Quantitative assessment of multiple sclerosis (MS) lesions in longitudinal MRI studies
Radiological Society of North America (RSNA), 97th Scientific Assembly and Annual Meeting, November 27–December 2, 2011, McCormick Place, Chicago
Abstract
An abstract is not available in the lab archive.
This 2011 RSNA conference abstract describes a computer-aided method for measuring multiple sclerosis lesion volumes across MRI examinations. Its retrospective evaluation included five patients; the reported results should be interpreted in light of that small sample.
Summary
The abstract describes an approach to detecting and measuring multiple sclerosis (MS) lesions on longitudinal magnetic resonance imaging (MRI). The method used proton density, pre-contrast T1-weighted, and post-contrast T1-weighted images. User-placed seeds were refined into lesion boundaries and extended to adjacent slices to extract three-dimensional lesions and calculate their volume and diameter.
The retrospective evaluation included five patients with MS, each with at least two follow-up MRI examinations. The abstract reports sensitivity of 98.68% and specificity of 99.88% for delineation, and a 0.41% error in estimated longitudinal lesion-volume change. These results come from a small retrospective sample and do not establish performance in other populations or prospective clinical settings.
Cite this work
U. Bagci, X. Chen, J. K. Udupa, B. Li, S. Messian, B. Turkbey, and O. Aras (2011). Quantitative assessment of multiple sclerosis (MS) lesions in longitudinal MRI studies. Radiological Society of North America (RSNA), 97th Scientific Assembly and Annual Meeting, November 27–December 2, 2011, McCormick Place, Chicago
BibTeX