Meeting Banner
Abstract #0715

Multi-Site Evaluations of a TRUST MRI Technique to Measure Brain Oxygenation

Peiying Liu1, Ivan E. Dimitrov, 12, Trevor Andrews2, 3, David E. Crane4, Jacinda K. Dariotis5, John Desmond6, Julie Dumas3, Guillaume Gilbert2, Anand Kumar7, Jean-Maxime Leroux8, Bradley MacIntosh4, 9, Shaolin Yang7, Guanghua Xiao10, Hanzhang Lu1

1Advanced Imaging Research Center, University of Texas Southwestern Medical Center, Dallas, TX, United States; 2Philips Medical Systems, Cleveland, OH, United States; 3University of Vermont College of Medicine, Burlington, VT, United States; 4Heart and Stroke Foundation Centre for Stroke Recovery, Sunnybrook Research Institute, Toronto, Ontario, Canada; 5Department of Population, Family, and Reproductive Health, Johns Hopkins University, Baltimore, MD, United States; 6Department of Neurology, Johns Hopkins University, Baltimore, MD, United States; 7Department of Psychiatry, University of Illinois at Chicago, Chicago, IL, United States; 8Department of Radiology, Hospital of the University of Montreal, Montreal, Quebec, Canada; 9Department of Medical Biophysics, University of Toronto, Toronto, Ontario, Canada; 10Department of Clinical Sciences, University of Texas Southwestern Medical Center, Dallas, TX, United States

Venous blood oxygenation (Yv) is critical for the quantification of CMRO2, a key biomarker for brain tissue viability and functionality. However, non-invasive quantification of Yv has proved to be challenging. Recently, a T2-Relaxation-Under-Spin-Tagging (TRUST) MRI technique was developed to measure global Yv in the superior sagittal sinus. In this work, we evaluated the applicability and reliability of TRUST in a multi-site setting. We compared Yv values (i.e. accuracy) and standard errors of Yv estimation (i.e. precision) across six imaging centers, and showed that the TRUST sequence can be effectively transferred and performed on a standard 3T scanner at remote sites.

Abstract PDF Presentation Video

Keywords

accuracy adding advanced affects although anterior arterial authors averaging bias blood brain broader calibration cerebral clinical college compatible compute conduct conducted confidence consistent control correlation crane custom dataset datasets decreases dependence described duration either error established estimation evaluate examined females final fitting flow foundation function functional gender gilbert gives global growing health healthy heart identify indicating instructed instructions johns knowledge label labeling males manually marker matched measure measured measures measuring medical medicine model multivariate need niche noise noisy observation operator oxygen oxygenation pairwise participating performance plot plotted population position positioning post potential precision preliminary preparation previously principle procedure processed processing psychiatry pulses pure radiology recently recovery regression relatively retest revealed sample scatter sciences several side since sinus site sites slice smoothly source southwestern spatial specifically spin stroke subject subjects subtraction suggested suggesting superior symbols system systems table technical tool toward trust variable variables variations venous versus vessels viability volunteers worse written