Transfer Learning with CNN-LSTM for detection of Cardiac Events in Arrhythmia Disease

Every year, a significant number of individuals lose their lives to a catastrophic heart condition referred to as Sudden Cardiac Arrest (SCA), often caused by Arrhythmia. Henceforth, the possibility of early prediction or diagnosis of SCA holds the potential to prevent numerous fatalities. The objective of our research is to create an efficient approach that conserves both space and computation time for the real-time categorization of SCA. To accomplish this, a novel strategy has been proposed by integrating transfer learning into our existing LSTM model. This approach entails the utilization of pre-trained neural networks that were originally developed for tasks like image recognition and language processing. The proposed networks are fine-tuned to cater to the specific requirements of Sudden Cardiac Arrest (SCA) classification. The objective is to leverage the wealth of knowledge embedded within these pretrained models, customize it to suit our domain, and potentially enhance both the efficiency and accuracy of our classification system.