AI-Driven Big Healthcare Analytics: Contributions and Challenges

AI-Driven Big Healthcare Analytics: Contributions and Challenges

Faiz Maazouzi (Department of Mathematics and Computer Science, University of Souk Ahras, Algeria) and Hafed Zarzour (LIM Research, Department of Mathematics and Computer Science, University of Souk Ahras, Algeria)
Copyright: © 2021 |Pages: 13
DOI: 10.4018/978-1-7998-4963-6.ch008
OnDemand PDF Download:
No Current Special Offers


With the increased development of technology in healthcare, a huge amount of data is collected from healthcare organizations and stored in distributed medical data centers. In this context, such data quantities, called medical big data, which include different types of digital contents such as text, image, and video, have become an interesting topic tending to change the way we describe, manage, process, analyze, and visualize data in healthcare industry. Artificial intelligence (AI) is one of the sub-fields of computer science enabling us to analyze and solve more complex problems in many areas, including healthcare. AI-driven big healthcare analytics have the potential to predict patients at risk, spread of viruses like SARS-CoV-2, spread of new coronavirus, diseases, and new potential drugs. This chapter presents the AI-driven big healthcare analytics as well as discusses the benefits and the challenges. It is expected that the chapter helps researchers and practitioners to apply AI and big data to improve healthcare.
Chapter Preview

Big Data

The term “Big data” seems to have been used for the first time in 1997 by Michael Cox and David Ellworth (Wang, Kung, & Byrd, 2018). It designates a system bringing together more than a trillion bytes (terabytes or 1012 bytes), or even petabytes (1015 bytes) or more which can include structured data, documents, images and sounds while also integrating specific search and search technologies. processing of these data.

In 2011, the health data stored around the world was estimated at 1017 bytes (Wang, Kung, & Byrd, 2018). We are now talking about 190x1018 bytes (Lajonchere, 2018).

In the last years, many studies have highlighted the utility of applying big data analysis to health informatics. For example, Asokan and Asokan (2015) discussed the importance of big data in improving the performance of the systems approach-based one health in the context of the health informatics science. In another study, Otero, Hersh, and Ganesh (2014) demonstrated how biomedical and health informaticians can work together in analytics and Big Data.

Key Terms in this Chapter

Healthcare Analytics: A emerging technology helping hospitals and healthcare organizations to enhance the prediction of patients' outcomes and treatments as well as the quality of their services.

Deep Learning: A specific part of machine learning models that employs multiple neural layers to solve more complex problems.

Big Data Analytics: A process of extracting useful information and pattern from big data.

Big Data: A large volume of structured, semi-structured, and unstructured data that come from different sources with the difficulties of storage and analysis.

Explainablity: A mechanism that can help users in trusting and understanding the artificial intelligence applications they use.

Machine Learning: A branch of artificial intelligence that is based on data acquisitions and algorithms developments to make decisions without explicit programs.

Artificial Intelligence: A popular sub-field of computer science that aims to make machine intelligent as a human.

Data Heterogeneity: A challenge that is caused during the process of data collection by unknown events.

Complete Chapter List

Search this Book: