Outlier detection in Wireless Sensor Network for Health Care: A statistical approach

A wireless sensor network in the healthcare system collects data from remotely observed patients using contemporary medical sensors and a reliable network. The sensors used to generate a large amount of data. Two types of unusual events that may occur during data generation are anomaly data from malfunctioning sensors that causes a false alarm or data from a reliable sensor that reports abnormal values that causes a medical emergency in the patient's health condition. To improve remote patient monitoring systems and provide better healthcare, researchers must find out how to distinguish between a genuine medical emergency and a false alarm. This study recommends utilizing Spearman's rank correlation and simple linear regression to detect outliers in medical wireless sensor networks. The outliers are identified using regression with high detection accuracy and a low false alarm rate.