Editors' ChoiceAlzheimer's Disease

Graph Theory Modeling for Diagnosing Presymptomatic Alzheimer's Disease

See allHide authors and affiliations

Science Translational Medicine  06 Nov 2013:
Vol. 5, Issue 210, pp. 210ec182
DOI: 10.1126/scitranslmed.3007931

Irreversible defects occur in the brain long before the symptoms of Alzheimer's disease (AD) appear. The only signs of impending AD are found in the cerebrospinal fluid, in which there is a decrease in two proteins, β-amyloid and tau. An invasive lumbar puncture is required to assess these markers.

A gentler way of examining the brain would be better. One candidate is resting state functional connectivity magnetic resonance imaging (rs-fcMRI ), which measures temporal correlations of blood oxygen level–dependent signals between brain regions. This method can reveal the degradation of brain networks as clinical symptoms of AD progress. Brier and coauthors have now analyzed functional connectivity MRI data with graph theory mathematical modeling to see whether they can predict future occurrence of AD.

Graph theory assesses the properties of systems that can be modeled as sets of vertices (brain regions) and edges (functional connections) and then generates a summary of network properties with respect to segregation and integration. The authors combine graph theory modeling of rs-fcMRI, cerebrospinal fluid biomarkers, and clinical examination to determine whether graph theory captures brain dynamics in AD and whether the same parameters are observed in preclinical disease. The authors demonstrate that graph theory analysis captures the deterioration in the brain not only in patients with symptomatic AD, but also in people before they show clinical symptoms. In addition, these changes in graph theory measures strongly correlate with advancing AD stage. These results raise hopes that mathematical algorithms could be used to diagnose AD without requiring a stressful lumbar puncture.

M. R. Brier et al., Functional connectivity and graph theory in preclinical Alzheimer's disease, Neurobiol. Aging, published online 18 October 2013 (10.1016/j.neurobiolaging.2013.10.081). [Abstract]

Navigate This Article