Skip to Main Content
Alzheimer's disease (AD) is the most common cause of dementia in the elderly and affects approximately 30 million individuals worldwide. With the growth of the older population in developed nations, the prevalence of AD is expected to triple over the next 50 years while its early diagnosis remains being a difficult task. Functional imaging modalities including singlephoton emission computed tomography (SPECT) and positron emission tomography (PET) are often used with the aim of achieving early diagnosis. However, conventional evaluation of SPECT images often relies on manual reorientation, visual reading of tomographic slices and semiquantitative analysis of certain regions of interest (ROIs). These steps are time consuming, subjective and prone to error. This paper shows a computer aided diagnosis (CAD) technique for the early detection of the Alzheimer's disease (AD) based on SPECT image feature selection and a random forest classifier. The dimension of the voxel intensities feature space is reduced by defining normalized mean squared error (NMSE) features over regions of interest (ROI) that are selected by a t-test feature selection with feature correlation weighting. A random forest classifier is then trained based on a carefully prepared SPECT database in order to classify a given unknown patient record. The proposed method yields an up to 96% classification accuracy, thus outperforming recent developed methods for early AD diagnosis.