Abstract #2163

Statistical Model for Predicting MS Cortical Lesion Detection Rates Based on Lesion Size & MRI Contrast & Resolution

Cherian Renil Zachariah¹, David Pitt², Katharine Teal Bluestein¹, Bradley Clymer³, Michael Knopp¹, Petra Schmalbrock¹

¹Wright Center of Innovation, Radiology Department, The Ohio State University, Columbus, OH, United States; ²Neurology Department, The Ohio Sate University, Columbus, OH, United States; ³Department of Electrical & Computer Engineering, The Ohio State University, Columbus, OH, United States

In Multiple Sclerosis, cortical lesions assessment is considered a potentially better marker for disease burden and progression than conventional white matter lesion assessment. However, because of their small size and low contrast relative to adjacent normal appearing cortex, cortical lesions are difficult to depict in vivo, and no objective measures verifying their presence exists. In this work, we built a statistical model based on MRI and histology of MS brain specimen. The model uses lesion size and MRI contrast and resolution to predict lesion detection rates.