Prediction of Alzheimer's disease in subjects with mild cognitive impairment from the ADNI cohort using patterns of cortical thinning

► Novel approach for subgrouping MCI subjects with respect to “time to conversion” ► New method for automatically selecting highly discriminative cortical features ► Cortical thickness extracted on all ADNI data with very small exclusion rate (2.7%) ► Patterns of cortical thinning in agreement with histological findings (Braak stages) ► Superior accuracy for predicting conversion to Alzheimer's disease depending on the time to conversion (70%-76%)