Machine learning approach to revolutionize use of Holistic Health Records for Personalized Healthcare

Abstract

 Last few decades have witnessed a noteworthy transition in Healthcare industry, the one in which paper based work was more & now the other where digital media is used very frequently.   Electronic Healthcare Records (EHRs) are proving as valuable resources for knowledge extraction & predictive modelling which enables support for cost-saving and timely decision making in offering a treatment to a patient. One can analyse & process these healthcare records to transform the way care is delivered and compensated. Integrating and transforming these heterogeneous data into single format needs efficient Machine Learning techniques. Precise analysis of medical records benefits patient care, early disease detection and community services. But, when the quality of medical data is incomplete, the analysis of accuracy gets reduced. With regards, machine learning (ML) based artificial intelligence plays a significant role which is used for large and complex issues and also consumes less computation time due to stochastic feature of the search methods.  However, we are facing issues like less communication efficiency that includes lack of cognitive bias, training, recognition, and cultural hierarchy in a medical application.  Also, there is a lack of tool development which has been neglected in most of the research and still relies on manual assessment. Hence there is a need of an effective model that will enhance the diagnosis performance for timely medical care which is followed by appropriate succeeding treatment, thus resultant in significant lifesaving. This paper focuses on the development of effective supervised learning-based ML framework which would serve as a valuable aid to detect/categorize disease and make a real-time effective clinical decision.