Abstract:A growing number of studies approach the brain as a complex network, the so-called 'connectome'. Adopting this framework, we examine what types or extent of damage the brain can withstand-referred to as network 'robustness'-and conversely, which kind of distortions can be expected after brain lesions. To this end, we review computational lesion studies and empirical studies investigating network alterations in brain tumour, stroke and traumatic brain injury patients. Common to these three types of focal injury is that there is no unequivocal relationship between the anatomical lesion site and its topological characteristics within the brain network. Furthermore, large-scale network effects of these focal lesions are compared to those of a widely studied multifocal neurodegenerative disorder, Alzheimer's disease, in which central parts of the connectome are preferentially affected. Results indicate that human brain networks are remarkably resilient to different types of lesions, compared to other types of complex networks such as random or scale-free networks. However, lesion effects have been found to depend critically on the topological position of the lesion. In particular, damage to network hub regions-and especially those connecting different subnetworks-was found to cause the largest disturbances in network organization. Regardless of lesion location, evidence from empirical and computational lesion studies shows that lesions cause significant alterations in global network topology. The direction of these changes though remains to be elucidated. Encouragingly, both empirical and modelling studies have indicated that after focal damage, the connectome carries the potential to recover at least to some extent, with normalization of graph metrics being related to improved behavioural and cognitive functioning. To conclude, we highlight possible clinical implications of these findings, point out several methodological limitations that pertain to the study of brain diseases adopting a network approach, and provide suggestions for future research.

Network community structure and resilience to localized damage: application to brain microcirculation

Resilience in Hierarchical Fluid Flow Networks

Towards Quantifying Neurovascular Resilience

Investigating structural and functional aspects of the brain's criticality in stroke

Brain Capillary Networks Across Species: A few Simple Organizational Requirements Are Sufficient to Reproduce Both Structure and Function

On propagation in networks, promising models beyond network diffusion to describe degenerative brain diseases and traumatic brain injuries

Network Localisation of White Matter Damage in Cerebral Small Vessel Disease

Using graph theory to compute Laplace operators arising in a model for blood flow in capillary network

Simulation of Robustness against Lesions of Cortical Networks

Resolving structural variability in network models and the brain

Graph Analysis and Modularity of Brain Functional Connectivity Networks: Searching for the Optimal Threshold

Brain Anomaly Networks Uncover Heterogeneous Functional Reorganization Patterns after Stroke

The Cerebrovascular Side of Plasticity: Microvascular Architecture across Health and Neurodegenerative and Vascular Diseases

Dynamical Signatures of Structural Connectivity Damage to a Model of the Brain Posed at Criticality

Microvascular plasticity in mouse stroke model recovery: Anatomy statistics, dynamics measured by longitudinal in vivo two-photon angiography, network vectorization

A multi-domain model for microcirculation in optic nerve: blood flow and oxygen transport

From time-series to complex networks: Application to the cerebrovascular flow patterns in atrial fibrillation

Stroke-induced slow oscillations in dynamic brain organization

Community structure and modularity in networks of correlated brain activity

Brain networks under attack: robustness properties and the impact of lesions

Criticality, Connectivity, and Neural Disorder: A Multifaceted Approach to Neural Computation