Barriers to brain volume measurement in the real world

Abstract:
This seminar aims to describe the most relevant technical barriers for the use of brain volume measurements in the clinical daily practice. It will be divided into four distinct sections.
In the first, an overall view of imaging techniques for assessing brain volumes and their changes over time will be provided. At the end of this first section, differences between segmentation and registration-based methods should be clear to the students, together with the pros and cons of the two approaches and to their fields of applicability (e.g. segmentation methods are more suitable for cross-sectional experiments, registration methods for longitudinal experiments).
In the second section, the interpretation of volume measurements as obtained from neuroimaging post processing of magnetic 

resonance images (MRI) will be provided. At the end of this second section, students will have clear that measures of volume and their changes reflect together real volume loss, biological fluctuations and measurement errors. Students should be aware about the importance of scaling factors in cross sectional estimates of volumes and importance of length of follow-up in longitudinal estimates of volumetric changes.
In the third section, the different sources of errors will be listed, paying particular attention to the differences between general errors related to the MRI acquisitions and errors specifically related to the MRI features of brain of MS patients. At the end of this section, students should know what acquisition and post-processing strategies can be applied to minimize these errors (e.g. inhomogeneity correction, lesion filling) and what errors are incorrigible (e.g. movement artifacts).
Finally, the last section will be focused on the pros and cons of the use of large data sets from MRI of healthy controls for assessing the deviation from normality at individual level for MS patient. At the end of this section, students will have clear that  different curve of normality can be obtained from different approaches. Thus, they should see that atrophy measures are not “objective” assessments of volumes or their changes, but depend on the software package used. 
At the end of the teaching course, the students will have clear that atrophy measurements need a long expertise in imaging post-processing and need to be tested with different software and in different laboratories to provide a clear evidence of their utility for individual quantitative monitoring the evolution of MS.