Machine learning for decision making in healthcare

Rizwan, A., Ozturk, M., Abu Ali, N., Zoha, A. , Abbasi, Q. H. and Imran, M. A. (2020) Machine learning for decision making in healthcare. In: Imran, M. A., Ghannam, R. and Abbasi, Q. H. (eds.) Engineering and Technology for Healthcare. Wiley-IEEE: Hoboken, NJ, pp. 95-115. ISBN 9781119644248 (doi: 10.1002/9781119644316.ch5)

Full text not currently available from Enlighten.

Abstract

This chapter presents a very important use‐case from Rizwan et al. to highlight the role of machine learning in making autonomous decisions for the provision of healthcare services. The scenario presented in this chapter involves use of the data collected for an important bio‐marker, Galvanic Skin Response measured with electrodermal activity sensors, and use of machine learning for auto diagnosis of hydration levels in the human body. The main steps of the bio‐electrical impedance analysis methodology followed in the development of the hydration level detection model are illustrated briefly. Like many real data‐based healthcare studies the main objectives of this chapter are the identification of the appropriate body posture and optimal interval of time for the data collection of bio‐markers and selection of the right combination of features and reliable algorithm for the model development for the auto diagnosis. In the light of the analytical study, the impact of these factors is discussed.

Item Type:Book Sections
Status:Published
Glasgow Author(s) Enlighten ID:Zoha, Dr Ahmed and Rizwan, Ali and Abbasi, Professor Qammer and Imran, Professor Muhammad and Ozturk, Mr Metin
Authors: Rizwan, A., Ozturk, M., Abu Ali, N., Zoha, A., Abbasi, Q. H., and Imran, M. A.
College/School:College of Science and Engineering > School of Engineering > Electronics and Nanoscale Engineering
College of Science and Engineering > School of Engineering > Systems Power and Energy
Publisher:Wiley-IEEE
ISBN:9781119644248
Published Online:27 November 2020

University Staff: Request a correction | Enlighten Editors: Update this record