An Innovative Method for Estimating Blood Pressure and Classifying Hypertension Levels Using PPG

An Innovative Method for Estimating Blood Pressure and Classifying Hypertension Levels Using PPG The present invention introduces a novel method for estimating systolic blood pressure (SBP) and diastolic blood pressure (DBP) using photoplethysmography (PPG) data, combined with an artificial neural network (ANN) model to classify hypertension levels. The system is designed for real-time, non-invasive blood pressure monitoring and operates on a low-cost hardware platform, including an Arduino UNO microcontroller and a PPG sensor. The empirical model used for blood pressure estimation is derived from normalized PPG data, allowing for personalized and accurate blood pressure readings. In addition to blood pressure estimation, the invention classifies hypertension into four categories: Normal, Pre-Hypertension, Stage-1 Hypertension, and Stage-2 Hypertension. This classification is achieved using an attention-based ANN model that processes estimated SBP, DBP, and heart rate data. The ANN’s architecture, which includes hidden layers and an attention layer, enhances classification accuracy by assigning appropriate attention weights to relevant outputs. The hardware system is integrated with a Wi-Fi module, enabling remote data transmission to a cloud platform, allowing users to monitor their blood pressure in real time via a mobile application. This invention provides a portable, affordable, and accessible solution for continuous blood pressure monitoring, offering a significant advantage over traditional cuff-based methods. It contributes to better management and early detection of hypertension, thereby improving cardiovascular health outcomes.