Meeting Banner
Abstract #2880

Sensitivity of Brain Volumetry: A FreeSurfer-Based Segmentation Study of Brain Data Acquired with the ADNI Protocol Vs. GE BRAVO

Zhaoying Han1, Julian R. Maclaren1, Alexander Brost1, Sjoerd B. Vos1, 2, Christoph Seeger1, 3, Nancy J. Fischbein1, Roland Bammer1

1Center for Quantitative Neuroimaging, Department of Radiology, Stanford University, Palo Alto, CA, United States; 2 Image Sciences Institute, University Medical Center Utrecht, Utrecht, Netherlands; 3Pattern Recognition Lab, Friedrich-Alexander-University Erlangen-Nuremberg, Erlangen, Germany

Biomarkers for neurodegenerative diseases derived from MRI brain structures are promising to translate into clinical practice by comparing a patients brain to those in a database. The aim of this study was to choose the well-defined standard protocol from the ADNI project and compare it to the standard GE BRAVO sequence used in clinical routine at our institution and assess comparability of volumetric measurements. Statistical analysis shows that ADNI and our clinical BRAVO protocol although tightly matched in terms of scan parameters have statistically significant differences in estimated volumes of the lateral ventricles and hippocampus with Freesurfer segmentations.

Abstract PDF Presentation Video

Keywords

accelerated acquisition advanced agreement although alto among analyses arrows assess asset audience available bland blue bottom boundary brain bravo brewer caused caution channel choose choroid classifying clinical coil comparability comparing conducted containing dale dashed database databases defined denotes detection disease diseases early examples exercised existing expert female foundation freely head healthy hemispheres identical included initiative institute institution interleaved involve jack larger lateral limits male matched matrix medical monitoring needs neurology normative orange output particular patient pattern percentage plexus plots popular population potentially powerful practice probably processed project protocol protocols quantitative question radiology recognition related researchers respectively routine scanner sciences segment segmentation segmentations segmented sensitivity since slice slices smaller specified statistical statistically structure structures subject subjects subsequently summarized table target technology terms thalamus tightly translating upper valid ventricle ventricles visual volume volumes volumetric whether white years yellow