Key result
Hybrid CNN-LSTM system achieves ~100% accuracy classifying five cardiac valvular conditions from phonocardiogram signals.
Why the study?
Automated and early diagnosis using PCG signals and deep learning models can help alleviate deadly complications of cardiovascular diseases.
Does a hybrid CNN-LSTM deep learning system improve the classification of cardiac valvular diseases from phonocardiogram signals compared to previous models?
Population
PCG signals from an open heart sound dataset and the PhysioNet/Computing in Cardiology 2016 challenge dataset
Comparison
Combined CNN and LSTM system vs previous works using the same databases
Design
Machine learning model development and validation study
Authors
Loading...
May advance AI phonocardiogram screening; leaves open prospective clinical validation before practice change.
Does a hybrid CNN-LSTM deep learning system improve the classification of cardiac valvular diseases from phonocardiogram signals compared to previous models?
A hybrid CNN-LSTM deep learning model demonstrated high accuracy (>98%) in classifying five cardiac valvular conditions using phonocardiogram signals, outperforming previous models.
Al-Issa et al. (2022) studied Cardiac valvular disease. Hybrid CNN and LSTM deep learning system was evaluated on Classification accuracy, F1-score, and AUC for heart valvular conditions. A hybrid CNN and LSTM deep learning system using phonocardiogram signals achieved 99.87% accuracy and 0.9985 AUC for classifying five cardiac valvular conditions with an augmented dataset.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: