A Rough Set Based Approach for ECG Classification
Transactions on Rough Sets IX
PVC discrimination using the QRS power spectrum and self-organizing maps
Computer Methods and Programs in Biomedicine
| Hi-index | 0.00 |
In this paper, ECG arrhythmia classification using principalcomponent analysis is proposed. Hebbian neural networks are usedfor computing the principal components of an ECG signal. Thisprovides an unsupervised feature extraction, dimension reductionand an improved computing efficiency. Results from 14 pathologicalrecords obtained from the MIT ECG database demonstrate thecapability of this method in differentiating between five differenttypes of arrhythmia despite the variations in signal morphology. Anaverage value for classification sensitivity and positivepredictivity were found to be Se% = 98.1% and +P% = 94.7%respectively.