Time-resolved functional PET-MRI fusion: temporally coupled metabolic and BOLD dynamics across task and naturalistic arousal

Motivation: Latest advances in functional PET (fPET)-FDG and PET-MRI have enabled us to map stimulus-driven metabolic and hemodynamic changes simultaneously within a single scan. However, existing PET-MRI studies have focused on the static, time-averaged (de)couplings of fPET-fMRI signals, neglecting the rich information embedded in their temporal dependence. Goal(s): The goal of this study is to propose and validate an analytical framework linking dynamic, concurrent variations in fPET and fMRI signals. Approach: The efficacy of the framework was tested on visual task and naturalistic arousal fPET-fMRI datasets. Results: Our results demonstrated that metabolic changes modeled by concurrent fMRI signals could successfully predict instantaneous fPET-FDG dynamics. Impact: The statistical framework proposed by our study will enable broad functional PET-MRI studies to elucidate the dynamic interplays amongst metabolic and hemodynamic processes that are otherwise obscured in the conventional, time-averaged analysis.