Tommaso Gili

This paper uses reservoir computing with noisy linear recurrent neural networks to create digital twins of brain activity from resting-state fMRI data, enabling prediction and decomposition of BOLD signals into low-dimensional latent states.

This study uses partial correlation on fMRI data to better align functional connectivity with structural connectivity, showing improved matches with regularization and thresholding.

This study identifies shared and distinct ultra-rare genetic risk factors for epilepsy subtypes using whole-exome sequencing of 20,979 patients and 33,444 controls.

This study identifies mesoscopic functional connectivity alterations in ASD by analyzing contrast subgraphs across brain networks of typically developed individuals and ASD patients.