Cuff-less blood pressure monitoring via PPG signals using a hybrid CNN-BiLSTM deep learning model with attention mechanism

Summary

This Scientific Reports paper from Shahid Beheshti University (Tehran) presents a hybrid CNN-BiLSTM model with multi-head attention for cuffless BP estimation from PPG signals. The hybrid architecture captures both local morphological features (CNN) and long-range temporal dynamics (BiLSTM), with attention providing input-adaptive feature weighting. The model is evaluated on 2,064 patients from the MIMIC-II database and achieves SBP MAE of 1.88 mmHg and DBP MAE of 1.34 mmHg, significantly outperforming CNN-only and LSTM-only ablations.