As the healthcare sector has been disrupted by technological innovation in the past decade, a considerable amount of patient-oriented data is generated at an exponential rate in a very short period of time. This data can be in either structured or unstructured format. It requires the power of statistical analysis to analyze such a big volume of data. In the healthcare sector, numerous big data analytics tools and techniques have been developed to handle these vast volumes of data. These tools help in extracting useful information out of such data. This chapter focuses on big data analytics tools along with applications and challenges of leveraging big data analytics in the healthcare sector. This study depends on secondary data that have been gathered from various sites, journals, books, and other available e-content and contributes to the existing literature on leveraging big data analytics in the healthcare sector.
TopIntroduction
Big data refers to databases with elements and patterns that are more complicated than standard tables and records, necessitating procedures that go far beyond simple calculation (Rehman et al, 2021). Medical datasets are a mixture of structured, semi-structured, and unstructured data (Bora, 2019). The characteristics that help define big data are Volume, Variety, Velocity, Value and Veracity (Shastri & Deshpande, 2019)
Volume: Volume is depicted by the quantity and size of generated and stored data. No fixed pattern is there for volume of this data. This term is connected with mass scale data and it should be stored, analyzed and managed.
Variety: Variety is the type of data i.e., unstructured or structured. Data can be in different forms i.e. images, text, video, audio or a combination of these.
Velocity: it refers to the speed of data generation and processing. Big data is produced more while compared to small data. Big data is generated and handled with the help of velocity.
Veracity: It is represented by the data's quality or dependability.
Value: The worth of the data being retrieved represents value. Big-data analytics results must be reliable and error-free.
As per IDC’s Data Sphere and Storage Sphere reports, big data in healthcare is expected to touch the height of 175 ZB in 2025.
Figure 1.
Increasing big data in healthcare
The variety and volume of big data are incompatible with traditional data management tools. To provide high quality healthcare services through patient-centered model, massive amounts of health data must be managed and analyzed (Kakandikar & Nandedkar, 2020). Big data analytics and applications in healthcare are still in their infancy, but rapid developments in tools and platforms can speed up their maturation (Raghupathi & Raghupathi, 2014). Generally, healthcare data is unstructured, arises in silos, and is stored in imaging systems, medical prescription notes, insurance claims data, EPR, and so on. Assimilating these data from disparate sources and incorporating them into advanced analytics is critical for improving healthcare services (Mathew & Pillai, 2015). The emerging landscape of big data and analytical techniques in the five sub-disciplines of healthcare, i.e., medical image analysis and imaging informatics, bioinformatics, clinical informatics, public health informatics and medical signal analytics (Rehman et al, 2021).
TopLiterature Review
Mathew & Pillai (2015) discussed that the lack of integration of heterogeneous data sources, skilled resources, privacy and security, infrastructure, data quality are the major challenges of Big Data adoption in healthcare. This research also presented various analytical tools that can be used to get benefits from the massive amount of healthcare data such as MapReduce, HDFS, HBase, Sqoop, SkyTree, Storm etc.
Nambiar et al (2013) highlighted several initiatives utilizing the potential of Big Data in healthcare like more effective treatment of asthma using a GPS sensor on top of an inhaler, battling the Flu through FluView which provide real-time information to help combat people at risk or already with metabolic syndromes using Reverse Engineering and Forward Simulation (REFS) technology and providing five warning signs like large waist size, high blood pressure, high triglycerides, low High density Lipoprotein (HDL), and high blood sugar.