System and method of data compression and non-linear sampling for implantable and battery-powered devices

• US 5,836,982 A
• Filed: 02/19/1997
• Issued: 11/17/1998
• Est. Priority Date: 02/19/1997
• Status: Expired due to Term

• Current Assignee: Michael J. Jaro
• Original Assignee: Medtronic Incorporated (Medtronic Plc)
• Inventors: Muhlenberg, Lambert, Weijand, Keon
• Primary Examiner(s): Kramm, William E.
• Assistant Examiner(s): LAYNO, CARL HERNANDZ
• Application Number:
  US08/801,335
• Time in Patent Office: 636 Days
• Field of Search: 607/9, 600/515
• US Class Current: 607/9
• CPC Class Codes: A61N 1/37276 characterised by means for reducing power consumption during telemetry
• CPC Class Codes: H03M 1/1205 Multiplexed conversion systems
• CPC Class Codes: H03M 1/1235 Non-linear conversion not otherwise provided for in subgroups of H03M1/12
• CPC Class Codes: H03M 1/127 at intervals varying with the rate of change of the input signal
• CPC Class Codes: H03M 7/42 using table look-up for the coding or decoding process, e.g. using read-only memory H03M7/4006 takes precedence

Abstract

A system and method of non-linear sampling of an analog signal, and for data compression, to provide overall compressed digital representative of analog signal such as a physiological signal obtained in a battery powered device such as an implantable pacemaker. Non-linear sampling is provided by generating a time varying threshold signal against which the analog signal is compared every clock sample, with the sample being skipped if the input analog signal does not exceed the threshold signal. The threshold signal can be either analog or a plurality of discrete levels, enabling up to a predetermined maximum number of samples to be skipped. The non-linearly sampled data is further compressed, using lossless techniques for compressing both the amplitude data and the data representative of skipped samples, to provide data for decompression. In a specific pacemaker embodiment, the data compression system recognizing a marker channel event, and includes data representative of the time and type of marker channel event.

Claims

1. A system of non-linear sampling of an analog signal for obtaining compressed digital data representative of said analog signal, comprising:

a source for producing sample clock signals at a predetermined rate;

sampling means for sampling said analog signal with clock signals to obtain digital representations of said analog signal;

holding means for holding each sampled value of said analog signal;

threshold means for generating, upon each sampling of said analog signal, a difference threshold signal representing a dynamically decreasing difference with respect to said held sampled value;

sample accept means for determining at each sample clock following a said sampling whether the difference between said analog signal and said held sample value is more than said threshold value, and for generating an accept signal when said difference is greater;

time means for determining the number of skipped samples between each said sampling and for generating a digital representation of said skipped samples; and

enabling means for enabling said sample means to sample said analog signal with a said clock signal only when a said accept signal is generated, whereby some samples are skipped and sampling is done at a non-linear rate with a reduction of samples depending upon time differences in said analog signal and the shape of decrease of said threshold signal.

2. The system as described in claim 1, comprising storage means for storing in relation to each other for each sample the digital representation of said analog signal and the digital representation of lapsed time.

3. The system as described in claim 1, wherein said threshold means comprises means for generating a signal which starts at a predetermined magnitude and decreases to substantially zero within a predetermined time corresponding to a predetermined number of sample clocks following the last sampling.

4. A compression system for adapting the effective sampling rate of an analog signal as a function of variations of said analog signal, comprising:

sampling means for sampling said analog signal at a predetermined sampling rate;

block means for holding a selected number of said samples;

delta means for determining a delta value for each said held sample, said delta value being representative of the difference between each said sample and the prior sample;

max means, for determining for each said block, the maximum delta value of said block;

threshold means for setting a plurality of thresholds corresponding to predetermined different delta values;

comparison means for comparing said maximum value with said thresholds and, on the basis of said comparing, keeping some of said delta values and discarding the others; and

writing means for writing compressed data made up of said maximum value and said kept delta values for each said block, wherein said threshold means comprises means for storing a plurality of respective discrete threshold values, and said comparison means comprises means for determining which two of said threshold values said maximum delta value is between, and for discarding delta values as a function of said determining.

5. The system as described in claim 4, wherein said writing means comprises telemetry means for direct telemetering of said compressed data.

6. The system as described in claim 4, wherein said writing means comprises storage means for storing said compressed data.

7. The system as described in claim 4, wherein said writing means has bit means for determining a number of bits for writing each kept delta value as a function of said determining.

8. The system as described in claim 4, wherein said system is housed in a pacemaker, and further comprising marker means for writing pacemaker marker data in said compressed data when a marker event has occurred during a said block.

9. An implantable pacemaker system, said system having pacing means for pacing a patient, sensing means for sensing physiologic analog signals, and data means for deriving from said analog signals compressed digital data representative of said analog signals, said data means comprising:

variation means for determining a measure of the variation of said analog signal;

non-linear means for obtaining sample data from said analog signal at a non-linear sample rate, said non-linear means comprising means for determining said non-linear sample rate as a function of said variation measure; and

writing means for writing data representative of said non-linear rate and said obtained samples, wherein said non-linear means comprises clock means for generating clock signals, sample means for sampling said analog signals coincident with clock signals, converter means for converting samples into digital data, and sample skip means for causing said sample means to skip sampling corresponding to Portions of said analog signals which contain reduced variations with respect to said variation measure, whereby the number of sampling operations is reduced to less than the number of said sample clocks further comprising means for obtaining data representative of said skipped samples, and skip compression means for compressing said skipped sample data.

10. The system as described in claim 9, wherein said skip compression means comprises delta means for obtaining skip delta data representative of the difference between each successive sample pair of skipped samples between the pair samples.

11. The system as described in claim 9, comprising lossless compression means for compressing said sample digital data and code means for coding said skip delta data.

12. A method of obtaining a compressed data representation of an analog signal, comprising:

establishing threshold data relating to anticipated values of said analog signal;

sampling said analog signal at an effective non-linear rate to obtain non-linear sample data, said non-linear sampling including comparing said analog signal to said threshold data and deriving said non-linear rate from said comparing;

obtaining a measure of said non-linear rate corresponding to each sample, and converting said measures into rate data;

losslessly compressing said sample data establishing discrete threshold data points, first sampling said analog signal at a fixed rate to obtain first sample data, comparing said first sample data with said threshold data points, and non-linear sampling said first sample data as a function of said comparison; and

writing said rate data and said sample data.

13. The method as described in claim 12, comprising establishing analog threshold data, and deriving said non-linear rate before sampling said analog signal.

14. The method as described in claim 12, comprising generating sample clocks at a predetermined clock rate, for each said clock determining whether to skip a sample or take a sample, and obtaining said non-linear rate measure by determining skipped samples.

15. The method as described in claim 14, comprising compressing said rate data by determining the differences of skipped samples between successive samples, and coding said differences.

16. An implantable medical device system comprising:

comprising a pacemaker, said pacemaker having marker means for obtaining marker channel data,

signal means for obtaining an analog physiological signal; and

data means for obtaining a compressed digital data stream representative of said analog signal, said data means having non-linear means for obtaining non-linearly spaced samples of said analog signal, first compression means for providing compressed data representative of said samples, and second compression means for providing compressed data representative of the spacing of said samples, thereby providing a data stream representative of said analog signal and insertion means for inserting said marker channel data into said data stream.

17. The system as described in claim 16, wherein said non-linear means comprises an analog non-linear sampling A/D converter which determines sampling rate before A/D conversion.

18. The system as described in claim 16, wherein said non-linear means comprises A/D sampling means for obtaining digital samples representative of said analog signal, and sample-skipping means for skipping some of said digital samples as a function of the rate of variation of said analog signal.

19. The system as describe in claim 16, wherein said non-linear means comprises threshold signal means for generating a threshold signal, said threshold signal being initialled upon each said sample.

20. The system as described in claim 19, comprising clock means for generating clock signals and comparator means operative at each said clock signal for comparing said threshold signal with said analog signal.

21. The system as described in claim 19, comprising means for adjusting said threshold signal as a function of signal history.

22. The system as described in claim 19, comprising program means for adjusting said threshold signal as a function of programmed data.

23. The system as described in claim 19, wherein said threshold signal has an initial value and a length, and adjusting means for adjusting either of said initial value and length.

24. The system as described in claim 23, wherein said adjusting means comprises accumulating means for accumulating data representative of said analog signal and for adjusting said threshold signal as a function of said accumulated data.