This study evaluates the use of support vector machines (SVMs) for the automatic classification of MRI scans in Alzheimer's disease (AD) and other dementias. The goal was to assess the effectiveness of SVMs in distinguishing AD from normal aging and other forms of dementia, including frontotemporal lobar degeneration (FTLD). The study used linear SVMs to classify grey matter segments from T1-weighted MR scans of pathologically confirmed AD patients and cognitively normal elderly individuals from two different centres with varying scanning equipment. The results showed that 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 separate analyses. SVMs were also able to differentiate between AD and FTLD with 89% accuracy. The study concluded that SVMs can effectively separate AD patients from healthy aging subjects, perform well in differentiating between two forms of dementia, and are robust across different centres. This suggests a potential role for computer-based diagnostic image analysis in clinical practice. The study highlights the potential of SVMs in improving the accuracy of dementia diagnosis using MRI scans.This study evaluates the use of support vector machines (SVMs) for the automatic classification of MRI scans in Alzheimer's disease (AD) and other dementias. The goal was to assess the effectiveness of SVMs in distinguishing AD from normal aging and other forms of dementia, including frontotemporal lobar degeneration (FTLD). The study used linear SVMs to classify grey matter segments from T1-weighted MR scans of pathologically confirmed AD patients and cognitively normal elderly individuals from two different centres with varying scanning equipment. The results showed that 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 separate analyses. SVMs were also able to differentiate between AD and FTLD with 89% accuracy. The study concluded that SVMs can effectively separate AD patients from healthy aging subjects, perform well in differentiating between two forms of dementia, and are robust across different centres. This suggests a potential role for computer-based diagnostic image analysis in clinical practice. The study highlights the potential of SVMs in improving the accuracy of dementia diagnosis using MRI scans.