Real-Time ECG Signal Classification Using Recurrent Neural Networks
Shavnam SharmaChandigarh Group of Colleges,Chandigarh Engineering College,Department of Computer Application,Mohali,Punjab,IndiaMayank NagarZaid AlsalamiThe Islamic University,College of Technical Engineering,Department of Computers Techniques Engineering,Najaf,IraqEshonov RavshanbekFergana Public Health Medical Institute,Muhammadmusayevich,Teacher of the Department of Biomedical Engineering,Fergana,UzbekistanChippy MohanCHRIST (Deemed to be University),School of Business & ManagementRajesh Kanna RCHRIST University,Department of Computer Science,Bangalore
2024en
ABI
Annotatsiya
This study investigates the use of Recurrent Neural Networks (RNNs) for real-time ECG signal classification. The CNN-RNN hybrid model achieved the highest accuracy of 97.5% and an F1-score of 96.5%, while the GRU model offered the best real-time performance with a latency of 40 ms and a throughput of 25 segments/s. Model optimization techniques reduced model sizes by 50%, facilitating deployment on edge devices. These results indicate that RNN-based models can enhance real-time cardiac monitoring, improving early detection and intervention for cardiovascular diseases.
Mavzular
Identifikatorlar
Iqtiboslar va manbalar
0 ta iqtibos12 ta foydalanilgan manba
Koʻrsatkichlar — AkademScholar · Tez orada