Added value of perfusion MRI in neonatal arterial ischemic stroke for delineation of biomarker and prediction of remote seizure

Motivation: Neonatal arterial ischemic stroke (NAIS) could lead to childhood epilepsy. It is critical to accurately predict remote seizures after NAIS diagnosis by adding understudied brain perfusion measures. Goal(s): To leverage multimodal MRI biomarkers to accurately predict remote seizures after investigating diffusion and perfusion MRI patterns in NAIS patients. Approach: Baseline diffusion and perfusion MRI scans acquired at initial presentation were analyzed to build an ensemble model for remote seizure prediction. Results: Unlike hypoperfusion pattern in adult brain stroke, predominant cases of hyperperfusion were identified in the NAIS population. Integrated optimized multimodal MRI-derived biomarkers effectively predict remote seizures. Impact: This study enhances understanding of NAIS by revealing unique perfusion characteristics and developing a predictive model for remote seizure after NAIS. It supports personalized interventions, potentially improving long-term outcomes and guiding future research on NAIS management and outcomes.