Medical diagnostic accuracies can be improved when the pattern is simplified through representation by important features. Features are used to represent patterns with minimal loss of important information. The feature vector, which is comprised of the set of all features used for describing a pattern, is a reduced-dimensional representation of that pattern. By identifying a set of salient features, the noise in a classification model can be reduced, resulting in more accurate classification. In this study, a signal-to-noise ratio (SNR) saliency measure was employed to determine saliency of input features of recurrent neural networks (RNNs) used in the classification of electrocardiogram (ECG) signals. In order to extract features representing four types of ECG beats (normal beat, congestive heart failure beat, ventricular tachyarrhythmia beat, atrial fibrillation beat) obtained from the Physiobank database, eigenvector methods were used. The RNNs used in the ECG beats classification were trained for the SNR screening method. The results of the application of the SNR screening method to the ECG signals demonstrated that classification accuracies of the RNNs with salient input features are higher than those of the RNNs with salient and non-salient input features.