Mild cognitive impairment (MCI) is considered a transitional state between the cognitive changes of normal aging and the earliest clinical features of Alzheimer's disease (AD). An important goal is to find features that predict which MCI patients will later convert to AD. Identification of such features will be increasingly important when treatments slowing down the progression of AD become available enabling early intervention. Brain imaging might be one possible predictor of conversion to AD. Functional imaging with positron emission tomography (PET) has shown that either normal elderly people carrying apolipoprotein E epsilon4 allele or people with MCI already show reduced cerebral glucose metabolism in those brain areas that are typically affected in AD. Investigations of different neurotransmitter systems might increase specificity and help in the differential diagnosis between dementing disorders. Dopamine transporter imaging to aid in the differential diagnosis between AD and dementia with Lewy bodies seems promising. Amyloid imaging is an example of "pathology specific" imaging that has great potential to enhance early detection of AD processes and to help in differential diagnosis. In the future, multi-tracer imaging or development of agents enabling imaging of other protein aggregations in neurodegenerative diseases could further help in the early and differential diagnostics and evaluation of novel treatments.