The Big-Data Revolution in US Health Care - Research Paper Example

Assignment:

Abstract

This study seeks to find out whether the healthcare system gets the most out of big data or not. Big data is inevitable in the current times due to increased population demands. It has challenges which must be addressed through technology in order to realise the importance of the big data. The study seeks to know these challenges and how they have been addressed. To realise the objective, the study carries out a survey research through key informant in the healthcare system for the case of Minnesota. This helps the study to understand the whole situation of database management as far as big data is concerned in the healthcare system. The study finds out that the healthcare system faces some challenges while handling the big data. The challenges range from technical to capacity problems. The study finds out that however, the healthcare system adopts new technological advances in the big data management. The technological methods help in improved data analytics to serve the intended target market. The study finds out technological research helps in improving the handling of the big data to serve the various systems that require it.

Introduction

In past few years there has been noted technological advances in the community. With growth, comes challenges. The health care sector is one of the fields that has had dynamisms of challenges ranging from new disease trends to need for faster response at hospital level.This calls for ways of getting and handling all forms of data. Minnesota, just like any other city that is careful with the health concerns.Based on the fact thatdata is collected from a variety of sources and from the public, Minnesota receives an average information flow of over 1 terabyte daily from the public and other health systems [1].Monitoring and controlling the health issues is attainable through the collection and management (analysing to suit persons of interest) of the data.This huge data tune of terabytes of data per day hence the name Big Data. The data is only useful when it has been analysed to something meaningful. Given the need to acquire, store and analyse this data under its characteristic properties is what poses to be a database management problem that will be pursued by the study.

Problem statement

There is need for a lot of data in the health sector. Big data requires a lot of technical expertise In data handling. It also requires a lot of resources to handle. In order to make the data useful, it must be well presented having been analysed. Based on the characteristics of big data, [1], handling data is a major challenge. The main problem if the healthcare system has been able to handle big data using the current technologies.

Importance of the study

This study is essential in understanding the reception of new advancements in big data handling. Handling big data involves, but not limited to analytics, storing and structuring. From the study one is able to understand possible technologies used in structuring the unstructured data as well the future of big data management in managing the healthcare needs. This importance can be attained when the study addresses some binding questions.

Research questions

What are the challenges facing management of big data in the healthcare system?

Is there any new research technologies adopted by the healthcare system?

What is the nature of databases used in the big data handling?

What is the future of the of application of big data in the healthcare system?

Theoretical Framework

This study is based on such concepts that big data can only be sufficiently stored in databases that are not structured. This makes the data unstructured. However, for easy use of this data, it must be structured for easy use by the end users. Managing the data from the big data requires a combine the two types of databases i.e structured and unstructured databases. These concepts are based on big data theory. It is the ‘ big data integration theory’. This theory is described as the theory of Semantics and Programming Languages, Database Mappings. He theory identifies the 4Vs of big data. The Vs include Volume, velocity, veracity and variety of the data. The characteristics of big data can be described from the dimensions of the 4Vs named[2, 3]. The application of the big data management can by far be applied to quite an extend of fields. This is because the handling is an attribute of technological advancement, which applies across all the fields.

Literature review

Handling of the big data depends on the characteristics of big data.An understanding of the characteristics gives a clear view into the possible challenges of handling big data.

Characteristics of big data

As earlier said, the characteristics of the big data culminate in the 4Vs. The big data is huge in volume, and the rate of coming in and utilization is real-time. The sources are varied eg from mobile phones, other healthcare systems worldwide, and direct hospital data.

As such, one needs to cope with noisy, biased, multisourced and very big data streams. As such, handling the data requires lots of space. The space has so far been attained through cloud computing and data compression computing [ 3,4].Coping with the real-time data creation and use requires a system that can analyse data very fast. It is important to note the other aspects such as volume and variety through which data has to be sorted through [2, 3, 4].

Challenges facing big data

The challenges, once more, are posed by the characteristics. The large volumes and varied sources result in heterogeneity of data.This can be explained from an understanding that big data is highly dimensional. The high dimensionality comes with cumulative noise, spurious correlations and incidental heterogeneity. Heterogeneity simply implies composition from dissimilar parts and thus highly varied [3]

A combination of high dimensionality with massive data from varied sources coming in at different speeds from a variety of technologies resultscomes with problems of algorithmic instability and heavy computational costs . For a clear view, it createsheterogeneity, experimental variations, statistical biases and also requires some development of robust adaptive procedures .Massive data has been associated with amplification of error rates that form part of inferential algorithms [4].

It is important to note that time constraints are a big challenge for big data. For example; very good delayed data is more useless than medium level data retrieved quickly [5]. This is a very important factor to the healthcare system especially when it comes to matters of urgency address.

Since the big data is an unavoidable, the challenges might have solutions, which may come with a few capacity issues.The most common capacity problems include expensive costs of the databases management,which may be unattainable by the healthcare department.The technical staff with sufficient expertise to handle the big data. [2]

Technological

In order to attain the desired heights indatabase management, the technology community has made some stridesin research. In order to keep large data, cloud computing has been a remarkable stride in big data. The technology has it that the cloud expands as data increases. There is need for databases that can handle high-speed data. The databases commonly used for accepting data at high speeds is the Mongo database.

Figure1: An illustration of the Mongo database

Some vendors use increased memory and parallel processing to crunch big data quickly. Others opt to put data in-memory but using grid computing approach.There is a remarkable software stride in the big data handling. One such a software is the Hadoop software which handles data in several ways including, but not limited to analytical roles [2].

Methodology

In order to understand the current situation of the big data management, survey research is the most applying method. Key informant interviews of the database management department in the healthcare system. The study used mixed questionnaire design (structured and semi-structured questionnaire) as shown in the appendix. The questionnaire is designed so as to understand the process of how the big data (structured and unstrucured) is converted to useful piece of information (structured) to the target end-users.

The design is meant to capture the information on the current situation of the database and possible qualitative future of the big data management. In our case, with reference to the healthcare system.

The results of the case study were recorded and analysed as reflects in the next section of this paper.

Findings and analysis

The case study finds out that there has been an almost revolutionary leap in big data. Majority of the respondents agreed that big data collection and storage was an unattainable dream in the recent past. The analysis was just recently unattainable but now, all desired end users can obtain the target information with ease. Collectively the informants agree that there has been a real change in the database systems. The change has involved combination of relational and non-relational databases.To be particular the sql (postgres) and mongo databases. It was found that the healthcare system receives data from variety of technology, unverified sourcesand large volumes per unit time. The data reception and usage is almost real-time. The received data is usually unclassified but the usable data has to beclassified and presented well.The healthcare system is a bit challenged in terms of expert sufficiency and infrastructure to handle big data.

Despite the challenges, the healthcare system has attained mileage in big datathrough the use of available technology. Softwaretechnology has been of great use. It was found that apache Hadoop is the handy tool that the healthcare department.It has made all the data handling processes very attainable. It comes with real-time data analysis and with very high efficiency and reliability.

Figure 2: The hadoop system which helps in an improved efficient and easy way of big data management for the end user.

From the given technological advancements in the big data research, it is apparent that database management is taking a good shape into the future.