ALZHEIMER´S DISEASE AND FRONTOTEMPORAL DEMENTIA DIFFERENTIAL AUTOMATIC DIAGNOSIS BASED ON SPECT IMAGES

Alzheimer´s disease (AD) and frontotemporal dementia (FTD) are the most frequent neurodegenerative cognitive disorders, but their differential diagnosis is difficult. The aim of this study is to evaluate an automatic method that returns the probability that a patient suffers from AD or FTD from the analysis of brain SPECT (single photon emission computed tomography) images. A set of 116 descriptors was calculated from the images of 173 patients. Its size was reduced to 8 (using PLS). A model based on LDA (linear discriminant analysis) was subsequently built. Validation was carried out by means of the leave-one-out protocol. 83% of patients was properly diagnosed. Performances favorably compared to those of experts (range is 65-73%). Since images were acquired in different hospitals, the impact of the medical center on the diagnosis of both the classifier and the physicians was investigated