Big Data Analytics for Healthcare Industry
Abstract
Big data comprises structures, semi-structured, and unstructured data generate from a wide variety of sources. The healthcare is one that does not only generate big data but also which would make great use of big data analytics. In the article titled Big Data Analytics for Healthcare Industry: Impact, Applications, and Tools, Kumar and Singh (2019) indicated that the main sources of big data are electronic health records, insurance companies, payer, and service providers. While the contextualization of big data was not in any way misplaced it failed to appreciate the greatest source of unstructured big data for the healthcare sector and that is the internet. This review considers that the mainstream sources of big data as mentioned above are sources of structured and semi-structured data on healthcare but the greatest source of data, which is primarily unstructured data, is the internet. People search for health solutions online following which many people self-medicate by purchasing drugs over the counter and online. There are a fewer number of instances in which people visit healthcare facilities than there is those who seek information online and go forth to self-medicate.
Introduction
Big data in healthcare plays an important role in the understanding of the state of public. Big data from the mainstream sources as mentioned by Kumar and Singh (2019) applies in making highly structured decisions on critical services sought by region, payer information, insurance information, and other data that pertains to information shared by the public to the mainstream healthcare providers. However, there is a lot bigger source of unstructured big data on the public and that is the internet.
Kumar and Singh (2019) studied the impact of big data in healthcare, and various tools available in the Hadoop ecosystem for handling it. In the study the researchers conceptualized the mainstream sources of healthcare care data and demonstrated how the Hadoop ecosystem can be used in big data analytics. The findings of the researcher were greatly insightful on the various types of big data available on the various systems and how the data can be used to understand the state of healthcare. However, there was a limitation to the researchers’ work in that it ignored two important aspects of big on health which are the fact that following the mainstream data would be skewed to providing information only on that part of the society that actively seeks healthcare support from facilities. Often women will seek services from healthcare facilities while men only do that when in a critical condition hence the approach ignores a critical segment of the society, data on men’s touch with the healthcare system. The second concern is that focusing on the ordinary sources of healthcare data obscures the online journey that many people before taking the health issues to a healthcare facility. Google and other search engines have become the free doctor that many people consult before making any other decision to seek healthcare either by visiting a healthcare or self-medicating. Such an approach is not just incomplete but also susceptible to errors in the understanding of the state of healthcare in the big data environment.
There are different analytics approaches that can be utilized by businesses in order to get a detailed and better understanding of big data for healthcare. Hadoop is by far one of the most critical analytics systems when considering big data analytics. Its application in the study by Kumar and Singh (2019) is a great indicators of how the system can be applied in gaining insights in big data. Other analytics systems include Weka Data Mining and Tableau. The application of such systems enables the big data experts to dig deep into data. These tools would be used in retesting the data that was used by Kumar and Singh (2019) in the study references herein as the primary study.
This document provides a critical assessment of the article titled Big Data Analytics for Healthcare Industry: Impact, Applications, and Tools. The critical assessment focuses on the strong points and achievements of the article as well as the weak points. The goal is to demonstrate how to better recognize big data for the healthcare sector and hos to improve the analytics and utilization of big data in the sector. To attain the said objectives, the critical assessment provides a critical assessment other research publications as they compare to the primary article as provided above. This allows for an in-depth analysis of the article and its impacts on big data analytics for the healthcare sector.
Strong Points and Achievements
Kumar and Singh (2019) demonstrated the applicability of tools for and important of accumulating, managing, analyzing, and assimilating large volumes of disparate, structured, and unstructured data for the purposes of application towards aiding the process of care delivery and disease exploration. The article demonstrates the variety of sources of big data for the healthcare sector, the nature stored in the different sources, interrelationships in data, and how to pick out both major and minor trends in the data using the Hadoop system and how his has led to the rise in image, signal, and genomics based analytics in the healthcare sector. Lastly, the article played an important role demonstrating the practical applicability of big data analytics in improving care. Each of these points addressed in this research.
The first strength and achievement of the researchers work was in demonstrating how big data is created, accumulated, managed, analyzed, and assimilated in different forms and in different mainstream sources of data. This is particularly on the interactions between a variety of systems such as electronic health records, doctors’ systems, payers’ systems, insurance systems, and public health systems among many other mainstream systems of generating and storing the data. From the demonstration of availability of big data to the demonstration of how the Hadoop ecosystem can be used in making use of big data, the study was a great success and important as a starting point for further research on big data for the healthcare sector.
Kumar and Singh (2019) described the implications of big data including right living, right care, right provide, right value, and right innovation. A feedback loop among all these applications of big data is a critical component of big data analytics. The demonstration of how these aspects work within the Hadoop systems is critical strength in the research.
All the sources of big data as identified in the work of Kumar and Singh (2019) have been continually reflected in other research studies indicating a general consensus on what comprises the critical data for the healthcare sector. Belle, Thiagarajan, Soroushmehr, Navidi, Beard, and Najarian (2015) reflected the applicability of the same sources and categorization of big data in the healthcare sector thereby showcasing the importance of these sources of data. However, Belle et al. (2015) also demonstrated other aspects of big data such as the use of images, signals, and genomics as components of big data. Some of these types of data may be available in sources identified by Kumar and Singh (2019) but others require going beyond the mainstream sources of big data.
The sources of data shown by Kumar and Singh (2019) and also reflected in Belle at al. (2015) have one challenge and that is the lack of new actionable information as captioned by Alharthi (2017). Alharthi (2017) did not intend to overlook the importance of the structured and semi-structured data that is generated from the mainstream sources of big data in healthcare. Instead, the researchers intended to demonstrate the importance of also looking into more unstructured data that comes from patients through the general use search engines. Recognizing the complexities in big data related to healthcare, the researcher pointed out the need to have data that can be used in predictive analytics especially of the health consumption behavior of the people. From an analytical perspective, the issue of attempting to predict people’s health choices based on data generated elsewhere other than the healthcare system, is a critical component of the field of big data for healthcare. This is only possible with the consideration of internet behavior as a source of big data for the healthcare sector and the recognition of what role this data would play in improving health outcomes.
The second strength is the demonstration of how to use different systems for the utilization of big data. The work of Kumar and Singh (2019) focused on the utilization of the Hadoop systems in unravelling trends in big data for the healthcare sector. The ability to provide demonstrations on the interrelationships of different types of big data was major success factor in the research. The important applicability of this observation is that it becomes possible to closely follow the demonstrations made by the authors in testing of actual data. In other words, critical assessment of the work provided by Kumar and Singh (2019) is a demonstration of how applicable different big data systems are in improving the understanding of healthcare as well as decision-making in different aspects of big data.
Big data analytics systems are similar in many different ways. For instance, there are many aspects of Hadoop that are similar to Tableau and Weka. The implications are that one may use the information provided by Kumar and Singh (2019) to experiment and make tests with other systems. This important aspect demonstrates just how open and universal big data can be. With the use of internet activity as the primary source of big data makes it even easier to implement the tests demonstrated in this research.
Archenaa and Anita (2015) indicated that without proper data analytics methods big data become useless. What the researchers communicated to was the need for determination of the mechanisms for big data analytics and the value proposition for the analytics. The implications are that there is the need to consistently follow through on the capabilities performing meaningful real-time analysis. For instance, the analysis of big data should enable the identification of predictions on emergency medical situations before they occur. This would be possible if the big data analytics are able to use prescriptive data as compared to historical data that is the main component of the mainstream data sources as considered and used by Kumar and Singh (2019).
Overall, the research was important in demonstrating how big data analytics can be used in improving care. Big data can be employed at different levels starting with the prompts and finding of services and healthcare solutions to policy and planning aspects by the government. Different users and interested parties in the healthcare system would find value in utilizing big data and the value would be maximized by focusing no only on the traditional sources of structured and unstructured healthcare data but also on the new and emerging sources of healthcare data, particular on data generated through search engines that have now become a main basis of self-medication information as well as virtual Helpance on Medicare. It is possible that understanding the trends in big data on healthcare in the internet would be able to shift the focus of policy in the healthcare sector.
Weak Points
The greatest weak point in the work of Kumar and Singh (2019) is that the researchers did not focus on unstructured healthcare data from sources other than the traditional sources as have been listed elsewhere in this report. This means that the analytics conducted based on the traditional sources is both incomplete and not real-time data. Its predictive power is therefore reduced by these characteristics.
The second weakness in the work of Kumar and Singh (2019) is that it focused on descriptive aspects of big data analytics, ignoring the most important prescriptive analytics of big data. Furthermore, the researchers failed to demonstrate using data how the Hadoop system operates and how the results are derived from the Hadoop system. The implications are that to persons with limited understanding of big data analytics and limited understanding of the analytics software it may be difficult to employ the skills learnt about big data analytics. These principles are the most important about big data analytics.
Conclusion
In concluding, this document provides a critical assessment of the work of Kumar and Singh (2019) on big data analytics. The article demonstrates the main sources of big data for the healthcare industry. However it failed to demonstrate how the data generated on the internet through search engines can be incorporated in big data analytics. With a huge population self-medicating after searching information online, big data analytics for healthcare should focus more on the unstructured data generated on the internet in spite of all its complexities. This would result in a major revolution on the understanding and utilization of big data analytics. The application of Hadoop system was highlighted in the article. However, there are other systems for testing big data including Weka and Tableau. The best use of big data is not in descriptive statistics but in predictive statistics use as it may be used in healthcare emergency situations.

Published by

Essays

View all posts