A New Algorithm of QT Interval Measurement Based on Multiscale Morphological Derivative Transform

Abstract

A new algorithm of QT interval measurement based on multiscale morphological derivative transform (MMDT) was described in this paper. After detection of QRS complex based on MMDT, we classified the waveform into four types based on its morphological characteristics and the optimum strategy for each type was described to detect the onset of Q wave. By introducing the 'wing' function, two referenced points of T peak were found, which was helpful to locate the T peak point and improve the performance of MMDT operator in detecting the bifid T wave and biphasic T wave. 80 data records from CSE database were used to evaluate the availability. The standard deviation of the onset calibration of QRS complex was limited in 6 ms while the standard deviation of the offset calibration of T wave was limited in 10 ms which were within the limits required by the CSE committee. Compared with those algorithms based on wavelet transform and adaptive threshold techniques, our algorithm needed less empirical parameters and calculation which exhibited good potentials for automated ECG signal analysis and cardiovascular arrhythmia recognition.