MRI-based parameter inference for cerebral perfusion modelling in health and ischaemic stroke

To date, perfusion simulations in the human brain have been limited to a relatively small number of patient-specific cases. Here, a
pipeline relying on magnetic resonance imaging (MRI) is proposed to overcome this limitation. The computational geometry is
adjusted using T1-weighted MRI, whereas the perfusion model parameters are tuned based on arterial spin labeling perfusion MRI. A
cohort of 75 patients is considered to demonstrate that the pipeline is suitable to generate virtual patients with statistically realistic
cerebral blood flow maps. In the future, in silico clinical trials will be conducted using similar virtual cohorts to improve ischaemic
stroke interventions.