Categorization of patients' health status in COPD disease using a wearable platform and random forests methodology

Chronic diseases are diseases of long duration and generally slow progression. Chronic Obstructive Pulmonary Disease (COPD) as one of these requires frequent examinations and hospital visits for its long-term management. CHRONIOUS, an integrated platform for chronic disease patients, provides functional information about their health status continuously, from the patients' environment. The main component of the system is the intelligent core of the wearable device that aims at the real-time characterization of the patients' health level based on the fusion of multiple data that acquired by wearable sensors, through patients smart device interface or retrieved from the specific system's database. The Decision Support System (DSS) is activated whenever new data are entered, and is functioning using by two classification methodologies composed by highly effective algorithms. These parallel classification techniques are: a rule-based expert system and a supervised Radom Forest classifier. The classification conclusion after the above analysis is twofold. The supervised technique provides with high accuracy the severity of patients' health level and the rule-based part, extracts the critical parameters or their combination which had been triggered leading in the improvement or the worsening of the patient's condition.