1*Research Scholar, Department of Electronics, Erode Arts and Science College, Erode.
2Controller of Examination, Erode Arts and Science College, Erode
3Associate Professor, Department of Electronics, Erode Arts and Science College, Erode
4Assistant Professor, Department of Electronics, Erode Arts and Science College, Erode
Received: 11th Dec, 2025; Revised: 11th Feb 2026; Accepted: 12th Feb, 2026; Available Online: 10th March, 2026
Continuous monitoring of various physiological parameters is vital to detect various health risks early on. The recent developments in embedded systems and machine learning have made it possible to develop intelligent healthcare monitoring systems that can process various physiological signals. This paper aims to propose an embedded machine learning (ML) based framework for personal health monitoring. The proposed framework includes a set of physiological sensors to measure various vital parameters such as heart rate, blood pressure, body temperature, ambient temperature, humidity and blood glucose levels. The proposed embedded device is connected to a set of sensors to collect data on various vital parameters. The collected data are then analyzed using a Random Forest (RF) classifier. A dataset of 300 physiological sample data was used to train the model. The Random Forest classifier is able to deal effectively with multiple physiological features and nonlinear relationships between health parameters. From the experimental results, it is clear that the system is able to effectively identify health hazards such as heat stress, dehydration, hypertension and abnormal glucose levels, thereby sending timely alerts to the users. The use of embedded sensors along with the RF model is an effective and affordable solution for continuous health monitoring.
Keywords: Embedded machine learning, Personalized health monitoring, Random Forest algorithm.
How to cite this article: Parventhan AC, Venkatachalam M, Gowthaman P, Sivasooriya V. Embedded Machine Learning Framework For Personalized Health Monitoring. Int J Drug Deliv Technol. 2026;16(3): 458-463. DOI: 10.25258/ijddt.16.3.53
Source of support: Nil.
Conflict of interest: None