Virtual brain network modeling in AD: from pathophysiology towards treatment

Abstract

AbstractBackgroundWhile much is known about the pathology and clinical symptoms of AD, their mutual relationship to impaired brain function during the AD disease course is incompletely understood. For deriving a coherent, integrated multi‐scale disease mechanism, computational neurophysiology is a promising candidate. Virtual models of neurons, neuronal circuits and large‐scale networks are being used at different scales and for different purposes. Ranging from biologically plausible neural mass models to more abstract models of complex system behaviour or pathology spreading patterns, their diversity and versatility can help to address issues that are hard to tackle experimentally.MethodIn this presentation I will review several recent developments that illustrate how virtual models can help to translate local neuronal activity to brain‐wide network level data, probe the effects of AD pathology on brain network behaviour, and predict neurophysiological outcomes of different types of treatment. More specifically, I will show how amyloid‐induced hyperexcitability is compatible with large‐scale oscillatory slowing in AD, and how we can simulate counteracting activity‐targeting interventions such as non‐invasive brain stimulation using virtual models.ResultAD‐like adaptations in neuronal excitability in neural mass model parameters led to the combination of high absolute power and large‐scale oscillatory slowing in nearly all tested scenarios. In contrast, the hypoexcitable neural mass scenarios did not produce AD‐like oscillatory changes. Furthermore, simulating various counteracting tDCS‐induced excitability changes led to highly variable ‘virtual treatment’ outcomes, supporting the idea that electrode location choice can play a substantial role in tDCS treatment success, and generating specific predictions to be falsified in patient studies.ConclusionVirtual models of altered brain activity can help to understand AD pathophysiology and facilitate treatment development.