Recovery of neural dynamics criticality in personalized whole brain models of stroke

Abstract

The critical brain hypothesis states that biological neuronal networks, because of their structural and functional architecture, work near phase transitions for optimal response to internal and external inputs. Criticality thus provides optimal function and behavioral capabilities. We test this hypothesis by examining the influence of brain injury (strokes) on the criticality of neural dynamics estimated at the level of single subjects using whole-brain models. Lesions engendered a sub-critical state that recovered over time in parallel with behavior. Notably, this improvement of criticality depended on the re-modeling of specific white matter connections. In summary, personalized whole-brain dynamical models poised at criticality track neural dynamics, alteration post-stroke, and behavior at the level of single subjects.

Excitatory-Inhibitory Homeostasis and Diaschisis: Tying the Local and Global Scales in the Post-stroke Cortex

“…While it is well established that neuromodulation can be used to improve functional recovery in stroke patients (Blicher et al, 2009;Romero Lauro et al, 2014), we suggest that a deeper understanding of its effects not only in excitability, but also FC, is essential to unlocking the full potential of these rehabilitation procedures. In addition, the use of adaptive systems, such as the Rehabilitation Gaming System (RGS) (Cameirão et al, 2010), has been shown to engage widespread areas across the human cortex (Prochnow et al, 2013) and stimulate cortical reorganization of motor networks (Ballester et al, 2017), being a further potentiator of FC recovery.…”

“…Pharmacological potentiation of AMPA receptors, in turn, has proved detrimental when applied at early stages, while a later application improved recovery, further supporting the delayed, long-term effect of these changes in excitability (Clarkson et al, 2011). Neuromodulation provides another option (Hummel and Cohen, 2006), and noninvasive techniques such as transcranial direct current (tDCS) and magnetic stimulation (TMS), aimed at increasing cortical excitability have been used to improve recovery in stroke patients (Blicher et al, 2009;Romero Lauro et al, 2014). An example of such techniques is theta burst stimulation (TBS) (Huang et al, 2005), based on the application of high frequency bursts of magnetic stimulation at intervals consistent with theta frequencies (i.e., ∼200 ms).…”

“…The therapeutic value of neuromodulation techniques aimed at increasing excitability is already known. Procedures such as tDCS and TMS (Blicher et al, 2009;Romero Lauro et al, 2014) have been applied in the past years, particularly targeted at modulation of motor cortex excitability, and future studies should focus on extending such therapies to different areas in the brain that might experience greater changes in normal levels of incoming excitation, possibly even in a patient-specific manner. Critically, it is important to stress that particular changes in excitability that can aid in the functional recovery of stroke patients might also have side effects.…”

“…Despite the fact that the mathematical formulation of plasticity was, in this case, different from more physiologically grounded rules ( Vogels et al, 2011 ), the working principle is quite similar to synaptic scaling and, particularly, to other implementations in whole-brain modeling ( Hellyer et al, 2016 ; Abeysuriya et al, 2018 ). Later, a similar study was performed in conjunction with behavioral analysis of longitudinal data from stroke patients ( Rocha et al, 2020 ). Homeostatic plasticity was implemented through a normalization of the weights of the structural connectivity matrix after a number of simulation timesteps, thus keeping the levels of incoming excitation to each cortical region relatively constant across time.…”

Criticality and network structure drive emergent oscillations in a stochastic whole-brain model

“…Recent numerical studies highlighted how the topological details of the underlying network are crucial in shaping the type of the transition. Indeed, for densely connected networks the system exhibits a first order discontinuous transition [34,35], with the activity of the network jumping discontinuously between a quiescent and an overactive phase, while for sparse networks the transition disappears [34,36], suggesting that brain dynamics in individuals affected by stroke will not be critical [37]. Nevertheless, an analytical understanding of such critical transition, and thus the disentangling of the role of the connectome structure, criticality and emergent properties is still missing.…”