Introduction: Clinical outcomes of several acute cardiac conditions, including myocardial infarction, can be significantly improved by timely diagnostics based on electrocardiography (ECG). However, the current diagnostic technologies do not permit this as they either require a large number of wired ECG electrodes placed by trained personnel or provide incomplete cardiac assessment due to a small number of electrodes in self-assessment devices.
Method: A handheld ECG device developed by HeartBeam records three quasi-orthogonal leads, thus opening the door to the self- assess-ment with unprecedented information about the cardiac vector. Here, we hypothesize that the information provided is sufficient for accurate syn-thesis of the 12-lead ECG. To test the hypothesis, we propose a synthesis algorithm based on the segment-by-segment 4-matrix (4M) transfor-mation applied to the P wave, QRS complex, ST segment, and T wave. The segments' transformation matrices ware calculated on the median beats of the 3- and 12-lead ECG signals using the least-squares method.
Results: The accuracy of the method was tested using data obtained from 64 healthy volunteers. The 4M method reconstructed the standard ECG with a 0.96 mean crosscorrelation for all 12 leads, showing the su-periority over the EASI method with the mean correlations in the range 0.56-0.98. Moreover, the 4M method rendered at least 2 times smaller mean difference error in amplitude and cardiac axis angle on all 4 seg-ments. We further show that the signal segmentation helps avoid false T-wave inversion, which may cause false alarms for acute MI. The T waves reconstructed by 4M approach had 4 times higher Pearson coefficients than those reconstructed by a single matrix.
Conclusion: We prove the capability of the mobile 3-lead 4M ECG technology to accurately synthesize the 12-lead ECG from a single meas-urement, which makes it potentially suitable for self-assessment in acute conditions.