Dynamic Sensitivity Analysis: Defining Personalised Strategies to Drive Brain State Transitions via Whole-Brain Modelling

Vohryzek, Cabral, Castaldo et al. — Computational and Structural Biotechnology Journal 21, 335–345, 2022.

Why I cared. We can describe brain states well enough. The clinically useful question is how to move someone from one to a healthier one — and that transition is exactly what most methods can’t quantify.

What we did. We introduced “Dynamic Sensitivity Analysis”: fit a whole-brain model to a person’s spatio-temporal dynamics, then apply systematic stimulations in silico and measure how well each one rebalances activity toward a target — say, healthy dynamics.

What we found. Transitions can be quantified by a stimulation’s ability to drive the brain toward the target state, and the framework generalises across stimulation paradigms — a way to search in silico for the best personalised strategy before touching a patient.

What it opened. If the model can rank stimulation strategies, how close is the in-silico optimum to what actually helps a real person — and how personal does the model have to be before that ranking can be trusted?