A New dispersion entropy and fuzzy logic system methodology for automated classification of dementia stages using electroencephalograms

• Published in: Clinical neurology and neurosurgery Vol. 201; p. 106446
• Main Authors: Amezquita-Sanchez, Juan P., Mammone, Nadia, Morabito, Francesco C., Adeli, Hojjat
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier B.V 01.02.2021; Elsevier Limited
• Subjects: Accuracy; Age; Alzheimer's disease; Automation; Brain research; Classification; Cognitive ability; Dementia; Dementia disorders; Differential diagnosis; Discrete wavelet transform; EEG; Electroencephalograms; Electroencephalography; Entropy; Fourier transforms; Fuzzy logic; Methods; Mild cognitive impairment; Neural networks; Neurodegenerative diseases; Neurology; Neurosurgery; Pattern recognition; Sensors; Signal processing; Support vector machines; Wavelet transforms; Fuzzy logic; Discrete wavelet transform; Alzheimer’s disease; Electroencephalograms; Mild cognitive impairment
• DOI: 10.1016/j.clineuro.2020.106446
• ISSN: 0303-8467; 1872-6968

•EEG-based methodology for distinguishing among the Alzheimer’s disease, Mild Cognitive Impairment, and healthy subjects.•Adroit integration of discrete wavelet transform, dispersion entropy index, and a fuzzy logic-based classification algorithm.•Effectiveness is evaluated employing a database of measured EEG data from 45 MCI, 45 AD, and 45 healthy subjects.•It differentiates MCI and AD patients from healthy subjects with an accuracy of 86.6–88.9 %, sensitivity of 91 %, and specificity of 87 %. A new EEG-based methodology is presented for differential diagnosis of the Alzheimer’s disease (AD), Mild Cognitive Impairment (MCI), and healthy subjects employing the discrete wavelet transform (DWT), dispersion entropy index (DEI), a recently-proposed nonlinear measurement, and a fuzzy logic-based classification algorithm. The effectiveness and usefulness of the proposed methodology are evaluated by employing a database of measured EEG data acquired from 135 subjects, 45 MCI, 45 AD and 45 healthy subjects. The proposed methodology differentiates MCI and AD patients from HC subjects with an accuracy of 82.6−86.9%, sensitivity of 91 %, and specificity of 87 %.