Patient monitoring and care delivery can undergo transformative changes through the adoption of Internet of Things (IoT) technologies in the healthcare industry. This study delves into the utilization of IoT for continuous health monitoring of senior patients, a particularly relevant endeavor during situations like pandemics and remote home treatments in intensive care units (ICUs). The research establishes a comprehensive IoT-based healthcare framework by integrating various sensors, machine learning algorithms, and cloud computing. The study encompasses a diverse group of 50 participants, including older individuals in good health and those with various health issues. Artificial neural network (ANN), long short-term memory (LSTM), and decision tree machine learning models are employed to analyze sensor data generated by temperature and pulse rate measurements. These models aim to proactively identify health irregularities, enabling timely intervention. The findings showcase that the ANN outperforms the LSTM and decision tree models in terms of predicting anomalies. The accuracy, recall, and F1 score of the ANN demonstrate its proficiency in recognizing health deviations while reducing false positives. This predictive capability holds the potential to revolutionize healthcare procedures by ensuring early identification and appropriate care. The research underscores the significance of incorporating predictive analytics into IoT-enabled healthcare frameworks. The continuous monitoring, proactive forecasting, and remote accessibility of such systems cater to the evolving needs of modern health care.