Transcranial direct current stimulation (tDCS) has proven to be effective in reducing the seizure frequency in patients with epilepsy. This therapy greatly benefits from designing personalized electrode montages (electrode positions and currents) that target the appropriate areas of the brain. In the state-of-the-art pipelines, a specialized physician selects these areas and a biophysical model of the patient’s head is used to find an optimal electrode montage that inhibits them. In the Galvani project, we aim to go a step forward and develop a pipeline that uses personalized Neurotwins for montage optimization. A Neurotwin is a mathematical model of the patient’s brain comprising physical and physiological aspects (in the form of a brain network model, BNM) personalized through data assimilation techniques. In our pipeline, first, a T1-weighted MRI is used to create a volume conductor model of the patient. Then, tractography is applied on a dMRI image to compute the so-called connectome (number of fibers connecting every pair of parcels). Finally, stereotactic-EEG (SEEG) data is used to classify the brain parcels according to their epileptogenicity, which, together with the connectome serves as the basis for the creation of a BNM consisting of coupled neural masses. The parameters of this BNM are adjusted using an evolutionary algorithm to match the functional connectivity associated with the seizure propagation pattern derived from SEEG recordings. Finally, the resulting Neurotwin is used to optimize a tDCS montage with the goal of stopping the seizure spread in the BNM while generating a large inhibitory field in the epileptogenic regions. The personalized neurotwins show a good fit with the empirical data (mean Pearson correlation coefficient of 0.6 between empirical and synthetic functional connectivity) Interestingly, in some of the patients, the models predicted that the seizure spread to areas that were not sampled by the SEEG electrode contacts. Our results show that the optimized montages obtained with the new pipeline, in addition to inducing an inhibitory field in the epileptogenic brain regions, also target brain regions that were not deemed important by the physicians when analyzing all the available data of the patients. This highlights the potential that the use of Neurotwins may have. Specifically, this approach could help identify subject-specific and pathology-specific extended networks in the brain to be targeted by the treatment.