The rapidly increasing availability of healthcare data is becoming the driving force for the adoption of data-driven approaches. However, due to a large amount of heterogeneous dataset including images (MRI, X-ray), texts (doctor’s note) and sounds, doctors still struggle against temporal and accuracy limitations when processing and analyzing such big data using conventional machines and approaches. Employing advanced machine learning techniques on big healthcare data anlaytics supported by Petascale high performance computing resources is expected to remove those limitations and help find unseen healthcare insights. This paper introduces a data analytics pipeline consisting of data curation (including cleansing, annotation, and integration) and data analytics processes, necessary to develop smart healthcare applications. In order to show its practical use, we present sample applications such as diagnostic imaging, landmark extraction and casenote generation using deep learning models, for orthodontic treatments in dentistry. Eventually, we will build smart healthcare infrastructure and system that fully automate the set of the curation and analytics processes. The developed system will dramatically reduce doctor’s workload and is smoothly expanded to other fields.
CITATION STYLE
Lee, C., Murata, S., Ishigaki, K., & Date, S. (2017). A Data Analytics Pipeline for Smart Healthcare Applications. In Sustained Simulation Performance 2017 (pp. 181–192). Springer International Publishing. https://doi.org/10.1007/978-3-319-66896-3_12
Mendeley helps you to discover research relevant for your work.