Classification of the long QT syndrome based on discriminant analysis of T-wave morphology

The long QT syndrome (LQTS) is a genetic disorder, characterized by a long QT interval in the ECG, due to abnormal cardiac repolarization. Apart from the QT interval, various parameters, describing T-wave morphology, have been shown to have diagnostic value, but no single ECG parameter has been sufficient. In this study we present a method for discrimination among normal genotypes and mutations in the KCNQ1 (KvLQT1) and KCNH2 (HERG) genes using parameters describing T-wave morphology in terms of duration, asymmetry, flatness and amplitude. Discriminant analyses based on 3, 4, 5, or 6 parameters all resulted in perfect discrimination in a learning set of 22 subjects with respectively 3, 3, 2, and 3 misclassifications in an evaluation set of 17 subjects. In a combined learning set of all 39 subjects both the 5 and 6 parameter classifiers resulted in a single misclassification