ABSTRACT
Cloud computing is a very promising area for researchers as smart and cost-effective healthcare systems are demanded in this pandemic era. This book chapter focuses on a Cloud computing–based smart healthcare system using electroencephalogram (EEG) signals.
The main objective of this work is to prepare a smart healthcare system so as to help separate sleep disorder subjects and normal subjects and alcoholic subjects and normal subjects based on fractal dimension calculation with minimum complexity and less time. The real-time EEG data can be uploaded on the hospital's Cloud. The Cloud-based healthcare system is trained with algorithms for finding various features that are given to the classifier automatically to detect whether the person is suffering from a sleep disorder, alcoholic disorder, or is normal.
This work uses fractal dimensions as a measure of complex EEG signals that detect and interpret transients. The work reported here is based on Higuchi's Fractal Dimension (HFD) and Katz's Fractal Dimension (KFD) methods. Signals are treated in a time domain to find HFD and KFD. Work is carried out on various combinations of three different classes viz: the EEG signals of 21 sleep disorder subjects, 20 alcoholic subjects, and 20 normal subjects. HFD provides a true detection rate of 90% for both the combination of sleep disorder subjects and alcoholic subjects and for the combination of sleep disorder subjects and normal subjects, whereas the true detection rate achieved for the combination of alcoholic and normal subjects is 62%. This work reveals that the fractal dimension algorithm is a strong and effective method for detection of different brain pathologies.
