Nicola Toschi

This study investigates brain age differences in youth with conduct disorder compared to controls, finding accelerated brain maturation in those with adolescence-onset CD.

This paper develops a deep learning model to generate synthetic TSPO PET images from structural MRI data, achieving accurate spatial patterns and maintaining reconstruction accuracy after normalization.

The paper validates an MRI framework for detecting axonal damage in rats and humans, showing increased axonal caliber in both pathological and normal-appearing white matter in multiple sclerosis patients.

This paper introduces a deep graph neural network combining GNNs and TCNs to model spatio-temporal dynamics in rs-fMRI data, showing explainable neurobiological insights.

This paper introduces ES-GC, a nonlinear Granger causality estimator using echo-state networks, showing improved detection of causal interactions in biological networks.

This paper develops a deep learning architecture for fusing multiple 3D neuroimaging modalities into informative latent embeddings, demonstrating their ability to cluster subjects into separable phenotypic strata.

This paper develops a parameter-efficient deep learning model to predict conversion from mild cognitive impairment to Alzheimer’s disease using multimodal data, achieving high accuracy and specificity.

This paper develops a machine learning tool to interpret cognitive differences using brain features from neuroimaging data.