There is an urgent need to improve diagnostic accuracy of Alzheimer's disease (AD) and frontotemporal lobar degeneration- (FTLD)-related disorders in a quantitative manner that does not rely on clinical features. In this study, we will develop an automated procedure for distinguishing individual individual FTLD and AD patients based on advanced MRI and supervised pattern recognition algorithms. Relative to traditional statistical methods, the main advantage of applying supervised machine learning to MRI data is that it allows a characterization at the level of the individual, therefore yielding results with a potentially high level of clinical translation. With the emergence of potential disease-modifying agents for neurodegenerative diseases, such studies are important to establish specificity so that patients may be appropriately entered into etiologically-specific clinical trials.