Abstract
We present a novel method for prediction of the onset of a spontaneous (paroxysmal) atrial fibrilation episode by representing the electrocardiograph (ECG) output as two time series corresponding to the interbeat intervals and the lengths of the atrial component of the ECG. We will then show how different entropy measures can be calulated from both of these series and then combined in a neural network trained using the Bayesian evidence procedure to form and effective predictive classifier.
| Original language | English |
|---|---|
| Title of host publication | Proceedings of the 2nd International Conference on Computational Intelligence in Medicine and Healthcare (CIMED2005) |
| Publisher | BIOPATTERN Network of Excellence |
| Pages | 376-382 |
| Number of pages | 7 |
| Publication status | Published - 2005 |
Bibliographical note
Second International Conference on Computational Intelligence in Medicine and Healthcare (CIMED2005), Lisbon (PT), 29 June - 1 July 2005.Keywords
- prediction, spontaneous, paroxysmal, atrial fibrilation, electrocardiograph, interbeat intervals, atrial component