An Algorithm for Robust and Efficient Location of T-Wave Ends in Electrocardiograms

Abstract

The purpose of this paper is to propose a new algorithm for T-wave end location in electrocardiograms, mainly through the computation of an indicator related to the area covered by the T-wave curve. Based on simple assumptions, essentially on the concavity of the T-wave form, it is formally proved that the maximum of the computed indicator inside each cardiac cycle coincides with the T-wave end. Moreover, the algorithm is robust to acquisition noise, to wave form morphological variations and to baseline wander. It is also computationally very simple: the main computation can be implemented as a simple finite impulse response filter. When evaluated with the PhysioNet QT database in terms of the mean and the standard deviation of the T-wave end location errors, the proposed algorithm outperforms the other algorithms evaluated with the same database, according to the most recent available publications up to our knowledge.

Analysis of QRS-T subtraction in unipolar atrial fibrillation electrograms

“…Firstly, the R peak, QRS onset and T wave endT-wave end were detected on the ECG (lead I for 6 patients and V5 for 2 patients) using the methods described by Madeiro et al [10] (for QRS segmentation) and Qinghua et al [11] (for detection of T wave endT-wave end). After segmentation of all QRS-T complexes in the ECG, the corresponding time locations of the fiducial points were used for the AEG signals ( Figure 2).…”

“…The detection of T-wave end was also based on an indicator of the area of the T-wave [11] and no additional pre-processing was necessary. Letting Q f and Q i be the samples (indices) related respectively to each QRS offset and the subsequent QRS onset, we define a moving sample k, Q f +w ≤ k ≤ Q i -p-1, where p is the number of samples corresponding to 16 ms and w is the number of samples corresponding to 128 ms [11].…”

“…Letting Q f and Q i be the samples (indices) related respectively to each QRS offset and the subsequent QRS onset, we define a moving sample k, Q f +w ≤ k ≤ Q i -p-1, where p is the number of samples corresponding to 16 ms and w is the number of samples corresponding to 128 ms [11]. Then, for each k value, the following metrics were calculated , ] …”

“…Otherwise, the T-wave end is located at the sample with the maximum amplitude of |A(k)| (monophasic T-waves) [11].…”

“…The method does not suffer from the limitations of the classical ABS approach nor of the 'QRS cancellation' method, as it is based on a segmentation of each individual QRS complex [10] and T-wave end on the ECG [11]. So, the QRS-T templates are matched both in shape and in length to the individual signals (both onset and offset of the QRS-T complex).…”