摘要：Diagnosis of Alzheimer’s disease (AD) in early stage is important to prevent progression of dementia in the aging society. Mild cognitive impairment (MCI) denotes an early stage of AD. In this paper, we aim to diagnose MCI patients during semantic verbal fluency tasks (SVFT) using functional near-infrared spectroscopy (fNIRS). To achieve our objective,t-values and correlations were calculated to find the region of interest(ROI)channels and brain connectivities. From the ROI channels HbO data were averaged over subjects, features (mean, slope, and skewness) were extracted for classification. Extracted features were labelled as two classes and classified via two classifiers, linear discriminant analysis (LDA) and support vector machine (S VM). The classification accuracies were 69.23 % for LDA and 73.07 % for SVM. The results show that there are significant differences of hemodynamic responses (HR) between MCI patients and healthy controls (HC). Therefore, these results suggest a possibility of using fNIRS as a diagnostic tool for MCI patients.
