Complexity of functional connectivity networks in mild cognitive impairment subjects during a working memory task

•We use magneto-encephalograms in patients with mild cognitive impairment (MCI).•We investigate complexity of functional connectivity network of MCI patients using two different measures: Graph Index Complexity and Efficiency Complexity.•Efficiency Complexity is superior to Graph Index Complexity and its value at theta band can be used for diagnosis of MCI.

ObjectivesThe objective is to study the changes of brain activity in patients with mild cognitive impairment (MCI). Using magneto-encephalogram (MEG) signals, the authors investigate differences of complexity of functional connectivity network between MCI and normal elderly subjects during a working memory task.MethodsMEGs are obtained from 18 right handed patients with MCI and 19 age-matched elderly participants without cognitive impairment used as the control group. The brain networks' complexities are measured by Graph Index Complexity (Cr) and Efficiency Complexity (Ce).ResultsThe results obtained by both measurements show complexity of functional networks involved in the working memory function in MCI subjects is reduced at alpha and theta bands compared with subjects with control subjects, and at the theta band this reduction is more pronounced in the whole brain and intra left hemisphere.ConclusionsCe would be a better measurement for showing the global differences between normal and MCI brains compared with Cr.SignificanceThe high accuracy of the classification shows Ce at theta band can be used as an index for assessing deficits associated with working memory, a good biomarker for diagnosis of MCI.