Estimation of Blood Pressure and Glucose through Photoplethysmography and Artificial Intelligence

Abstract

This article presents the development of a system for non-invasive measurement of blood pressure and glucose levels using photoplethysmography (PPG) and machine learning models. The system, called CARDIOPRESS, combines an optical sensor to capture heart rate and oxygen saturation (SpO2) data with machine learning algorithms to predict systolic and diastolic blood pressure, as well as blood glucose levels. Experimental results demonstrate a good approximation in the prediction of both blood pressure and glucose levels, validating its potential for continuous monitoring of cardiovascular and metabolic health.