Using a Cell-to-ECG model to evaluate ischemia detection from different lead sets

In this study, we designed a Cell-to-ECG forward model to simulate ischemia condition from cell level, and then defined five different ischemia locations including inferior, anterior, septal, lateral, and apex on an actual 3-D heart geometry from CT images. For simulating the early ischemia, the ion channels are adjusted through concentration and conductance. The extent of ischemia is defined with a ten-layer model from endocardial to epicardial. The model can generate ECG signal from any position on the torso, therefore different lead sets can be configured based on the selected subset from the torso. All reduced lead sets are converted to 12-lead ECGs by published conversion matrixes. The criteria of acute ischemia detection are based on AHA/ACC recommendation of 2 contiguous leads 200 uV ST elevation for chest leads, and 100 uV ST elevation for limb leads. The model reveals the consistent ECG pattern with the accepted criteria, especially the reciprocal ST changes from the simulated transmural ischemia cases. The reduced lead sets show comparable detection accuracy and sensitivity to the standard 12-lead for most ischemia locations except for anterior MI, where more specific criteria can adapt to the changes of the ECG morphology from reduced lead sets.