linked
Writing in APA Style 7th Edition Example Paper.
Student Name
University Name
Course Name
Instructor Name
Date
Introduction
Analytical of big data has recently become a crucial topic in the IT and field and generally worldwide. The power of big data and analytics has not yet been comprehensively explored, and researchers and scholars are just at the tip of the iceberg concerning big analytical data. Big data is central in several pillars that make it so powerful when it comes to problems solving and decision making. These pillars are mathematics and statistics, computer science, and Information Technology. Big data’s statistical and mathematical aspects are used to access the model’s accuracy, reliability, and generalization to a broader community. Computer science and IT assist in algorithm generation and creation of logical scripts used to wrangle the data to the required specification and form.
Researchers have explored the applicability of big data in the health sector and found a more comprehensive space where big data and analytics can b applied. The concept behind big data analysis is the application of the concept of correlation in statistics to deduce useful and meaningful information from the data. Additionally, the algorithms used in big data are correlated, and the end product of the algorithms results from the correlations existing in the data. One area researchers have been exploring is the application of big data to combat the growth of the cancer virus. Worldwide
, cancer has become a killer disease, second after cardiovascular disease. The main problem with the cancer virus is late diagnosis and treatment starting late. If cancer can be identified early enough, treatment methods can eradicate cancer entirely at the early stage. The problem at hand is, how do we encourage people to go for health check-ups continually? Are the people in the community are of cancer virus? Is this awareness dependent on other crucial factors? What other factors affect cancer awareness? What factors affect cancer treatment among patients?.
All these aspects can be comfortably be generated using big data analytical techniques. The basis of big data is the data. Whenever the information is not available, big data has no use to society. However, data availability is not a problem in a recent word where a lot of data is being generated daily. The estimated data generated per day is around 2.5 quintillion bytes of data in the social media platforms. In order to infer useful information from these data, complex algorithms are needed to pass in the data and obtain the required information. Additionally, powerful data analysis platforms and software are also in need. Python and R software are widely used when it comes to big data analysis since they have a string computational complexity are they can handle large datasets required in big data analysis.
Data exist in several forms, structured data, and unstructured data. Studies have shown only a small section of the data being generated on a daily basis that is structured. Around 80% of the information being generated is highly unstructured. For data analysis to take place, the data being analyzed should always comply with a particular structure assumed by the algorithms and the analytical techniques. Data transformation is therefore required to transform the data into the needed format. Data cleaning is regarded as the most tiresome and time-consuming in data analysis deal with ensuring the data is clean and ready for analysis. Data scientists spent a lot of time in data cleaning since the data obtained usually is dirty and unstructured.
Big Data Overview
The term “big data” came to light in early 1997 and was first used by researchers and analysts in early 2001due to the phenomena increased volume of data being generated. The concept of “3Vs” was the underlying principle behind big data. The 3Vs referred to volume, velocity, and variety. The three aspects have been widely employed by researchers and scholars to define the concept of big data. The idea of volume dealt with the amount of data being generated, such that the data was too huge to be handled by the early data handing techniques. The velocity of the data, on the other hand, dealt with the speed of data generation, sharing processes hand integration, and finally, data variety dealt with the heterogeneity of the data, data types, and data sources.
In some other cases, the scholar introduced the fourth V refereeing to data veracity. Data velocity focuses on the quality of the data uses the quality of evidence derived from the data. The definition of big data was clearly defined and updated by Gartner in 2012, where the meaning of big data is made precise and clear. This definition, big data, is defined as the use of high volume data, high velocity, and information assets of high quality in discovering processes and optimization. Further, the indication that big data means completely different things with respect o the group of people. Another overview of big data is captured in a report handle to the US congress that defined big data as extensive variable data of large volume and high complexity that uses advanced methods and techniques as well as technology to mine information. Generally, big data is a full 360 degree of data, and all pertinent data, ranging from data collection, strange and analysis. Some of the areas big data has been comprehensively used include astronomy, retail sales, search engines, politics, and health.
Data Collection
Several techniques exist concerning the data collection process and data collection methods. The methods entirely depend on the research study and the objectives of the research. After data is collected is usually stored in large databases for future use. The databases keep on growing as more and more data keeps in coming. Social media platforms are the first place where big data can be generated. A survey is a statistical technique used in data collection where the researcher is interested in accessing a population’s views concerning the topic of interest. In the heart of the survey, we have a questionnaire that contains the survey questions. The survey questions in big data represent the data variables of the datasets. Respondents are typically required to respond to the items to their best of knowledge.
An online questionnaire is mostly used whenever broad data is required since it’s easier to perform and less costly than one survey. The data collection process is critical and essential since it determines the data that is collected, and the data determines the analysis results. If a researcher gathers data that is not correct and full of errors, the analysis results from big data analytics will also be misinforming. In the health sector, a lot of data do exist concerning the patient’s condition and history. Normally this data is stored in the hospital databases and is continuously updated depending on the patient’s condition. For instance, whenever you attend a hospital, your biodata is usually taken, and your health history also captures. This data huge since it accounts for all the patients who attended the hospital for a specified time period.
Other useful information can also be taken and be combined with the bio-data in the health records to make the data more vast and comprehensive. Nowadays, health facilities are creating APIS for data collection for their patients; the APIS is then integrated with analytical algorithms where the data is fed, and the output is displayed or sent back to the patients. For instance, for the pregnant mother, monitoring of the fetus is crucial to the mother’s life and the baby’s life. Buy this, constantly monitoring g the health of the fetus is critical and crucial. Big data analytics can be used to monitor the condition of the baby by accessing the condition of the mother, an SMS API integrated with the hospital data and clinicians knowledge can therefore be fed information from the mother, in terms of how the mother is feeling and the feedback is then sent to the mother. If the condition is critical, a recommendation to the hospital is therefore given.
Data streams in medical health can be divided into three classes; this data originates from the available health systems EMRs. This class includes family and personal historical data, medication history, reports generated from the laboratories, and pathology results. These big data analysis objectives are to lay a comprehensive understanding of disease risk factors and their outcomes. By doing so, the treatment cost is significantly reduced, and efficiency in health care service is enhanced. In the field of medicine, large-scale data in the field of molecular and biological studies are referred to as Omics; the core aim in the OMIC analysis is to investigate the disease mechanism and enhance the specialization of medical treatment such as permission medicine.
Data from the health sector can originate from administrative records, biometric data, patients, report data, medical imaging done via the internet, biomarker data, EMRs, clinical registers, and prospective and retrospective cohort studies.
Big Data in HealthCare
The evolution and the application of technology in the healthcare sector have enabled the collection of a huge amount of data related to patients’ and patients’ history and other important information. Big data in health care focused on the data stored in the electronic health care databases in structural complexity and difficult to manipulate and handle using the traditional methods. Additionally, traditional data handling software is also incapable of handling big data in healthcare. Thus the focus of big data analysis is grounded on handling these big volume data in health sectors the best way possible and derive insightful information from the data. Big data’s core aims to use the externally large data analyses computationally and find the existing patterns, association, and trends. Additionally, data visualization is also very powerful in deriving visual plots of phenomena under study. Studies indicate that. The data generated in the healthcare sectors is the Exabyte levels; this is overwhelming due to the data volume and the data’s diversity. These data have to be analyses and manages and thus the need for big data analysis techniques.
The highly growing data will test the data storage capacity in healthcare, and the innovation of Google Glasses will add a huge amount of data to the behavioral and social aspects of big data. RRMs contain a plethora of datasets ranging from genomic data, demographic data, and clinical trial data know for evaluating the health care services being offered in health care. To comfortably manage and utilize this data, the integration of Information technology and big data is needed for data analysis and retrieval, as well as communication of analysis results.
Nowadays, the application of big data and related skills in the healthcare field is regarded as a method of performing reliable data analysis and decision-making processes in health matters. More important, increasing the gap between the cost of treatment and treatment outcome is needed. In developed countries, researchers and health practitioners are working tirelessly to fill this gap and ensure better healthcare services. Analysis of these gaps indicates that the existing gap between treatment outcome and treatment cost results in poor usage of the available medical evidence, poor management of research insights, and poor recording of care experience. All these aspects lead to wastage of resources, wastage of opportunities, and harm to the patients.
The benefit of big data in health has focused on the following aspects; disease prevention, identifying potential disease risk factors, designing and discovering an interventional treatment for change in health behaviors. Further application of data analysis in big data is applied in the following manner, predictive modeling, management of the population, improving medical devices and surveillance, support for clinical decisions and precision, improving quality of healthcare services, research applications.
Processing of Big Data
Distributed Data Analysis
Data analysis techniques such as the conventional data wrangling methods normally do not scale to achieve the data’s required processing conditions. Conventional processing methods employ a two-step analysis approach such that the problem under investigation is divided into several sub-problems with the same characteristics (“map-reduce”). Additionally, the sub-problems solution is combined to create the overall problem output; the process is referred to as the reduce step. A practical application of Map Reduce in healthcare has been the analysis of electrophysiological data obtained from epilepsy research studies. The Map-Reduce analysis technique is often used to analyze complex biological data that requires complex computations such as genome sequencing. Research studies give evidence that big data analysis techniques and methods out ways the conventional data analysis and processing to enhance the increasing datasets.
Predictive Analytics
The background of predictive analytics mathematical statistics and machine learning. Under predictive analytics, data analysis techniques from statistics are combined together and used as a whole. Predictive analytics, therefore, uses historical and real-time data to give a prediction about a future outcome. Different sectors of the economy have employed predictive modeling, and the usefulness is being realized. Predictive analytics supports the development and invention of health programs that have an immense impact on patients’ behavior.
Crowdsourcing
The idea behind crowdsourcing is the employment of a huge number of analysts who work together to collect, clean, and analyze large datasets for a common agenda. The process is enhanced by the internet, where the employed analysts can exchange ideas while processing the data. The problem under analysis is normally presented in the simplest form possible to give room for validation and accuracy assessment.
Big Data Algorithms
The big data algorithms applied in bid data analytical are mostly centered around statistics and machine learning. Statistical techniques provide techniques such as regression, non-linear regression, time series analysis, etc. the machine learning aspects complement statistical techniques by providing other techniques such as clustering, classifying, and visualization. The two fields are almost inseparable since they complement each other to attain full potential. In machine learning algorithms, several techniques applied in big data analysis exist, such as random forest classifiers, clustering techniques such as the KNN clustering, support vector machine, and black-box algorithms such as the neural networks. Machine learning algorithms’ strength lies in the data splitting, where the data is split into training, testing data, and validation data. With this, the data scientists are able to test and validate the model created. In a classification problem, stratified data splitting techniques are applied to ensure that each data split contains an equal proportion of the target variable.
Clustering algorithms deal with the clustering of data that has similar characteristics. The algorithm is provided with unlabeled datasets, and the clustering of the data is performed based on data correlation and association. Apart from clustering algorithms, basic models such as a logistic regression model are also applicable and widely applied in the medical field. In predicting the status of cancer cells (benign and malignant), the logistic regression model has been found to be very insightful in predicting the cancer cells. Such a scenario represents a classification problem where the data is classified into classes based on the provided explanatory variables.
Challenges for Healthcare Big Data
The health care sector, population health, and biomedical research sectors are constantly generating large amounts of data. These data being generated will pose advantages and disadvantages in equal measures. The datasets’ diversity and the complexity of the data required vast and rich in structure databases for data storage. Implementing such databases required resources and skilled labor as well as time. Some of the problems expected to arise from big data include measurement errors, missing data, irrelevant data, error in information coding presented in textual format. This will shift the analysis to mainly focus on the data analysis, results presentation, and interpretation.
Moreover, the fact that patients have different characteristics and analysis results of some parameters and hypotheses cannot be generalized to every patient. Some additional information, therefore, might be required for generalization. The distribution of health care data will also pose a great challenge in big data analytics. Studies have indicated that medical data and related datasets are located in different sectors that are governed by different states and administration. Additionally, medical data is also dependent on other data, such as insurance data, social, economic data. The integration of these data from different areas will call for restructuring of the current data infrastructure to accommodate these data and enhance data communication and sharing.
The expected change of data and data structure will attract additional functions to the already existing systems such as dealing with poor quality data, privacy enhancement, willingness to share data, methodological hitches, data inconsistencies, and legal matters.
Conclusion
Big data analysis has vast applications in healthcare and related fields. A comprehensive understanding of big data analytics and its use in scientific research is needed, especially in this time of increasing medical data and related. Big data and big data analytics will continue to offer a proliferation of datasets and accelerate healthcare innovation. The rise of big data in higher learning institutions and the integration of big data and technology and e-learning will see the healthcare sector improve healthcare services and enhance healthcare services efficiency and reliability.
Comprehensively utilizing big data will help save time, reduce the cost of healthcare, reduce readmission, treatment optimization, and future planning. Although big data saves lives, healthcare and medical fields are still lagging behind in embracing and implementing big data due to issues such as data privacy, security, governance, and establishing standards. Nevertheless, projections indicate that there will be a widespread use and implementation of big data in the future across the healthcare industry in response to several challenges. Finally, the application of big data analytics and technology is inevitable, and the health sectors are expected to be one of the most beneficiaries of big data analytics and innovation in technology.