Key result
A discrete wavelet transform coupled with 1D-CNNs achieved 99.17% accuracy, 98.90% sensitivity, and 99.17% specificity for classifying normal sinus rhythm, AF, and non-AF from short-term ECGs.
Why the study?
Manual interpretation of short-term single-lead ECGs is subjective and susceptible to inter-observer variabilities, and signals are often contaminated by noise.
Does a 1D-CNN coupled with DWT accurately detect atrial fibrillation from short-term single-lead ECGs?
Does a 1D-CNN coupled with DWT accurately detect atrial fibrillation from short-term single-lead ECGs?
A 1D-CNN model coupled with DWT demonstrated high accuracy (>99%) in detecting atrial fibrillation from short-term single-lead ECGs.
No takes yet. Share an insight, caveat, or question.
Supports feasibility of CNN-based AF detection from short ECGs; leaves open prospective validation before clinical use.
Nurmaini et al. (2020) studied Atrial fibrillation. Discrete wavelet transform (DWT) coupled with 1D-CNNs vs. Other approaches was evaluated on Three-class classification (NSR, AF, NAF) accuracy. A discrete wavelet transform coupled with 1D-CNNs achieved 99.17% accuracy, 98.90% sensitivity, and 99.17% specificity for classifying normal sinus rhythm, AF, and non-AF from short-term ECGs.
Synapse has enriched 4 closely related papers on similar clinical questions. Consider them for comparative context: