Key result
An automated ECG beat classification system using Deep Auto-Encoders and Deep Neural Networks achieved 99.73% accuracy, 91.20% sensitivity, and 99.80% specificity for cardiac disease detection.
Why the study?
Prior machine learning techniques for ECG beat classification required manual feature selection, expert input, and time-consuming processing steps.
Does a Deep Learning architecture using DAEs and DNNs improve ECG beat classification compared to shallow models?
Population
10 classes of imbalanced data from ECG signals
Comparison
Proposed deep auto-encoder and deep neural network model vs shallow models and deep learning approaches
Authors
Loading...
Supports automated ECG classification; leaves open prospective clinical validation before practice change.
Does a Deep Learning architecture using DAEs and DNNs improve ECG beat classification compared to shallow models?
A novel deep learning architecture combining Deep Auto-Encoders and Deep Neural Networks demonstrates high accuracy (99.73%) for automated ECG beat classification.
Nurmaini et al. (2019) studied Cardiac Disease. Deep Learning architecture (Deep Auto-Encoders and Deep Neural Networks) vs. Shallow models and other DL approaches was evaluated on Classification accuracy. An automated ECG beat classification system using Deep Auto-Encoders and Deep Neural Networks achieved 99.73% accuracy, 91.20% sensitivity, and 99.80% specificity for cardiac disease detection.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: