On-demand Data Analytics Support for Hemorrhagic Stroke Patients Using Wearable IoT Device and Fog Computing Technology

Stroke is one of the high-risk diseases with a global concern that has caused the death of millions of individuals and is still ravaging lives of people in many countries-like fire in a wild forest. Stroke occurrences are highly influenced by age and other factors such as blood pressure, heart pulses, and feeding habits. Hemorrhagic stroke is prevalent among the age groups of 16 and 50 years-stimulated by chronic high blood pressure, heartbeat problems, and aging blood vessels. These health factors can be collected from patients and processed to monitor their health conditions in real-time-such as to reduce the number of death cases caused by stroke. In this research, an IoT-based healthcare delivery system for hemorrhagic stroke patients was developed using a wearable device and a fog computing approach for monitoring vital symptoms of the patients and administering on-demand health care services to the users. The IoT device is capable of detecting irregular blood pressure and heartbeat using a Continuous Wavelength Transform (CWT) signal and performing analytics on the data collected using a rule-based mining algorithm on the wearable firmware enhanced with a robust database model and a fog layer for efficient data analytics. The system was implemented in the form of a wearable armband for the user and supported with a dynamic web page for device allocation and real-time monitoring of the patient's health status. The wearable device was able to communicate to the user in real-time-notifying them of their health status, recommending preventive actions to be taken, and alerting their doctors when there are emergencies- using the fog computing layer to reduce data latency and provide fast response time. In evaluating the performance of our system, the readings from the wearable armband for monitoring the blood pressure and heartbeat were compared with the patient's reading using the standard Sphygmomanometer and Pulse oximeter, the results show high accuracy relative to the outputs of our wearable IoT device.