A. Rakotomamonjy et al., AUTOMATED NEURAL-NETWORK DETECTION OF WAVELET PREPROCESSED ELECTROCARDIOGRAM LATE POTENTIALS, Medical & biological engineering & computing, 36(3), 1998, pp. 346-350
The aim of the study is 50 investigate the potential of a feedforward
neural network for detecting wavelet preprocessed late potentials. The
terminal parts of a simulated QRS complex are processed with a contin
uous wavelet transform, which leads to a time-frequency represenation
of the QRS complex. Then, diagnostic feature vectors are obtained by s
ubdividing the representations into several regions and by processing
the sum of the decomposition coefficients belonging to each region. Th
e neural network is trained with these feature vectors. Simulated ECGs
with varying signal-to-noise ratios are used to train and test the cl
assifier. Results show that correct classification ranges from 79% (hi
gh-level noise) to 99% (no noise). The study shows the potential of ne
ural networks for the classification of late potentials that have been
preprocessed by a wavelet transform. However, clinical use of this me
thod still requires further investigation.