INTERANTIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT · Jul 2026
A Comprehensive Review of Machine Learning Techniques for Non-Invasive Blood Pressure Prediction and Hypertension Detection
By Khandelwar NKN
Blood pressure prediction has become a major research focus due to the growing need for early hypertension detection and continuous health monitoring. Traditional cuff-based measurements are often inconvenient, intermittent, and unsuitable for real-time analysis. Recent advancements in machine learning provide powerful alternatives for non-invasive and continuous blood pressure estimation using physiological signals such as photoplethysmography (PPG), electrocardiography (ECG), and demographic attributes. This review paper presents a detailed analysis of machine learning and deep learning models applied to blood pressure prediction, including regression methods, ensemble classifiers, and advanced architectures like CNN and LSTM. It explores feature extraction techniques, dataset characteristics, evaluation metrics, and current challenges. The study highlights recent developments, research gaps, and future opportunities for improving accuracy, reliability, and real-time clinical applicability.