EEG Window Length Evaluation for the Detection of Alzheimer’s Disease over Different Brain Regions

• DOI: 10.3390/brainsci9040081
• Publication Date: 2019-04-15T15:15:58Z
• Authors: Katerina D. Tzimo...; Nikolaos Giannakeas; Alexandros T. Tza...; Loukas G. Astrakas; Theodora Afrantou; Panagiotis Ioannidis; Nikolaos Grigoriadis; Pantelis Angelidis; Dimitrios G. Tsal...; Markos G. Tsipouras
• Keywords: Random Forests; moderate; detection; window length; Neurosciences. Biological psychiatry. Neuropsychiatry; 02 engineering and technology; Article; 3. Good health; 03 medical and health sciences; 0302 clinical medicine; classification; 0202 electrical engineering, electronic engineering, information engineering; mild; Alzheimer’s Disease; EEG; 10. No inequality; dementia; RC321-571

Alzheimer's Disease (AD) is a neurogenerative disorder and the most common type of dementia with rapidly increasing world prevalence. In this paper, ability several statistical spectral features to detect AD from electroencephalographic (EEG) recordings evaluated. For purpose, clinical EEG 14 patients (8 mild 6 moderate AD) 10 healthy, age-matched individuals are analyzed. The signals initially segmented in nonoverlapping epochs different lengths ranging 5 s 12 s. Then, group calculated for each rhythm (δ, θ, α, β, γ) extracted, forming feature vector that trained tested Random Forests classifier. Six classification problems addressed, including discrimination whole-brain dynamics separately specific brain regions order highlight any alterations cortical regions. results indicated high accuracy 88.79% 96.78% classification. Also, was higher at posterior central than frontal area right side temporal lobe all problems.