Predictive & Preventive Healthcare: A Granular Distributed Model Utilizing An Empirically Effective Activation Function

Abstract

Healthcare analytics is most confidently considered as an specific domain that has interconnected healthcare analytics for numerous areas, as well as precautionary health, well-being and unwellness management. In particular to chronic  diseases management, affected  populations in several  disease classes,   the   field  of  healthcare analytics  assists   target the customized management procedures  and  conventions that  may   alleviate  the   chronic   disease, in addition as  hamper  the  onset  of  affiliated  medical   conditions. Due  to increase   in  size  of  patient  related data  in  hospitals  over  a  decade, development of fast  information processing   systems   for  this  massive, noisy and  fuzzy natured data is the  need of the hour. The prototype design and analysis proposed  in this paper  aims to explore various  relations among  patient attributes, disease  risk  factors  which can be mapped to built  automated disease categorization and preventive healthcare systems. This paper presents the results obtained after performing experimental simulations on a standard healthcare related dataset. To  present  the  advantage  of the proposed  model,  the evaluated results are being compared with  some existing  approaches implemented  in the  past years.