Benefits and Challenges of Big Data in Healthcare

Big data has come a long way to reach its present-day status. It was first mentioned back in 1997 at an IEEE conference to explain data visualization and the challenges it posed for computers.32 During the 1990s, technological improvements and IT advances made possible the use of large amount of data but in parallel little knowledge could be gained by its creation. The beginning of 2000s until 2008, saw the breakthrough period for big data advances. Big data was defined for the first time in terms of volume, velocity and variety.

After this definition the development of the necessary software for handling the boom in data creation. This software enabled end users to process huge amount of data within and between different 32 Cox,M., Ellsworth, D., 1997. Application-controlled demand paging for out-of-core visualization.

Proceedings of the 8th IEEE Conference on Visualization. IEEE Computer Society Press, Los Alamitos, CA. organizations cooperatively and in real-time. Around that period, started the digitalization of medical records and clinical data in electronic databases, thus making medical data accessible, usable, and interactive so that healthcare providers could be assisted in their work.

2009 was the year that saw the breakthrough point in big data analytics. 33 Banks and e-commerce enterprises were the first to implement big data analytics computer systems and experienced an improvement in business processes, workforce effectiveness, reduced costs and in attracting new customers. By 2011 the amount of stored healthcare data worldwide had reached 150 exabytes (1 EB = 1018 bytes), the majority of which electronic health records.

Big data, however, still has a long way to reach its true potential since simulation techniques and prediction models for analyzing heath data have not been developed in a sufficient manner.

The most recent development in big data analytics is cloud use in connection with data.

The trend in healthcare industry is the shift from the use of structured-based to semi structured-based (e.g., sensor wireless devises, telehealth, home monitoring) and unstructured data (e.g. notes, video, images). The use of sensors and monitors is the key in supporting home healthcare services because of the data that will be generated through these sensors, thus improving the healthcare services quality by increasingly accurate analysis and prediction.

At this point it is important to mention the sources of big data in healthcare. Therefore, the most important big data sources are Electronic Health Records (EHRs), medical imaging data, clinical notes, and genetic data.

Describing the architecture, components and functionalities of big data analytics is the necessary step in order to explain its capabilities and benefits. Wang, Y., et al., in their study proposed a life cycle framework for healthcare big data analytics architecture that includes, data capture, continues with data transformation and last leads to data consumption.36 The proposed architecture is shown in Figure 3.

1. 33 Bryant, R.E., Katz, R.H., Lazowska, E.D., 2008. Big-data computing: creating revolutionary breakthroughs in commerce, science, and society computing. Computing Research Initiatives for the 21st Century. Computing Research Association 34
2. Wang, Y., et al., Big data analytics: Understanding its capabilities and potential benefits for healthcare organizations, Technol. Forecast. Soc. Change (2016)
3. 35 Manogaran, Gunasekaran & Thota, Chandu & Lopez, Daphne & Vijayakumar, V & M. Abbas, Kaja & Sundarsekar, Revathi. (2017). Big Data Knowledge System in Healthcare. 10.1007/978-3-319-49736-5_7. 36
4. Wang, Y., et al., Big data analytics: Understanding its capabilities and potential benefits for healthcare organizations, Technol. Forecast. Soc. Change (2016)