A Semi-Automated QT Interval Measurement Based on Wavelet and Energy Analysis

In this paper, a semi-automated multiscale algorithm based on rescaled continuous wavelet transform is presented in order to determine QT interval. According to our previous works, the relation between the duration of ECG waves and their wavelet transforms and dominant scales are used to determine QRS complex vicinity and to denoise ECG signal based on detected vicinity of QRS complex. Then, a simple mathematical sinusoid model for T-wave is considered to determine T-wave domain which is based on variance deviation of T-model and ECG between two successive QRS complexes. A simple analysis of signal’s energy leads to rescaled Maximum Energy Density (MED) curve which is used to determine onset and offset of QT-interval. We evaluated the algorithm on the PTB database. The proposed approach is achieved about 53.7 ms of RMS error. The preliminary results are sent to PhysioNet/Computers in Cardiology Challenge 2006.