 hour, carrying out said setting of gain and sensitivity threshold after at least a predetermined amount of data has been obtained and stored in each of said histograms, and reducing the data in each of said histograms by a respective predetermined measure after each said setting of gain and sensitivity threshold.