Analysis of the ST-T complex of the electrocardiogram using the Karhunen—Loeve transform: adaptive monitoring and alternans detection

Abstract

The Karhunen-Loève transform (KLT) is applied to study the ventricular repolarisation period as reflected in the ST-T complex of the surface ECG. The KLT coefficients provide a sensitive means of quantitating ST-T shapes. A training set of ST-T complexes is used to derive a set of KLT basis vectors that permits representation of 90% of the signal energy using four KLT coefficients. As a truncated KLT expansion tends to favor representation of the signal over any additive noise, a time series of KLT coefficients obtained from successive ST-T complexes is better suited for representation of both medium-term variations (such as ischemic changes) and short-term variations (such as ST-T alternans) than discrete parameters such as the ST level or other local indices. For analysis of ischemic changes, an adaptive filter is described that can be used to estimate the KLT coefficient, yielding an increase in the signal-to-noise ratio of 10 dB (u = 0.1), with a convergence time of about three beats. A beat spectrum of the unfiltered KLT coefficient series is used for detection of ST-T alterans. These methods are illustrated with examples from the European ST-T Database. About 20% of records revealed quasi-periodic salvos of ischemic ST-T change episodes and another 20% exhibit repetitive, but not clearly periodic patterns of ST-T change episodes. About 5% of ischemic episodes were associated with ST-T alterans.

Automatic Detection Of Electrocardiogram ST Segment: Application In Ischemic Disease Diagnosis

“…The RR-Interval is the distance between two subsequent QRS complex and represent the Heart Rate (HR) variability. In our system, we used a robust real-time QRS detection algorithm popularly known as Pan-Tompkins algorithm [14] and added searchback compare method [7].…”

“…The several methods for ischemia parameter detection (T wave and ST complex) have been proposed. In generally, all of them are based on the spectral estimation [6] and signal point from the ST segment better characterizes ischemic patterns [3,7]. The various methods have been applied to the ECG for ischemia analysis and detection: used the First Fourier Transform (FFT) to analyze the frequency component [8], fuzzy-logic, neural network, genetic algorithm, support vector machines (SVM), wavelet transform and many more [9].…”

Computerized Interpretation of Cardiovascular Physiological Signals

“…In order to disregard the effects of insignificant factors, such as pulse rate and age, they proposed to extract and normalize the QRS complex for pattern recognition. This idea, namely morphological analysis instead of metric measurement, was revisited more than once in the following years (Bemmel et al, 1973;Maitra & Zucker 1975;Suppappola et al, 1997;Laguna et al, 1999). In those systems, the ECG waveforms or their components (e.g., QRS complex, ST-T segmentation, etc) were regularized and normalized beat by beat for morphological analysis.…”

ECG-based detection of body position changes in ischemia monitoring

“…In this paper, the KLT has been applied to the QRS complex and the entire ST-T complex in order to analyze how BPCs affect the ECG signal. The details on how this transform was developed and applied to the ECG segments are described in [14], [15].…”

“…These functions are simply the distance series between each KLT coefficients vector (in which only the first four components are considered) and a mean reference value ( ) estimated from the onset of the recording (6) with being the th-order KLT coefficient at estimated for th lead beat. The coefficient series is estimated using adaptive filtering to remove noise uncorrelated to the signal, thus improving the KLT estimation [14]. A compromise between noise reduction and convergence time is reached using a step-size parameter for the LMS algorithm of , that yields a SNR improvement in the series of more than 6 dB, with a convergence time of one beat [14].…”