Figure 2.
The architecture of our artificial neural network (ANN). The input features to the ANN are as follows: mean NN: mean normal R-peak to normal R-peak interval; SDNN: standard deviation of NN; RMSSD: square root of the mean squared difference of successive NN; pNN50: proportion of the interval differences of successive NN intervals greater than 50 ms; VLF: very low frequency; LF: low frequency; HF: high frequency; SD1: standard deviation of the points perpendicular to the axis of the line of identity; SD2: standard deviation of the points along the axis of identity.