Introduction

Thousands of patients die in hospital each year in the world, being often preceded by changes in physiological parameters, such as pulse, blood pressure, temperature and respiratory rate, these deaths are potentially preventable. If these changes in the patient’s vital parameters are recognized early, serious adverse events like cardiac arrest and deaths can be prevented. Our aim here is to create early warning systems that should enable a more timely response to, and monitoring of patients. Applying techniques and algorithms based on machine learning to the analysis of vital-sign data collected from patients being administered in hospital. As data can be collected from wearable sensors as used by people to monitor and keep track of physiological parameters, it can be used to trigger early warnings like patient have 70% chances of developing diabetes so she should see a doctor. It may benefit patients significantly by predictive and personalized monitoring systems.