To be diagnostically useful, structural MRI must reliably distinguish Alzheimer's disease (AD) from normal aging in individual scans. Recent advances in statistical learning theory have led to the application of support vector machines to MRI for detection of a variety of disease states. The aims of this study were to assess how successfully support vector machines assigned individual diagnoses and to determine whether data-sets combined from multiple scanners and different centres could be used to obtain effective classification of scans. We used linear support vector machines to classify the grey matter segment of T1-weighted MR scans from pathologically proven AD patients and cognitively normal elderly individuals obtained from two centres with different scanning equipment. Because the clinical diagnosis of mild AD is difficult we also tested the ability of support vector machines to differentiate control scans from patients without post-mortem confirmation. Finally we sought to use these methods to differentiate scans between patients suffering from AD from those with frontotemporal lobar degeneration. Up to 96% of pathologically verified AD patients were correctly classified using whole brain images. Data from different centres were successfully combined achieving comparable results from the separate analyses. Importantly, data from one centre could be used to train a support vector machine to accurately differentiate AD and normal ageing scans obtained from another centre with different subjects and different scanner equipment. Patients with mild, clinically probable AD and age/sex matched controls were correctly separated in 89% of cases which is compatible with published diagnosis rates in the best clinical centres. This method correctly assigned 89% of patients with post-mortem confirmed diagnosis of either AD or frontotemporal lobar degeneration to their respective group. Our study leads to three conclusions: Firstly, support vector machines successfully separate patients with AD from healthy aging subjects. Secondly, they perform well in the differential diagnosis of two different forms of dementia. Thirdly, the method is robust and can be generalized across different centres. This suggests an important role for computer based diagnostic image analysis for clinical practice.

To investigate the effect of age on global and regional brain volumes and rates of atrophy, and to compare directly results based on cross-sectional and longitudinal data. Thirty-nine healthy control subjects (age range, 31-84 years) underwent serial magnetic resonance imaging assessments. Measurements included the whole-brain, temporal lobe, hippocampal, and ventricular volumes at baseline and for repeat scans. We found significant decreases in cross-sectional whole-brain (P<.001), temporal lobe (P<.001), and hippocampal (P =.003) volumes and a significant increase in ventricular volume (P<.001) with increasing age. Cross-sectional and longitudinal estimates of atrophy rates were similar. We also found directional evidence of acceleration in atrophy rates with increasing age in all analyses, with the most marked changes occurring after 70 years of age. This increase in rates after 70 years of age was particularly marked in the ventricles (P<.001) and the hippocampi (P =.01). We found a significant age-associated decrease in global and regional brain volumes. Some evidence indicates that this decline in brain volumes may be due to a nonlinear acceleration in rates of atrophy with increasing age. A better understanding of this process may help to discriminate normal age-related changes from neurodegenerative diseases.