Key result
1D Self-Operational Neural Networks achieved 98% and 99.04% average accuracies for supra-ventricular and ventricular ectopic beat classifications on the MIT-BIH dataset, surpassing 1D CNNs.
Why the study?
Compact deep learning systems with real-time ability and high accuracy for patient-specific ECG classification remain scarce, and conventional 1D CNNs are limited by their basic linear neuron models.
Population
MIT-BIH arrhythmia benchmark database
Comparison
1D Self-organized Operational Neural Networks vs conventional 1D CNNs
Loading...
May enable compact real-time patient-specific ECG classification; leaves open prospective clinical validation before practice change.
1D Self-ONNs provide highly accurate, real-time patient-specific ECG classification for arrhythmia detection, outperforming conventional 1D CNNs.
A 2021 study studied Arrhythmia. 1D Self-Operational Neural Networks (1D Self-ONNs) vs. 1D Convolutional Neural Networks (1D CNNs) was evaluated on Classification accuracy of supra-ventricular and ventricular ectopic beats. 1D Self-Operational Neural Networks achieved 98% and 99.04% average accuracies for supra-ventricular and ventricular ectopic beat classifications on the MIT-BIH dataset, surpassing 1D CNNs.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: