Model-based machine learning of critical brain dynamics

TL;DR

This paper demonstrates that deep neural networks trained on brain models can effectively identify criticality in empirical brain data, revealing the dynamical state during sleep transitions.

Contribution

It introduces a novel approach using neural networks trained on models to detect critical brain dynamics in real fMRI data.

Findings

High correlation between predicted criticality and cluster size exponents

Neural network accurately classifies brain state as subcritical during sleep transition

Conceptual models can be used to infer neural dynamical regimes

Abstract

Criticality can be exactly demonstrated in certain models of brain activity, yet it remains challenging to identify in empirical data. We trained a fully connected deep neural network to learn the phases of an excitable model unfolding on the anatomical connectome of human brain. This network was then applied to brain-wide fMRI data acquired during the descent from wakefulness to deep sleep. We report high correlation between the predicted proximity to the critical point and the exponents of cluster size distributions, indicative of subcritical dynamics. This result demonstrates that conceptual models can be leveraged to identify the dynamical regime of real neural systems.