Prediction of Alzheimer’s Disease in Subjects with Mild Cognitive Impairment Using Structural Patterns of Cortical Thinning*

Abstract

Predicting Alzheimer’s disease (AD) in patients exhibiting early symptoms of cognitive decline may have great influence on treatment and drug discovery. Structural magnetic resonance imaging (MRI) has the potential of revealing early signs of neuro-degeneration in the human brain and may thus aid in predicting and diagnosing AD. Surface-based cortical thickness measurements from T1w MRI have demonstrated high sensitivity to cortical gray matter changes. In this study we investigated the possibility for using patterns of cortical thickness measurements for predicting AD in patients with mild cognitive impairment (MCI). We used a novel technique for identifying cortical regions potentially discriminative for separating subjects with MCI, which progress to AD, from subjects with MCI, which do not progress to AD. Cortical thickness measurements from these selected regions were used in a classifier for testing the ability to predict AD. The classification showed an overall accuracy of 72% for predicting AD conversion in MCI patients 12 months in advance, which is better than recently published results on similar data.