Quality Assessment of Ambulatory Electrocardiogram Signals by Noise Detection using Optimal Binary Classification

In order to improve the diagnostic capability in Ambulatory Electrocardiogram signal and to reduce the noise signal impacts, there is a need for more robust models in place. In terms of improvising to the existing solutions, this article explores a novel binary classifier that learns from the features optimized by fusion of diversity assessment measures, which performs Quality Assessment of Ambulatory Electrocardiogram Signals (QAAES) by Noise Detection. The performance of the proposed model QAAES has been scaled by comparing it with contemporary models. Concerning performance analysis, the 10-fold cross-validation has been carried on a benchmark dataset. The results obtained from experiments carried on proposed and other contemporary models for cross-validation metrics have been compared to signify the sensitivity, specificity, and noise detection accuracy.