Abstract Healthcare acquired its current influence regarding big data technology due to the fact that the data sources involved in healthcare are well-known for their volume, heterogeneous complexity, and high dynamism. In the context of big data, the success of healthcare applications depends solely on the underlying architecture and utilization of appropriate tools, as evidenced in pioneering research attempts. Big data technology has many areas of application in healthcare, such as predictive modeling and clinical decision support, disease or safety surveillance, public health, and research. Big data analytics in medicine and healthcare covers integration and analysis of large amounts of complex heterogeneous biomedical data and electronic health records data. Since the volume of data is huge, it must be used as an advantage to provide the right intervention at the right time and establish personalized care for the patients. Thus this benefits all components of the healthcare system. This chapter aims to provide a better healthcare framework harnessing the benefits of big data analytics. The huge medical data is processed effectively by using various analytical tools, thereby arriving at deeper insights into the data. The existing systems employ algorithms like artificial neural networks, logistic regression, and fuzzy based algorithms. Some techniques used swarm optimization for classification. The clustering involved usage of self-organizing maps and k-means. This chapter proposes a new framework for healthcare using support vector machines and analyzes the performance for big data in terms of accuracy, sensitivity, error rate, and area under the ROC curve. The proposed system is compared with the conventional system in terms of RMSE and MAPE value and the results prove promising for the newly proposed system.
Paper
Full text
Medical Big Data Mining and Processing in e-Healthcare
Semantic Scholar · Medicine · 2019
Abstract
Abstract Healthcare acquired its current influence regarding big data technology due to the fact that the data sources involved in healthcare are well-known for their volume, heterogeneous complexity, and high dynamism. In the context of big data, the success of healthcare applications depends solely on the underlying architecture and utilization of appropriate tools, as evidenced in pioneering research attempts. Big data technology has many areas of application in healthcare, such as predictive modeling and clinical decision support, disease or safety surveillance, public health, and research. Big data analytics in medicine and healthcare covers integration and analysis of large amounts of complex heterogeneous biomedical data and electronic health records data. Since the volume of data is huge, it must be used as an advantage to provide the right intervention at the right time and establish personalized care for the patients. Thus this benefits all components of the healthcare system. This chapter aims to provide a better healthcare framework harnessing the benefits of big data analytics. The huge medical data is processed effectively by using various analytical tools, thereby arriving at deeper insights into the data. The existing systems employ algorithms like artificial neural networks, logistic regression, and fuzzy based algorithms. Some techniques used swarm optimization for classification. The clustering involved usage of self-organizing maps and k-means. This chapter proposes a new framework for healthcare using support vector machines and analyzes the performance for big data in terms of accuracy, sensitivity, error rate, and area under the ROC curve. The proposed system is compared with the conventional system in terms of RMSE and MAPE value and the results prove promising for the newly proposed system.