Evaluating the Discriminative Power of Cardiovascular and PPG-Derived Biomarkers for Blood Pressure Classification: A Machine Learning Approach

Hypertension is a major global health concern and a leading risk factor for cardiovascular diseases, stroke, and kidney failure. Accurate and continuous blood pressure (BP) monitoring is crucial for early detection and effective management of hypertension-related conditions. Traditional cuff-based BP measurement methods, although widely used in clinical settings, are impractical for continuous tracking due to their intermittent nature and potential discomfort. Consequently, there is growing interest in developing non-invasive, cuffless BP monitoring techniques that rely on physiological signals such as photoplethysmography (PPG) and other cardiovascular biomarkers. In this context, identifying reliable markers that can accurately classify BP levels is essential for advancing cuffless monitoring technologies. This study aims to assess the effectiveness of selected cardiovascular and PPG-derived biomarkers in distinguishing normal from abnormal BP, following a structured approach that includes signal preprocessing, feature extraction, and machine learning-based classification.