A Novel Deep Learning-Based Approach for Hypertension Level Detection Using PPG
Published in IEEE SILCON 2023, 2023
In the contemporary era, a significant portion of individuals endure cardiovascular ailments (CVDs). Hypertension stands as the principal cause behind blood pressure (BP) irregularities and diverse CVDs, making continuous BP monitoring an urgent priority. This work devises an efficacious deep learning-powered automated technique for measuring BP (specifically systolic blood pressure (SBP) and diastolic blood pressure (DBP)), leveraging potentially cost-effective technology.
Two Long Short-Term Memory (LSTM)-based regression models are formulated to prognosticate SBP and DBP based on recorded photoplethysmogram (PPG) readings. An attention mechanism-based TLSTM (tanh Long-Short Term Memory) model is further proposed that can accurately predict distinct stages of hypertension — Normal, Pre-Hypertension, Hypertension stage 1, and Hypertension stage 2. The attained root-mean-squared error (RMSE) values are 10.503 and 9.284 for SBP and DBP respectively, while mean absolute error (MAE) values are 7.529 and 4.218. The proposed attention-based TLSTM model exhibits a classification accuracy of 96%. The novelty of this work lies in incorporating an attention module into the TLSTM network to increase its classification accuracy.
Recommended citation: De, S., Mukherjee, P., & Halder Roy, A. (2023). "A Novel Deep Learning-Based Approach for Hypertension Level Detection Using PPG." IEEE SILCON.
Download Paper