Peter S. LaViolette1, Mitchell Daun2, Melissa A. Prah1, Kourosh Jafari-Khouzani3, Pavlina Polaskova3, Elizabeth R. Gerstner4, Steven M. Stufflebeam3, Kathleen M. Schmainda1
1Radiology, Medical College of Wisconsin, Milwaukee, WI, United States; 2Neurology, Medical College of Wisconsin, Milwaukee, WI, United States; 3Radiology, Massachusetts General Hospital, Charlestown, MA, United States; 4Neurology, Massachusetts General Hospital, Boston, MA, United States
A new biomarker of brain tumor vasculature derived from independent component analysis (ICA) applied to DSC MRI data has been recently described. It has been shown to occur preferentially in tumor, and is predictive of response to bevacizumab. This study varies the number of ICA components modeled to determine what number is most appropriate for modeling AVOL. The study was done in 27 patients with two scans days apart prior to therapy. We also compared simultaneously acquired SE and GE data. We find that modeling 4 or 5 components results in highly repeatable GE maps, while SE maps of AVOL are less repeatable.