Performance analysis of the automatic RTapex and RTend measurement

Reliable definitions of T wave apex and end are necessary when the ventricular repolarisation beat-to-beat variability or its inter-lead dispersion is evaluated. Synthesised ECG signals were corrupted with additive white noise (WN), sinusoidal base line drift (BLD) and sinusoidal amplitude modulation (AM) in order to test the sensitivity of both RTapex and RTend to noise. RTapex was found to be more robust with respect to additive WN while RTend displayed a better behaviour with AM. Both measurements were reliable with BLD, if a suitable base-fine removal algorithm was applied. Real data from 13 healthy subjects were analysed by comparing the results of RTapex and RTend variability in the X, Y, Z orthogonal leads. Both variabilities displayed LF (/spl sim/0.1 Hz) and HF (/spl sim/0.3 Hz) rhythms synchronous to those of RR variability. RTend displayed a variability amplitude two fold that of RTapex and a better correspondence among the results obtained from different leads.