Figure 4.

Proposed ECG identification architecture using LSTM-based RNN Model. The inputs are raw signals preprocessed from datasets, segmented into ECG components with a window size of T, and trained at the LSTM-based RNN model.