Abstract #4705

Reliability and Validity of MRI-Based Automated Volumetry Software Relative to Manual Measurement of Subcortical Structures in HIV-Infected Patients from a Multisite Prospective Study

Jeffrey Dewey¹, George Han², Troy Russell¹, Jared Price¹, Daniel McCaffrey¹, Ekta Sem¹, Joy Chinazom Anyanwu¹, Charles Guttmann¹, Bradford Navia³, Ronald Cohen², David Tate¹

¹Center for Neurological Imaging, Brigham and Women's Hospital, Boston, MA, USA; ²Warren Alpert School of Medicine at Brown University, Providence, RI, USA; ³Tufts New England Medical Center, Boston, MA, USA

While software packages designed to automatically identify subcortical structures can greatly increase the efficiency of volumetric studies, notable differences in the performance of these tools have previously been reported. The purpose of this study was to compare the accuracy and consistency of volumetric data for several clinically relevant subcortical structures obtained using three methods: manual tracing, Freesurfer, and SPM. Freesurfer output was more accurate and consistent relative to manual tracings than that of SPM in the majority of structures. However, the inaccuracy and variability of both tools suggest that researchers supplement their use with visual inspection to ensure reliable data.