Automated identification of occlusion location in the cuprit coronary artery
A diagnostic ECG system analyzes lead traces for evidence of ST elevation in the lead signals. The pattern of ST elevation in leads having predetermined vantage points to the electrical activity of the heart and, in some instances, the presence of ST depression in certain other leads, identifies a specific coronary artery or branch as the culprit coronary artery for an acute ischemic event. ECG measurements which are associated with the identity of specific arterial occlusion locations are calculated and used to form a classifier of the probability of occlusion at different locations. The location identified as having the highest probability is indicated to a user as the most likely occlusion location.
1. A method of identifying the location of an occlusion of a culprit coronary artery associated with an ischemic event by means of an n-lead ECG system comprising:
receiving ECG signals of the n-leads;
analyzing the ECG lead signals for ST elevation;
calculating a plurality of ECG measurements which are useful in the classification of a culprit coronary artery and the location of its occlusion;
using the ECG measurements to develop a classifier of the probability of the location of an occlusion in different coronary artery locations;
operating the classifier to identify a location with the highest probability of an occlusion; and
identifying to a user a specific coronary artery location which is likely to be the site of an occlusion.
2. The method of claim 1 , wherein using the ECG measurements to develop a classifier further comprises producing contour plots of probability surfaces of the likelihood of occlusion in different coronary artery locations.
3. The method of claim 2 , wherein the different coronary artery locations are non-proximal RCA, proximal RCA and LCx.
4. The method of claim 2 , wherein operating the classifier further comprises:
accessing the contour plots with ECG measurements to identify a probability value at each of a plurality of locations; and
identifying a location from the contour plot with the highest probability value.
5. The method of claim 1 , wherein developing a classifier further comprises developing a logistic regression classifier.
6. The method of claim 1 , wherein the different coronary artery locations are proximal LAD and mid/distal LAD.
7. The method of claim 6 , wherein analyzing further comprises analyzing the ECG lead signals for characteristics of NSTEMI.
8. The method of claim 7 , wherein calculating further comprises calculating measurements of deeply inverted and symmetric T-waves.
9. The method of claim 1 , further comprising:
analyzing an ST elevation pattern for an indication of the obstruction of a specific coronary artery or branch; and
identifying to a user a specific coronary artery or branch as a culprit coronary artery.
10. The method of claim 9 , wherein analyzing the pattern of ST elevation further comprises identifying ST elevation in an anterior lead group including one or more of leads V2, V3, V4, V5 which may be accompanied by ST elevation in lateral lead group including one or more of leads V6, I, aVL as indicating the obstruction of the left anterior descending (LAD) coronary artery; and
wherein identifying further comprises identifying the LAD coronary artery as a culprit coronary artery.
11. The method of claim 9 , wherein analyzing the pattern of ST elevation further comprises identifying ST elevation in inferior lead group including one or more of leads II, III, aVF and in posterior lead group including one or more of leads V7, V8, V9; and
analyzing the traces of anterior lead group including one or more of leads V1, V2, V3 for ST depression and identifying ST depression in the lead group;
wherein identifying further comprises identifying the left circumflex (LCx) coronary artery as a culprit coronary artery.
12. The method of claim 9 , wherein analyzing the pattern of ST elevation further comprises identifying ST elevation in inferior lead group including one or more of leads II, III, aVF and/or in right chest leads including one or more of leads V3R-V5R, wherein the level of ST elevation in lead III is greater than the level of ST elevation in lead II; and
wherein identifying further comprises identifying the right coronary artery (RCA) as a culprit coronary artery.
13. The method of claim 9 , wherein analyzing the pattern of ST elevation further comprises identifying ST elevation in lead aVR; and
further comprising identifying ST depression in a plurality of other leads; and
wherein identifying further comprises identifying the left main (LM) coronary artery as a culprit coronary artery.
14. The method of claim 1 , wherein n is twelve.