Abstract
Hypothesis/objective: Prolonged QT interval is an index of propensity for dangerous ventricular tachyarrhythmias. The aim of this article is to establish an automatic algorithm for QT interval measurement.
Method: The proposed method is based on the continuous wavelet transform. In this method, the concepts of the rescaled wavelet coefficients and dominant scales of the electrocardiogram (ECG) components are used to perform detection of ECG characteristic points. A new concept of rescaled maximum energy density is introduced so as to perform the estimation of the QT interval.
Results and conclusion: We have applied the algorithm to the PTB database of the Physiobank∖Physionet in lead II. Then, the results were evaluated using pertinent reference QT. The criterion used for evaluation of the method's performance is the root mean square (RMS) error. The method approached the RMS error of 27.89 ms for 549 subjects. The proposed method is fast, simple and is applicable to a wide range of ECG cardio cycle morphologies.
Acknowledgement
The authors would like to express their deep gratitude to Prof. George Moody from Harvard-MIT division of Health Sciences and Technology (HST) because of his consideration on this project.