Brain Function Entropy Analysis during Dual‐Tasking: Application of Functional Near Infrared Spectroscopy to Screen Alzheimer’s Disease

Abstract

AbstractBackgroundDespite a high prevalence, over half of patients with dementia never receive an evaluation, indicating a need for a quick and objective routine test for screening cognitive decline in older adults. Due to neural death and disconnections, Alzheimer’s disease (AD) is associated with a low value of nonlinear complexity of brain function, which can be quantified using multiscale entropy analysis.MethodOur proposed solution is an upper extremity function dual‐task screening method, during which, brain function is measured by functional near infrared spectroscopy (fNIRS). The aim of this study was to investigate association between fNIRS entropy during upper‐extremity dual‐tasking and cognitive status in older adults. The sample population for this data includes 40 older adults with ages ranging from 65‐96 years. Participants were categorized into cognitive normal (CN), mild cognitively impaired (MCI), and AD based on MoCA test results. The fNIRS device took measurements over the frontal (right and left) and parietal (right and left) brain regions. Participants were first instructed to rest for three minutes, during which they were asked to relax and not think about any specific topic. Participants were then asked to flex and extend their right arm consistently for three minutes, while counting backwards by intervals of three.ResultWe used multivariable repeated measures analysis using mixed effects modeling to assess the effect of cognitive status on multiscale entropy, with task condition (rest and dual‐task) as within‐subject parameter and sex as covariate. We observed a significant effect of cognitive status on multiscale entropy of left frontal (p = 0.0196) and right frontal brain regions (p = 0.0448), with 0.198, 0.267, and 0.301 entropy values on average across AD, MCI, and CN, respectively. There was also a significant effect of task on multiscale entropy of right frontal (p = 0.0165), left frontal (p = 0.0209) and left parietal regions (p = 0.0011), with 39.79% higher entropy during dual‐task compared to rest trials.ConclusionThe results of this study indicate potential for the use of fNIRS multiscale entropy analysis as a screening tool for cognitive impairment. More research is necessary to better understand other potential confounding factors as well as validate this approach in a larger sample size.