Quantification of cerebral blood flow using non-invasive image-derived arterial input function with [15O]-water PET/MRI

Motivation: Arterial sampling for cerebral blood flow (CBF) quantification in PET imaging is invasive and error-prone. Image-derived input function (IDIF) could serve as a noninvasive alternative Goal(s): Develop a semi-automated blood-free pipeline using MRI-guided PET to derive an IDIF for CBF estimation Approach: Carotid artery segmentation and partial volume correction were applied to [15O]-water PET images co-registered with MRI to derive IDIF using eight healthy subjects Results: IDIF-based CBF measurements showed consistency in the test-retest scans, demonstrating the potential for IDIF substitute arterial sampling Impact: This MRI-guided blood-free IDIF method has the potential to serve as a noninvasive alternative for CBF quantification in PET, improving patient comfort and reducing procedural risks and complexities in clinical and research settings.