Dynamic Data mining for Smart mHealth

The traditional patient monitoring systems works with threshold alarming mechanism. When certain value of the signal reaches to the threshold, system raises alarm. However such systems are not true all the time and leads to false positive prediction. In addition of this, these systems do not support mobility; hence real time monitoring of the patient is not possible. To overcome these drawbacks window based mining of vital signals is possible, where the mining approaches are applied on the signal window in order to extract risk components and predict worst health condition. Also mining systems requires past prediction data of the patient in order to predict the future risk patterns, which is a bottleneck of the system where predicting the risk for new patient is impossible. This mining bottleneck is overcome with a novel Real Time system, where the risk pattern of completely new patient would be predicted. This book provides you the design of real time patient monitoring system to monitor the patient in real time intelligently and ubiquitously where the signals are mined using online mining algorithms on mobile devices.