Alzheimer's disease (AD), the most common neurodegenerative disorder worldwide, is the sixth most common overall cause of death in the USA. It ranks third in health care costs in the USA after heart disease and cancer, and claims an estimated $156 billion USD in direct and indirect costs annually (Wimo et al.,2006). An estimated 13 million elderly will be diagnosed with dementia of the Alzheimer's type (DAT) by the year 2050 in the USA alone (Hebert et al., 2003). As a result of the global aging of the population of all developed countries, the socioeconomic impact of DAT will continue to rise. By the time DAT is clinically diagnosed with current criteria, AD pathology has already spread widely in the brain. To ameliorate the personal and economic impact of DAT, we need to improve on our abilities to diagnose and treat patients as early as possible. This requires improved neuroimaging methods to track pathology in the living brain, and improved computational methods to identify factors that accelerate or resist disease progression.

Mild cognitive impairment (MCI) is an intermediate cognitive state prior to dementia onset, in which people experience some cognitive changes but continue to do well in their daily activities. MCI carries a 4–6-fold increased risk of future diagnosis of dementia. Approximately 10–15% of MCI subjects transition into DAT annually (Petersen, 2007; Petersen et al., 2001), making the MCI state the single most important risk factor of future diagnosis of dementia.