MR image postprocessing for multiple sclerosis research

Abstract

Computerized image analysis is becoming more routine in multiple sclerosis research. This article reviews the common types of task that are performed when producing quantitative measures of disease status, progression, or response to treatment. These tasks encompass uniformity (bias) correction, registration, segmentation, image algebra and fitting, diffusion tensor imaging and tractography, perfusion assessment, and three-dimensional visualization. The aim of these steps is to output reproducible, quantitative assessments of MR imaging scans that can be performed on data generated by the many different scanning sites that may be involved in multicenter studies.