Synthesis and Evaluation of Acute Myocardial Ischemia Using a Dynamical ECG Model

RINCÓN SOLER, AI; VALVERDE, ER; FERNÁNDEZ BISCAY, C; BONOMINI M.P.; ARINI P.D.

Córdoba

Congreso; XXI Congreso Argentino de Bioingeniería, X Jornadas de Ingeniería Clínica; 2017

Dynamical modelscapable to mimic real physiological situations are useful tools to generate setsof signals with specific characteristics in time and frequency. An applicationof this synthetic signals involves training and improvement of algorithms todetect different pathophysiological conditions. To obtain the best results, synthesizedsignals need to be highly comparable with the real ones. In this studyimplements a procedure to generate synthetic electrocardiogram (ECG) signalsfrom patients referred to percutaneous transluminal coronary angioplasty (PTCA)and their respective control registers using the ECGSYN model. In this order ofideas, the T-wave spectral variance (TSV) index was used as a method to comparethe information that synthetic signals contain in relation to their realcounterparts, and in this way, test the reliability of the ECGSYN. Results indicatehigh similitude between the real T-waves and those produced by the model with amaximum error of 11% in the worst case. TSV index in real and synthetic signalsreveal similar values in control and PTCA. Additionally, no statisticaldifferences (p = NS) were found between these values, indicating that theECGSYN is capable to replicate a physiological scenario with acute myocardialischemia.