Real Time IoT and Machine Learning Based Monitoring System for Bedridden Patients

Monitoring bedridden patients during their recovery often requires constant attention to notice any significant changes in their health and provide the necessary action on time. Traditional monitoring relies almost entirely on manual observation and any deficiencies in staff could delay necessary actions and reduce their effectiveness. This paper describes the development of a Real-Time Continuous Health Monitoring System for bedridden patients utilizing IoT and Machine Learning. The monitoring system is designed with multiple biomedical sensors, including heart rate, temperature, oxygen saturation and motion sensors, interfaced with an IoT driven microcontroller allowing real-time collection and transmission of data to a cloud platform. Then, the collected data is analyzed using machine learning algorithms to detect anomalies and predict risk for health deterioration. In the event of an abnormal finding, alerts are automatically forwarded to the clinicians or family members of the patient mobile applications. This intelligent system will eliminate the possibility of human error and allow for 24/7 monitoring of bedridden patients which will allow for early detection of health deterioration and shortening of response times.