An intelligent healthcare system with peer-to-peer learning and data assessment

Modern e-healthcare systems are prevalent in many medical institutions to reduce physicians' workload and enhance diagnostic accuracy, which leverages affordable wearable devices and Machine-Learning (ML) techniques. The healthcare systems collect various vital biosignals (e.g., heart rate and blood pressure) from wearable devices of users (e.g., chronic patients living alone at home) and analyze these patients' data in real-time by different ML classifiers (e.g. Support Vector Machine (SVM) or Hidden Markov Model (HMM)). The automatic diagnosis effectively improves the physicians' performance in terms of diagnostic efficiency and accuracy. There are three challenges impacting these healthcare systems -- the increasing number of patients, new diseases and the changes of existing disease patterns, which are caused by population aging as well as the alteration of environment and lifestyle. This research is intended to explore a novel healthcare system with advanced ML solutions that can solve the challenges and exhibit high accuracy and efficiency.<br>