The term connectomics broadly refers to the study of networks of structurally and functionally connected regions within the central nervous system. Connectivity can be measured and inferred using both neuroimaging and neurophysiological methods such as diffusion tensor imaging (DTI), functional MRI (fMRI), electroencephalography (EEG), magnetoencephalography (MEG), electrocorticography (ECoG), and awake brain mapping [15]. Such studies have yielded novel insights into brain regions traditionally regarded as ‘non-eloquent’ that may actually be essential for brain function, including anatomical regions involved in mentalizing, semantic processing, and language expression [4]. In addition, individual connectomic analyses have expanded our understanding of anatomical–functional correlation, including the identification of motor speech areas outside of the traditional topographic location of Broca’s area, characterization of the medial frontal cognitive control networks, and establishment of the second ventral stream of language processing [16]. Structural connectivity is typically based on tractography (e.g., DTI) and provides an estimation of axonal fiber or tract connections between topographic brain regions [15,17]. Functional connectivity can be assessed using various aforementioned modalities, including fMRI, EEG, and MEG. While structural connectomes provide organic pathways for neuronal activity, functional connectomes may inform indirect connections, multiple inputs, or synaptic changes [18]. Therefore, both structural and functional connectomics are informative and complementary approaches to better resolve our understanding of brain connectivity. Although discordance between modalities may be observed, it is important to consider that each method is governed by specific principles and should not be interpreted as the failure of an individual method [15]. Information gathered from awake neuro-oncological, epilepsy, and DBS surgeries is used to reinforce radiological and neurophysiological models of brain networks. Network analysis can be used to better understand the structural and functional connections linking distinct brain areas in general, and in the context of an intra-axial lesion in particular.

Assembling a connectome using any of the aforementioned approaches utilizes an approximately similar pipeline (Figure 1). The brain is first split into distinct regions through a process known as parcellation. In the case of functional connectomic methodologies, such as fMRI, a blood-oxygen-level-dependent (BOLD) time series is extracted from each parcel and compared with the temporal data from the remaining parcels [15]. In contrast, generating a structural connectome involves applying each parcel as a seed within the tractography iteration and the number of fibers subsequently informs putative connections between regions [15]. Through either approach, connectivity between distinct brain areas is quantified, often illustrated as a connectivity matrix. This can then be further processed using techniques such as graph theory, whereby specific regions (nodes or hubs) and the links between these regions (edges) are studied [19,20].