Health Informatic Essay

1647 Words4 Pages

Health Informatics has been witnessing a tremendous modernization by leveraging the information technology and networking. Big Data tools offer a platform for organizing huge volume of data generated out of the medical informatics systems. They offer mechanism to store data in a distributed manner and offer parallel processing environment to process the large amount of data. Even though such platforms offer scalable way of managing large volume of data, those tools do not provide mechanism to get value from the large volume of data. Health care data is peculiar in nature because it contains many links to within themselves, such as symptoms, practitioners and medication etc. Processing such data using traditional RDBMS, Big Data tools to get …show more content…

Health care has reached next level of it supremacy, when information technology is applied on it to result in Health informatics. Health Informatics or Health Information Systems is the modern branch that is emerged out of the fusion of computer science, information technology, social science and health care. Advent of electronic records, use of sensors is shaping the health care industry into a potential place where big data can be rightly applied. A lot of issues like efficient data management to effective decision making out of the data is still a big challenge. The advent of health informatics has been creating a tremendous change in the field of health care [1]. It started with modernizing the health care systems with the introduction of Electronic Medical Records (EMR). EMR brought much digitization in the field of health care which resulted in great improvements at the hospitals, patient care methodologies and medication [2]. Since the recent past, sensors play a vital role in the field of health care. Body Sensor Network (BSN) is an amalgamation of wireless wearable smart devices. Such devices are meant to be worn by the chronic patients and the wireless smart devices transmit the symptoms to the remote location based on the configurations [3]. Such technology aims to minimize the presence of individual human care takers for the patients. …show more content…

When healthcare data of size more than TB is fed for processing, overhead is caused in the successive phases. For exam-ple, Shuffle phase generates larger map outputs like many more GBs per node. When it comes to process multiple TB of healthcare data, additional customization like increasing the block size is unavoidable. Also data transferred frequently between shuffle and sort phases adds to the overhead of overall performance. When health care data with a list of patients along with their diseases is fed at Map Reduce sys-tem, each patient will be mapped a disease at the end of the Map phase. There is higher likelihood for the existence of duplicate entries at end of Map phase. Let <K,V> be the set of <disease, patient> pairs, and for the given healthcare data there will be a list of such pairs with same disease as key but with different patients as val-ues. There will be a list of redundant entries in the keys’ list. Yet there is no provision to group all keys with same disease. Existing Map Reduce based system does not provide mechanism to derive such links from within the healthcare data. Health care data is idiosyncratic in nature, because it contains multiple links within itself. Map Reduce process maps each disease from the set {d1,d2,…,dn} to a patient from the set {patient1, patient2, …,

Open Document