Inter subject variability analysis in normal ageing and prodromal AD

Abstract

Inter subject variability examines the individual similarity/difference in neural dynamics within a population, both during rest and cognitive tasks. The current study investigates inter subject similarity using Inter Subject Correlation (${\mathrm{ISC}}_{\mathrm{corr}}$), Inter Subject Coherence (${\mathrm{ISC}}_{\mathrm{coh}}$) and Inter Subject Functional Connectivity (ISFC) in prodromal AD and healthy controls. Electroencephalogram (EEG) signals were recorded from 13 prodromal AD and 20 healthy subjects belonging to an ageing population. Inter subject correlation and inter subject coherence were estimated for all the EEG channels for each subject pair. Weighted phase lag index (WPLI) based functional connectivity is used for estimating ISFC. Reduction in values of ${\mathrm{ISC}}_{\mathrm{corr}}\mathrm{and}{\mathrm{ISC}}_{\mathrm{coh}}$ was observed in all the recording conditions for the patient group with respect to the healthy controls, with statistically significant difference being observed only in the resting state. ISFC analysis showed a reduction in inter subject similarity in the prodromal AD patient group in comparison to the healthy controls in both resting and cognitive tasks. Results from the study infers the reduction $\mathrm{of}{\mathrm{ISC}}_{\mathrm{corr}}\mathrm{and}{\mathrm{ISC}}_{\mathrm{coh}}$ and ISFC in the patient group indicating a reduction in intersubject similarity. The reduction in intersubject similarity observed in prodromal AD group may result from the heterogeneity of brain dynamics within the population during the progression of AD. The study further highlights that the reduction in intersubject similarity is more pronounced during resting states compared to cognitive tasks. Future studies shall benefit from examining intersubject similarity within the patient group which could aid in the development of biomarkers for early detection of the disease.