Abstract:
Medical data is one of the most rewarding and yet most complicated data to analyze. Medical
images, biomedical signals and handwritten prescriptions are available and can be used for prediagnostic tasks on the existence of chronic disease by assuming big data analytic concepts.
Hence, the main objective of this study was to design a big data analytics prototype that process
and visualize the huge amount of dataset by using R-studio programming software. Big data
processing and visualization is a challenge that needs new way of tackling which otherwise cannot
be solved with current practice of data management because data deluge and data creation
frequency in varieties of formats are inevitable scenarios.
A big data analytics system that descriptive the occurrence of chronic disease from the big medical
data was developed by using different methods and tools. In this study data computation techniques
is applied and descriptive analysis were employed. The major new data management techniques
are applied to ensure the quality of data and integrate data from different sources. Experimental
research design was employed for this study. In addition, (descriptive) analysis approach based
on a logistic activation function is employed to build the model. This study achieved as it is possible
to manage big data regardless of size and nature of data.
The major challenge faced during conducting this study is dealing with heterogeneous data in
order to generate insights for improved health-care outcomes or visualization of data. The other
most challenging task was the fact that data preserved in Jimma Medical center are disorganized
and distributed since it comes from various sources and having different structures and forms. The
researcher strongly recommend that prototype with the capability of analyzing and visualizing
heterogeneous big data should be developed. As new area of study, it is strongly recommended
further studies in specific contexts.
