Table 3 Comparison between the most recent studies presented to predict PAF termination
Study | Database | Short description of methods | Diagnostic |
|---|---|---|---|
accuracy | |||
This work | Cinc/Challenge 2004 [19] | CTM from the first differences scatter plot of the wavelet coefficient vector associated to the AF frequency scale of the AA | 96% |
Alcaraz & Rieta 2009 [34] | Cinc/Challenge 2004 [19] | Regularity analysis via SampEn from the MAW of the AA signal | 93% |
Sun & Wang 2008 [39] | Cinc/Challenge 2004 [19] | Combination of features extracted from the ECG recurrence plot quantification making use of a multilayer perceptron neural network | 96% |
Alcaraz & Rieta 2008 [29] | Cinc/Challenge 2004 [19] | Regularity analysis via SampEn of time and wavelet domains of the AA | 93% |
Alcaraz et al 2008 [40] | Own Database with 50 episodes: 21 non-terminating and 29 terminating | Analysis of time and frequency parameters obtained from the AA | 92% |
Nilsson et al 2006 [41] | Cinc/Challenge 2004 [19] | Analysis of time and frequency parameters and non-linear indices obtained from the AA | 90% |
Petrutiu et al 2004 [42] | Cinc/Challenge 2004 [19] | Experimental combination of AA peak power evolution within the two last seconds of the episode with the DAF | 93% |