Measuring Cortical Thickness

Introduction The overall goal of this master's project is to study the effect of aging on the thickness of the cerebral cortex. There are three inherent challenges in this project: Creating a coherent definition of cortical thickness. Implementing that definition numerically in the context of real MRI data. Applying the thickness implementation to a real dataset of an aging population, and interpreting the results. Defining cortical thickness is a surprisingly non-trivial task. The question is moreover a fundamental one: what is it, ideally, that one wants to measure? I will try to argue that previous approaches of drawing a straight line from the surface to the white matter are inadequate, and that thickness is not a property of just the surface, but instead should be definable at any point within the cortical mantle, and that a thickness metric ought to try to model the functional unit of the cortex, the cortical column. The numerical/computational implementation of any definition of cortical thickness raises further issues. There are the inherent limitation of MRI to consider, such as limited resolution and partial volume effect. Important also are implementation issues involving concerns such as choice of algorithms and use of system resources. Lastly, the more theoretical and computational issues also have to be applied to a practical goal: the study of how cortical thickness changes with age. We know from previous MRI studies that there is a noticeable decline in the cortical mantle with age, though the cellular mechanism that underlie that decline still need to be elucidated. Animal studies appear to suggest that it is not a