Method for discriminating between ventricular and supraventricular arrhythmias

1. A method of observing and analyzing cardiac function, the method comprising:

capturing electrical cardiac signals using at least first and second electrodes implanted in a patient;

selecting portions of the cardiac signals for analysis including a set of samples each indicative of a cardiac event;

generating a number of scores related to the correlation of individual samples to a cardiac template;

calculating a set variance metric related to a variance in the correlation scores of the set of samples;

comparing the set variance metric to a threshold; and

if the set variance metric does not exceed the threshold, determining that the patient is likely experiencing a malignant cardiac event.

2. A method as in claim 1 , further comprising:

identifying a score related to the correlation of a most recent sample to the cardiac template;

comparing the identified score to a threshold; and

if the identified score does not exceed the threshold, determining that the patient is likely experiencing a malignant cardiac event.

3. A method as in claim 1 , further comprising observing a cardiac triggering event, the cardiac triggering event being the presence of a cardiac event rate in excess of a predetermined level, wherein the cardiac template is a cardiac template representative of non-malignant cardiac events occurring after a cardiac triggering event.

4. The method of claim 3 , wherein the cardiac template is dynamically updated.

5. The method of claim 1 , further comprising comparing an individual score threshold to one of the generated number of scores to determine whether an associated individual sample is indicative of a malignant cardiac condition.

6. A method of cardiac signal analysis comprising:

repeating the method of claim 1 for a number, Y, of temporally related sample sets; and

determining whether electrical cardiac stimulus is indicated by counting a number of sample sets, X, in which it is determined that the patient is likely experiencing a malignant cardiac event, and comparing X to Y.

7. A method of cardiac signal analysis comprising:

repeating the method of claim 2 for Y temporally related sample sets; and

determining whether electrical cardiac stimulus is indicated by counting a number, X, of sample sets in which it is determined that the patient is likely experiencing a malignant cardiac event.

8. A method of cardiac signal analysis comprising:

collecting a set of electrical signal samples of a patient's cardiac function from implanted electrodes;

comparing a number of individual signal samples to a cardiac signal template to generate a comparison score related to a correlation between the signal sample and the cardiac signal template;

for a number of individual signal samples, marking the individual signal sample as indicative of a malignant cardiac condition if at least one of the following conditions is met:

(a) a subset of signal samples including the individual signal sample and a number of prior signal samples illustrates a variance of comparison scores in excess of a predetermined threshold; and

(b) a cardiac rate triggering event has been detected, the cardiac rate triggering event comprising the detection of a cardiac rate in excess of a predetermined level, and a correlation between the individual signal sample and a post-event template falls below a predetermined level; and

observing whether at least a threshold number of signal samples are marked as indicative of a malignant cardiac condition within a predetermined number of signal samples.

9. An implantable cardiac stimulus device comprising:

operational circuitry contained within an implantable housing;

a plurality of electrodes coupled to the operational circuitry such that the operational circuitry can captured cardiac electrical data therefrom when implanted in an implantee; wherein the operational circuitry is configured to perform the following method of determining whether the implantee is in need of therapy:

a) capturing electrical cardiac signals from the plurality of electrodes;

b) selecting portions of the cardiac signals for analysis including a set of samples each indicative of a cardiac event;

c) generating a number of scores related to the correlation of individual samples to a cardiac template;

d) calculating a set variance metric related to a variance in the correlation scores of the set of samples;

e) comparing the set variance metric to a threshold; and

f) if the set variance metric does not exceed the threshold, determining that the patient is likely experiencing a malignant cardiac event.

10. The implantable cardiac stimulus device of claim 9 wherein the operational circuitry is further configured to:

repeat steps a) through f) for a number, Y, of temporally related sample sets;

determine whether electrical cardiac stimulus is indicated by counting a number of sample sets, X, in which it is determined that the patient is likely experiencing a malignant cardiac event, and comparing X to Y.

11. The implantable cardiac stimulus device of claim 9 wherein the operational circuitry is configured to:

identify a score related to the correlation of a most recent sample to the cardiac template;

compare the identified score to a threshold; and

if the identified score does not exceed the threshold, determining that the patient is likely experiencing a malignant cardiac event.

12. The implantable cardiac stimulus device of claim 9 wherein the operational circuitry is further configured to:

repeat the method of claim 11 for Y temporally related sample sets; and

determine whether electrical cardiac stimulus is indicated by counting a number, X, of sample sets in which it is determined that the patient is likely experiencing a malignant cardiac event.

13. The implantable cardiac stimulus device of claim 9 wherein the operational circuitry is further configured to observe a cardiac triggering event, the cardiac triggering event being the presence of a cardiac event rate in excess of a predetermined level, wherein the cardiac template is a cardiac template representative of non-malignant cardiac events occurring after a cardiac triggering event.

14. The implantable cardiac stimulus device of claim 13 wherein the operational circuitry is further configured such that the cardiac template is dynamically updated.

15. The implantable cardiac stimulus device of claim 9 wherein the operational circuitry is further configured to: compare an individual score threshold to one of the generated number of scores to determine whether an associated individual sample is indicative of a malignant cardiac condition.

16. An implantable cardiac stimulus device comprising:

operational circuitry contained within an implantable housing;

a plurality of electrodes coupled to the operational circuitry such that the operational circuitry can captured cardiac electrical data therefrom when implanted in an implantee;

wherein the operational circuitry is configured to perform the following method of determining whether the implantee is in need of therapy:

collect a set of electrical signal samples of a patient's cardiac function from the plurality of electrodes;

compare a number of individual signal samples to a cardiac signal template to generate a comparison score related to a correlation between the signal sample and the cardiac signal template;

for a number of individual signal samples, marking the individual signal sample as indicative of a malignant cardiac condition if at least one of the following conditions is met:

(a) a subset of signal samples including the individual signal sample and a number of prior signal samples illustrates a variance of comparison scores in excess of a predetermined threshold; and

(b) a cardiac rate triggering event has been detected, the cardiac rate triggering event comprising the detection of a cardiac rate in excess of a predetermined level, and a correlation between the individual signal sample and a post-event template falls below a predetermined level; and

observe whether at least a threshold number of signal samples are marked as indicative of a malignant cardiac condition within a predetermined number of signal samples.