Key result
A semi-supervised stacked label consistent autoencoder yielded better reconstruction and classification of ECG and EEG signals and was more than an order of magnitude faster than compressed sensing.
A novel autoencoder-based framework offers a faster and more accurate alternative to compressed sensing for the simultaneous reconstruction and classification of biomedical signals like ECGs.
No takes yet. Share an insight, caveat, or question.
May enable efficient real-time ECG/EEG processing; extends methods but leaves open clinical validation.
Gogna et al. (2016) studied Arrhythmia and seizure (ECG and EEG signals). Semi-supervised Stacked Label Consistent Autoencoder vs. Compressed sensing (CS) based methods and traditional classification methods was evaluated on Reconstruction and classification of biomedical signals. A semi-supervised stacked label consistent autoencoder yielded better reconstruction and classification of ECG and EEG signals and was more than an order of magnitude faster than compressed sensing.
Synapse has enriched 4 closely related papers on similar clinical questions. Consider them for comparative context: