Big Data in Healthcare

Question: Discuss about the Big Data in Healthcare. Answer: Introduction Today, the concepts of big data are a vital aspect of business success; in fact, the services and solutions offered by big data are no longer optional company capabilities but necessities for the survival of big businesses and organisations. This notion is evidenced by the financial and operational edge provided by big data solutions (ATKearney 2017). Studies such as those done by the likes of Bain Company (2013), suggests that over 400 companies around the globe and with revenues of over $1 billion depend on this new technology to deliver their own services. However, what makes it suitable for businesses today as compared to other traditional methods? According to Taylor (CEO FICO) (2013), big data enables organisations to leverage on available data/information to make better decisions. Furthermore, the drive for smarter and better decisions is facilitated by customers information, which is now readily available. Now, this report assess the application of big data solutions and the services they offer to the healthcare sector while comparing them with other traditional data management systems. Through this assessment, processes used to identify big data solutions/technologies are provided as well as the impacts of the said technologies. In general, the term big data refers to data tasks or systems that manage large information assets that normal and conventional database systems are unable to handle. However, through a technical eye, big data will encompass technologies such as NoSQL, MapReduce and even Hadoop which offer solutions to the existing problems of data particularly those that have large volumes of unstructured information (Gaffney 2014). In application, two forms are used; online and offline big data systems. Online systems will have real-time support where data is created, ingested and transformed by the said system. On the other hand, offline systems will perform the same roles but in a batch mode that has an interactive output. Figure 1: Big data outline (Oracle 2013) The modern healthcare facilities and the expenditures involved have forced the industry to adopt systems that are big data driven. These technologies offer increased efficiencies in decision-making processes, which attract more economists and facilitates a rapid pace of innovation. Moreover, the healthcare sector is also guaranteed of other benefits offered by the advanced analytics seen in both online and offline big data systems. Consider the MapR Converged data platform, a platform that offers a wide range of solutions including; a united system that monitors fraud and manages resources, services that streamline system records, and integrates internet of things into the healthcare sector (McDonald 2017). Moreover, sub-components of the application such as UnitedHealthcare are already utilised by more than 51 million people and in over 6100 hospitals across America (McDonald 2017). Differences between online and offline BD system In online big data systems, the data is sourced online, therefore, creating new information. This outcome necessitates low-latency level in order to meet the user requirements more so the SLA stipulations. As an application, online big data will have a wide range of services including; product catalogues, websites, data sensors and analytics, e-commerce among many others. Examples of online big data include; MongoDB and NoSQL (Leone 2013). Offline big data, on the other hand, does not create any new data and instead will use the interactive platform to solve the problems raised by the users. Again, this outcome means they produce static (fixed) solutions that are presented as the end outputs. Therefore, they can go offline without impacting the overall goals of the systems. Examples of offline big data are Spark, Hadoop and other business intelligence tools (MongoDB 2016). Selecting a big data application System performance is always the ultimate goal of any big data application. The leadership in the healthcare sector will have little concerns over the different components of the system so long as it meets its overall objective (Regola, Cieslak Chawla 2013). Similarly, businesses and projects will emphasise on the desired outcome as compared to the elements of the applications. To meet this objective, some items are necessary and must be observed during the selection, they are: Big data platform Architecture components define the capabilities and the analysis aspects met by the big data application. In addition to this, the system architecture will determine the systems organization and functions. Therefore, the system design should come first to determine the overall outline of the big data system. As a good practice, the architecture should be able to consume myriad data sources in an efficient manner. Example: Figure 2: Big data architecture (George 2014) Storage methods Having identified the architecture, its also appropriate to select a storage system based on the size of the data and users. Online or offline system This is a critical assumption that will eventually determine the latency level, therefore, define the application delays. Furthermore, offline applications can also have in-memory solutions, which are faster, and process data at nearly real time pace (George 2014). User accessibility User integration will define the interface considered. For instance, NoSQL databases require certain interfaces to access them. Therefore, the access method should align with the tools used to develop the applications Data type Data serialisation should be considered when using an unstructured approach that includes streaming data such as that found in social media. Data serialisation will facilitate capture and representation of such data, which occurs in high velocity (Millman 2017). System integration If the big data application is set to use an already existing data warehouse, then data integration tools must be considered. As an added advantage, vendors who deal with big data platform will also provide these solutions thus provide a support for the integration process. (Modified from Nick Millman work 2017). Big Data technologies Big data analytics has expanded over the past few years to include mainstream consumers unlike before when it was used by large organisations but with minimal customer impact. Today, big data technologies are categorised based on their demand and the potential for growth. In healthcare, for instance, technologies such as those seen below are used to store and process records. Furthermore, they streamline information captured by sensors or medical machines attached to patients. These features improve examination methods such as the mapping of human genome among other many applications (Singh, Singh, Garg Mishra 2015). Technologies: Column oriented databases (COD) Due to modern volumes of data that are ever growing traditional row-oriented database systems fall short in query performance. This necessitates the importance of COD that use column modules allowing faster query time and improved data compression rates (Rodrigues 2012). NoSQL database A key big data technology that enables data documentation/storage through graph database systems. This technology is appropriate for assessment and analysis, a key practice of business management (Rodrigues 2012). Hadoop An open source platform that is used to handle big data. Moreover, its the most popular method of implementing MapReduce which among its benefits has a high flexibility to function with multiple data sources (Gill Press 2016, Rodrigues 2012). MapReduce Having highlighted it above, its a programming technique that enables users to execute scalable and massive jobs using many different servers. It will function into two steps; one mapping where inputs are converted into datasets of specific values and two, reduce where outputs are combined to form reduced sets of data (Modified from Gill Press 2016). Hive A bridge technology that allows normal BI programs to run queries using Hadoop clusters. Therefore, through its operations it enhances the reach of Hadoop platform improving its application among BI users (Rodrigues 2012). Business Impact of Big Data Information technology has revolutionised business more so by creating an extensive global market. This global marketplace has many supplier and consumers, which have produced the vast amount of information available today. In fact, the volume of data available today doubles every 18 months (IDC 2010). This flood of information is the root concept of big data, which can challenge businesses or provide considerable opportunities. Challenges of big data One of the predominate challenges with big data is poor data quality where businesses are unable to obtain accurate information based on their requirements. In fact, the price of obtaining accurate information is the actual admission fee of entering into the business market (ATKearney 2017). Moreover, when one considers the aggregate challenges that are experienced due to the quality of data, the price of business intelligence becomes significantly high which is a common problem of starting a business. Furthermore, consider other challenges that are produced by this problem; inaccurate prospects, excessive data sources that require extended time for analysis and long development time (IDG Connect 2014). Opportunities Through big data, enterprises can identify and filter customers based on their requirements. This makes the customer the heart of business thus improving the level of customer intimacy, which generally increases the customer base, and in return an organization can increase its revenue. A hospital, for instance, can increase its throughput through efficient management of customer information (ATKearney 2012). Furthermore, with big data services, an enterprise is able to utilise the vast wealth of information collected throughout the years. Unlike before, this unused data is used to produce better products and services that are focused on the customers needs. Therefore, big data increases product innovation a critical element for business survival. Finally, big data can manage business operations through assessment technologies such as those that use sensors and radio frequency identifiers (ATKearney 2017) Organisational Impact of Big Data. Big data is generally thought to improve an organisations performance, however, this impact is dependent on the organisations resources visa vie those of the application used. In some cases, the big data application may lower the productivity of an organisation if its set as the sole managerial product and not as a support mechanism (Ghasemaghaei, Hassanein Turel 2015). According to a study by McKinsey Global Institute (MGI) (2015), most organisations have zero to negative big data impact, as their analytics are limited to tests and shallow analysis in some slices of their businesses. However, in other organisations such as those in the healthcare sector big data increases operational margins by over 60 percent. In addition to this, the US healthcare sector has experienced reduced operational costs of over 8 percent through the services offered by big data especially those of data analytics and quality advancements (Court 2015). Nevertheless, an organisational impact is dictated by the strategy used to capture information that is later used for analysis. As a start, organisation transformation will start with a plan based on the demand at hand. In addition to this, cultural challenges (organisation culture) must be addressed prior to the incorporation of the new data system. This outlook will provide a positive impact on an organisation leading to success. Conclusion Big data services, solutions and technologies offer many benefits as compared to other traditional data management systems. Data analytics a defining characteristic of big data enables organisations to make better decisions that enhance business intelligence. Consider the sector highlighted, the healthcare sector, this industry has been able to forecast patient activities, which have helped them, meet the current medical demand. Today, hospitals and other clinical facilities that use big data technologies such as MapR can increase their services through efficient mechanisms, which in return have increased their throughput. Therefore, when properly selected and implemented, big data increases the overall efficiency of an organisation particularly those in the business sector. References ATKearney, 2012. Big Data and the Creative Destruction of Todays Business Models. Online. Available at: https://www.atkearney.com/documents/10192/698536/Big+Data+and+the+Creative+Destruction+of+Todays+Business+Models.pdf/f05aed38-6c26-431d-8500-d75a2c384919 ATKearney, 2017. Big Data: The Next Leading Edge in the Financial Industry. Financial institutions. Online. Available at: https://www.atkearney.com/financial-institutions/ideas-insights/featured-article/-/asset_publisher/4rTTGHNzeaaK/content/big-data-the-next-leading-edge-in-the-financial-industry/10192?_101_INSTANCE_4rTTGHNzeaaK_redirect=%2Ffinancial-institutions%2Fideas-insights Avanade, 2010. Global Survey: The Business Impact of Big Data. Online. Available at: https://www.avanade.com/~/media/asset/point-of-view/big-data-executive-summary-final-seov.pdf Court. D, 2015. Getting big impact from big data. Mckinsey Quarterly. Online. Available at: https://www.mckinsey.com/business-functions/digital-mckinsey/our-insights/getting-big-impact-from-big-data Gaffney. B, 2014. What is Big Data? Himss Clinical and business intelligence. Available at: www.himss.org/file/1242441/download?token=sQoZJ5uB Ghasemaghaei. M, Hassanein. K Turel. O, 2015. Impacts of Big Data Analytics on Organizations: A Resource Fit Perspective. Impact of Data Analytics on Organizational Performance. Available at: https://aisel.aisnet.org/cgi/viewcontent.cgi?article=1030context=amcis2015 George. L, 2014. Getting Started with Big Data Architecture. Cloudera. Online. Available at: https://blog.cloudera.com/blog/2014/09/getting-started-with-big-data-architecture/ IDG Connect, 2014. Big Data for Marketing Sales: Data Accuracy to Business Impact. Online. Available at: https://info.avention.com/rs/onesource/images/Whitepaper_IDGBigDataForMarketingandSales.pdf McDonald. C, 2017. 5 Big Data Production Examples in Healthcare. Online. Available at: https://www.mapr.com/blog/5-big-data-production-examples-healthcare Millman. N, 2014. 8 considerations when selecting big data technology. Online. Available at: https://www.computerworld.com/article/2475840/big-data/8-considerations-when-selecting-big-data-technology.html MongoDB, 2016. Big Data: Examples and Guidelines for the Enterprise Decision Maker. Available at: https://s3.amazonaws.com/infomongodbcom/10gen_Big_Data_White_Paper.pdf Oracle, 2013. Oracle: Big Data for the Enterprise. Oracle white paper. Online. Available at: https://www.oracle.com/us/products/database/big-data-for-enterprise-519135.pdf Press. G, 2016. Top 10 Hot Big Data Technologies. Online. Available at: https://www.forbes.com/sites/gilpress/2016/03/14/top-10-hot-big-data-technologies/#4fbab1f065d7 Regola. N, Cieslak. D Chawla. N, 2013. The Need to Consider Hardware Selection when Designing Big Data Applications Supported by Metadata. Online. Available at: https://www3.nd.edu/~nchawla/papers/bigdata13.pdf Rodrigues. T, 2012. 10 emerging technologies for Big Data. Tech republic. Online. Available at: https://www.techrepublic.com/blog/big-data-analytics/10-emerging-technologies-for-big-data/ Signh. S, Singh. P, Garg. R Mishra P, 2015. Big Data: Technologies, Trends and Applications. International Journal of Computer Science and Information Technologies. 6(5). Available at: https://ijcsit.com/docs/Volume%206/vol6issue05/ijcsit20150605101.pdf Taylor. J, 2013. Delivering Customer Value Faster With Big Data Analytics. Decision management solutions. Online. Available at: https://www.fico.com/en/wp-content/secure_upload/DeliveringCustomerValueFasterWithBigDataAnalytics.pdf Wegener. R Sinha. V, 2013. The value of Big Data: How analytics differentiates winners. Bain and Company. Online. Available at: https://www.bain.com/Images/BAIN%20_BRIEF_The_value_of_Big_Data.pdf