We assessed characteristics of atrial and ventricular activity from the ECG for predicting the oflset of atrial fibrillation for the 2004 PhysioNet/Computers in Cardiology Challenge. Seven parameters were analysed with five based on the statistical characteristics of the RR interval (mean, standard deviation, skewness, kurtosis and median beat-to-beat change), and fibrillation frequency and atrial signal amplitude. The power of the parameters to predict termination of the arrhythmia was assessed individually and in combination using linear discriminant analysis ( D A ) and a?t$cial neural network (ANN) techniques. Fibrillation fiequency with a threshold value of 5.55 Hz was able to idenrify 10/10 learning set records which terminated immediately (T) and 8/10 of non-terminating records (N) and was the best of the individual parameters. Classifcations for the test set for event 1 of the challenge for algorithms based on fibrillation frequency alone, LDA and ANN received scores of 24/30, 18/30 and 23/30 respectively. Low jibrillation frequency is an indicator of spontaneous termination of atrial fibrillation.
|Name||Computers in Cardiology|
|Conference||Computers in Cardiology Conference |
|Period||19/09/04 → 22/09/04|