Method for efficiently encoding and compressing ECG data optimized for use in an ambulatory ECG monitor
A method for efficiently encoding and compressing ECG data optimized for use in an ambulatory electrocardiography monitor is provided. ECG data is first encoded and compressed in a lossy process and further encoded and compressed in a lossless process. A compression ratio significantly higher than other Holter-type monitors is achieved. Requirements for storage space and power cell consumption are reduced, contributing to the long-term availability of the monitor.
1. A computer-implemented method for encoding and compressing electrocardiography values, comprising the steps of:
obtaining a series of electrocardiography values, wherein each of the electrocardiography values can be one of a raw electrocardiography value, processed electrocardiography value, filtered electrocardiography value, averaged electrocardiography value, and a sampled electrocardiography value;
defining a plurality of bins, each bin comprising a lower threshold electrocardiography value, an upper threshold electrocardiography value, and a code for the bin;
setting a serial accumulator to a pre-determined value;
processing each of the electrocardiography values from the series of the electrocardiography values, comprising the steps of:
selecting the electrocardiography value next remaining to be processed;
taking a difference of the selected electrocardiography value and the serial accumulator;
identifying the bin in the plurality of the bins corresponding to the difference;
representing the selected electrocardiography value by the code for the identified bin;
adjusting the serial accumulator by a value derived from the identified bin; and
writing each of the represented codes into a sequence in a non-volatile memory.
2. A method in accordance with claim 1 , further comprising the step of encoding the sequence into a single number between 0 and 1.
3. A method in accordance with claim 2 , further comprising encoding the sequence by the steps of:
setting a range of 0 to 1 for an initial code from the sequence of the represented codes; and
processing each of the codes remaining in the sequence, comprising the steps of:
obtaining an estimation of probabilities of next codes;
dividing the range into sub-ranges, each sub-range representing a fraction of the range proportional to the probabilities of the next codes;
obtaining a next code;
selecting the sub-range corresponding to the next code;
representing the next code by the selected sub-range; and
continuing the steps of the process using the selected sub-range in place of the range.
4. A method in accordance with claim 3 , wherein the probabilities of the next codes are obtained through a sample electrocardiography data or the sequence of the represented codes.
5. A method in accordance with claim 3 , further comprising the step of adjusting the probabilities during at least one of recursive range divisions.
6. A method in accordance with claim 1 , wherein the value derived from the identified bin may be based on one of the lower threshold electrocardiography value, the upper threshold electrocardiography value, a modification of the lower electrocardiography value with a constant number or a variable number, and a modification of the upper threshold electrocardiography value with a constant number or variable number.
7. A method in accordance with claim 1 , wherein the predetermined value is a center value of an electrocardiography recorder.
8. A method in accordance with claim 1 , wherein each successive bin is continuing forward from a previous bin so as to cover a contiguous range of the electrocardiography values.
9. A method in accordance with claim 1 , wherein the bin is identified when the difference is equal to or larger than the lower threshold electrocardiography value and smaller than the upper threshold electrocardiography value of the identified bin.
10. A method in accordance with claim 1 , wherein the bin is identified when the difference is larger than the lower threshold electrocardiography value and equal to or smaller than the upper threshold electrocardiography value of the identified bin.
11. A computer-implemented method for encoding and compressing electrocardiography values, comprising the steps of:
obtaining a series of electrocardiography values;
defining a plurality of bins, each bin comprising a lower threshold electrocardiography value, an upper threshold electrocardiography value, and a code for the bin;
setting a serial accumulator to predetermined value;
processing each of the electrocardiography values from the series of the electrocardiography values, comprising the steps of:
selecting the electrocardiography value next remaining to be processed;
taking a difference of the selected electrocardiography value and the serial accumulator;
identifying the bin in the plurality of the bins corresponding to the difference;
representing the selected electrocardiography value by the code for the identified bin;
adjusting the serial accumulator by a value derived from the identified bin;
concatenating each of the represented codes into a sequence;
encoding the sequence of the represented codes into a single number between 0 and 1, further comprising the step of:
setting a range for an initial code from the sequence of the represented codes; and
processing each of the codes remaining in the sequence, comprising the steps of:
obtaining an estimation of probabilities of next codes;
dividing the range into sub-ranges, each sub-range representing a fraction of the range proportional to the probabilities of the next codes;
obtaining a next code;
selecting the sub-range corresponding to the next code;
representing the next code by the selected sub-range; and
continuing the steps of the process using the selected sub-range in place of the range; and
storing the encoded single number into the non-volatile memory.
12. A method in accordance with claim 11 , wherein the probabilities of the next codes are obtained through a sample electrocardiography data or the sequence of the represented codes.
13. A method in accordance with claim 11 , further comprising the step of adjusting the probabilities during at least one of recursive range divisions.
14. A method in accordance with claim 11 , wherein the value derived from the identified bin may be based on one of the lower threshold electrocardiography value, the upper threshold electrocardiography value, a modification of the lower electrocardiography value with a constant number or a variable number, and a modification of the upper threshold electrocardiography value with a constant number or variable number.
15. A method in accordance with claim 11 , wherein the predetermined value is a center value of an electrocardiography recorder.
16. A method in accordance with claim 11 , wherein each successive bin is continuing forward from a previous bin so as to cover a contiguous range of the electrocardiography values.
17. A method in accordance with claim 11 , wherein the bin is identified when the difference is equal to or larger than the lower threshold electrocardiography value and smaller than the upper threshold electrocardiography value of the identified bin.
18. A method in accordance with claim 11 , wherein the bin is identified when the difference is larger than the lower threshold electrocardiography value and equal to or smaller than the upper threshold electrocardiography value of the identified bin.
19. A method in accordance with claim 11 , wherein the electrocardiography value can be one of raw electrocardiography value, processed electrocardiography value, filtered electrocardiography value, averaged value, and sampled value.