Abstract

AbstractBackgroundCognitive dysfunction in Alzheimer’s Disease (AD) is caused by dysfunctional brain region interactions. Although functional connectivity alterations in AD have been described in several studies, their associations to tau pathology have not been completely understood. Using whole‐brain network modeling we can mechanistically associate the functional impact of tau on brain function in silico.MethodUsing tau PET and fMRI data from the ADNI3 study, we linked regional tau burdens to fMRI activity using a Hopf dynamical model of brain activity using linear regressions. The Hopf model faithfully recreates oscillatory brain activity and adjust the bifurcation of the regional oscillations to either to a noisy (subcritical) or a more regular (supracritical) regime (Deco et al., 2017). The similarity of simulated and empirical time series was assessed by comparing the resulting (dynamical) functional connectivity differences using root mean squared error and Kolmogorov‐Smirnov statistics (as described in Kobeleva et al., 2021). We compared the resulting bifurcation parameters of the Hopf model between healthy controls with no SUVR above 2, healthy controls with at least one region with a SUVR above 2 (considered as at‐risk subjects with a high tau burden) and patients with mild cognitive impairment (MCI).ResultWe achieved a good fit of the tau‐dependent model of brain activity, ensuring a realistic representation of brain activity in the model. Mean bifurcation parameters differed across the three groups (controls with low and high tau burden, and MCI patients). Specifically, we found increasingly more subcritical bifurcation parameters in controls with high tau burden and MCI patients, quantitatively associating a higher tau burden associated with noisier brain oscillations.ConclusionOur results show that biology‐inspired models of AD contain parameters with clinically relevant information. We observed a shift to a more subcritical regime in subjects with high tau and even more so in MCI, showing that tau is associated with nosier brain oscillations, possibly contributing to impaired information transmission and observed cognitive dysfunction. These results constitute a missing link to the branching process theory of cortical network function in AD and contribute to understanding impaired cortical excitability in AD.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call