Mild Cognitive Impairment is an intermediate condition between non-altered cognition and dementia, and the progression rate of MCI to PD-D over 4 – 12 years is 45-60% (D. Aarsland et al. 2001)(Levy et al. 2002)(Buter et al. 2008). Identifying PD patients with incipient cognitive decline is of major importance for clinicians and researchers.

EEG data for groups of 27 PD patients with MCI and 43 PD patients without cognitive impairment were recorded using a high-resolution recording system. 66 Frequency and 330 Phase Lag Index measures were extracted. Random Forest algorithm was applied to identify important features for classifying PD-MCI from Non-MCI patients. Scores from five cognitive domains and overall cognitive scores were correlated with Phase Lag Index (PLI) and frequency measures.

Phase Lag Index is an effective QEEG feature in identifying PD patients with Mild Cognitive Impairment.