DEVELOPMENT OF A STROKE PREDICTION SYSTEM BASED ON IOT AND MACHINE LEARNING

Abstract

Stroke is one of the leading causes of disability in many Asian countries, with low and middle-income countries bearing a higher burden of mortality. Worldwide Cerebrovascular accidents (stroke) are the second leading cause of death and the third leading cause of disability, where in Bangladesh it is the third leading cause of death. Effective prevention strategies include targeting the key modifiable factors like hypertension, diabetes, smoking, and high cholesterol. Due to the high cost of our diagnosis system, the majority of our people cannot go for checkups. Nowadays, IoT has unarguably transformed the healthcare industry and is highly beneficial for doctors, and patients. This project proposes a prototype IoT-based Brain Stroke Prediction System by analyzing the key risk factors of stroke and predicting the associated risk status using machine learning technology. Blood glucose level, hypertension status, heart disease status, smoking status, marital status, age, gender, BMI, working type, and residence type are the risk factors employed for this proposed system. The proposed prototype is able to detect blood glucose level which is one of the key risk factors of stroke non-invasively using NIR spectrology. It then analyzes other key risk factors of stroke along with blood glucose level using machine learning technology. The final outcome of this proposed protype is the associative risk status of a person in positive (high) or negative (low) format.