Ef-Zin: A hybrid framework for ubiquitous management of comorbidity and multimorbidity in chronic diseases

The existence of comorbidity and multimorbidity increases the diagnostic uncertainty and has a variety of negative social and economical impacts. This paper proposes the Ef-Zin framework that aims to manage patients suffering from chronic conditions by means of (a) creating collaborative virtual groups through medical and paramedical professionals and (b) delivering the appropriate therapy to the individual patient. Ef-Zin involves two distinct processing phases for parallel evaluation of patient´s contextual information. For the evaluation it uses rule-based algorithms and Random Forest (RF) machine learning technique for categorizing patients into groups according to the severity levels, making decisions about the services that will be delivered and notifying the appropriate specialized healthcare professionals for patient´s current health status. We have carefully drafted an architecture of the proposed Ef-Zin framework and qualitative evaluation has been conducted in a common use case scenario such as Chronic Obstructive Lung Disease (COPD) and a cardiovascular disease (hypertension) that is the most frequent and significant disease that coexists with COPD.