Longitudinal designs are widely used in medical studies as a means of observing within-subject changes over time in groups of subjects, thereby aiming to improve sensitivity for detecting disease effects. Paralleling an increased use of such studies in neuroimaging has been the adoption of pattern recognition algorithms for making individualized predictions of disease. However, at present few pattern recognition methods exist to make full use of neuroimaging data that have been collected longitudinally, with most methods relying instead on cross-sectional style analysis. In this thesis we develop a feature construction method that uses longitudinal high dimensional data to improve the predictive performance of pattern recognition algorithms when classifying early neurodegeneration. Our method can be applied to data from a wide range of longitudinal study designs and permits an arbitrary number of time-points per subject. We apply the method to two problems: discriminating subjects with mild cognitive impairment (MCI) from healthy controls and discriminating subjects at risk for Parkinson’s disease from healthy controls. We show substantial improvements in predictive accuracy relative to cross-sectional classifiers for discriminating disease subjects from healthy controls on the basis of structural magnetic resonance (MR) images. In addition, our method allows for the transfer of longitudinal information from one set of subjects to make disease predictions in another set of subjects. The proposed methodology is simple and, as a feature construction technique, flexible with respect to the choice of classifier, imaging modality and image registration algorithm.
|Date of Award
|David Lythgoe (Supervisor) & Andre Marquand (Supervisor)