EEG signal analysis for early diagnosis of Alzheimer disease using spectral and wavelet based features

Abstract

Alzheimer disease is one of the most common and fastest growing neurodegenerative diseases in the western countries. Development of different biomarkers tools are key issues for diagnosis of Alzheimer disease and its progression, in early stages. Electroencephalogram (EEG) signal analysis can be well suited for automated diagnosis of Alzheimer’s disease. This paper focuses on spectral and wavelet features for diagnosis of Alzheimer using EEG signals with effective increase in diagnostic accuracy for improvement in diagnosis in early stage. The use of spectral and wavelet based features are proposed in this paper, with effective increase in classification accuracy by use of supervised classifiers giving 94% diagnostic accuracy for early Alzheimer’s diagnosis. As compared to MRI, PET based diagnosis; the EEG based method of diagnosis of Alzheimer disease is much more cost effective. Frequency bump modeling is used to observe the effect of early stage diagnosis of Alzheimer disease using EEG. This research work of Alzheimer disease using EEG signals, explores new tool in the form of bump modeling of EEG signals; for Alzheimer disease diagnosis in early stages.