qEEG and Functional Connectivity as a Translational Bridge Between Humans and Dogs in Epilepsy and Associated Disorders: From Spontaneous Model to Automatic Classification—An Integrative Review

Quantitative electroencephalography (qEEG) converts the raw EEG signal into reproducible numerical descriptors (spectral power, hemispheric symmetry, coherence and signal complexity) and has emerged as a candidate translational biomarker linking human and canine neurology. This integrative review examined the diagnostic, prognostic, pharmacological and translational value of qEEG, with emphasis on functional connectivity, and assessed the comparability of the dog as a natural model of human disease. Seventy-five studies were included, spanning epilepsy, acute brain injury, neurodegeneration, rehabilitation and paroxysmal disorders. In both species, epilepsy was consistently associated with altered spectral power and with reduced or reorganized coherence, and interictal abnormalities were demonstrable even in the absence of visible epileptiform discharges. Dogs reproduced the human patterns closely: phenobarbital induced the spectral redistribution predicted by human pharmaco-EEG data, canine cognitive dysfunction reproduced the slowing and the sleep-architecture changes described in Alzheimer’s disease, and a single machine-learning pipeline classified human and canine recordings with comparable accuracy. Conversely, acute brain injury remains virtually unexplored in the dog, and no canine normative database comparable to the human ones is yet available. The evidence was limited by heterogeneous acquisition protocols, small samples and scarce longitudinal veterinary data. qEEG, and coherence in particular, appears to be a promising cross-species biomarker of network dysfunction and supports the dog as a translational platform, although standardized validation remains necessary.