Multimodal virtual brain: when hd-EEG drives simulations of brain dynamics closer to patients’ reality

Motivation: The creation of virtual brain models able to fit individual resting-state fMRI (rs-fMRI) time series is needed, aiming to obtain digital twins of patients useful for clinical translation. Goal(s): In this work we develop a multimodal virtual brain for a group of mild cognitive impairment patients, a new step towards the creation of patients' digital twins. Approach: For each subject hd-EEG data were integrated in the multimodal virtual brain together with structural connectivity to drive brain dynamics simulations. Results: Multimodal virtual brain revealed its superior ability to fit rs-fMRI time series recorded in patients. Impact: With its capacity to perform simulations closer to reality, this new model opens new prospectives in the use of virtual brains as digital representations of patients, a crucial tool for the development of personalized interventions.