Meeting Banner
Abstract #1058

A Comparison of Patient-Specific Carotid Arterial Input Functions in Head & Neck DCE Examinations

Marco Borri1, Maria Schmidt1, David Collins1, Matthew Orton1, James d'Arcy1, Ceri Powell2, Dow-Mu Koh1, 3, Angela Riddell3, Kate Newbold2, Martin O. Leach1

1CR-UK and EPSRC Cancer Imaging Centre, Institute of Cancer Research and Royal Marsden Hospital, Sutton, Surrey, United Kingdom; 2Head & Neck Department, The Royal Marsden NHS Foundation Trust; 3Radiology Department, The Royal Marsden NHS Foundation Trust

Reliability of functional parameters depends on accuracy of DCE modelling approaches, and the arterial input function is an important component of the perfusion model. In this work we compare two different approaches to measuring patient-specific AIF in the carotid artery: (i) using a pre-bolus (10% of the dose) prior to DCE acquisition and (ii) using a local enhancing vessel present in the DCE examination. Both approaches are compared in a cohort of patients with histologically proven head and neck carcinoma undergoing radical chemoradiotherapy and enrolled in a longitudinal study, and reproducibility and inter-patient variability are assessed.

Abstract PDF

Keywords

accounting accuracy accurately acquisition acquisitions adjusted affected agreement always approaches arterial artifacts avoiding axial better biomedical blood bolus calculation cancer carcinoma carotid chemotherapy clear cohort common compensation computed concentration confer consistent constant constants construct contrast convert correlation cosine curve curves decreased diagnosis direct directly distributions dose dynamic enable enhanced enrolled error estimation evaluation examination examinations example exception expected extracted find fitting flow foundation fraction free full function functional functions funding gadolinium general good graph head identified indicates induction injection input intra investigated lesion local longitudinal made martin measured measuring median model neck nevertheless nominal paired partial pass patient phys pixel pixels placed plot poor population presence profile progression proved published quality quantify radical radiology reduce reduced reduction reliable repeated represent required residuals resolution respectively royal sampling scaled scanned scanner section seems sense significantly slice software statistically straight superior surrey taken temporal tool treatment trust variability velocity vessel volume washout whole written