This document describes a degree project that explores using a morphable brain model to detect Alzheimer's disease based on deformation fields from brain scans. The project tested the method on low-quality MRI scans and achieved 94% accuracy, 97% sensitivity, and 92% specificity in classifying diseased and healthy brains, which is comparable to other methods using higher quality images. The morphable model approach has the advantage of being robust to intensity errors, making it applicable to lower quality clinical scans where data is more readily available.