J. P. V. Madeiro

New approach for T-wave peak detection and T-wave end location in 12-lead paced ECG signals based on a mathematical model

This paper presents an innovative approach for T-wave peak detection and subsequent T-wave end location in 12-lead paced ECG signals based on a mathematical model of a skewed Gaussian function. Following the stage of QRS segmentation, we establish search windows using a number of the earliest intervals between each QRS offset and subsequent QRS onset. Then, we compute a template based on a Gaussian-function, modified by a mathematical procedure to insert asymmetry, which models the T-wave. Cross-correlation and an approach based on the computation of Trapeziums area are used to locate, respectively, the peak and end point of each T-wave throughout the whole raw ECG signal. For evaluating purposes, we used a database of high resolution 12-lead paced ECG signals, recorded from patients with ischaemic cardiomyopathy (ICM) in the University Hospitals of Leicester NHS Trust, UK, and the well-known QT database. The average T-wave detection rates, sensitivity and positive predictivity, were both equal to 99.12%, for the first database, and, respectively, equal to 99.32% and 99.47%, for QT database. The average time errors computed for T-wave peak and T-wave end locations were, respectively, -0.38±7.12 ms and -3.70±15.46 ms, for the first database, and 1.40±8.99 ms and 2.83±15.27 ms, for QT database. The results demonstrate the accuracy, consistency and robustness of the proposed method for a wide variety of T-wave morphologies studied.

Evaluating Gaussian and Rayleigh-Based Mathematical Models for T and P-waves in ECG

This paper presents a comparative study of modelling and segmentation of P and T waves in electrocardiograms, using three different mathematical models: Gaussian function, a composition of two Gaussian functions and Rayleigh probability density function (Rayleigh pdf). In order to evaluate the adaptability and the matching degree between each model and each characteristic wave, we compute the normalized root mean square (RMS) error, as well as the evolution of the fitting parameters related to each kernel throughout ECG records from the well-known QT database. Concerning the delineation of P and T-waves, we apply Wavelet Transform for estimating T-wave and P-wave peak locations and combine each developed model with an approach based on the computation of Trapeziums area to locate the end point of each T-wave and the beginning and end point of each P-wave. The composition of two Gaussian functions has produced the most accurate results concerning wave modelling, providing average values of normalized RMS errors equal to 9,15% and 18,70%, respectively for T-wave and P-wave. Rayleigh pdf provided the most stable fitting parameters. For T-wave end location, the most accurate results were computed when using the kernel composition of two Gaussian functions, for which the average time error was 4,49 ± 14,32 ms. For P-wave begin and P-wave end locations, the most accurate results were computed when using kernel Rayleigh pdf, for which the average time errors were, respectively, -4,23 ± 14,84 ms and 2,26 ± 13,14 ms.