Diagnosis of Alzheimer Disease Progression Stage from Cross Sectional Cognitive Data by Deep Neural Network

The use of deep learning in diagnostic and modeling the progression of neurodegenerative diseases has had a significant boom in the last years. The model complexity, due to the large quantity and diversity of data necessary for their training, do not make them affordable in conditions where obtaining clinical data and magnetic resonance imaging are expensive and complex. However, under these conditions it is feasible using scores from cognitive functions. These techniques are cheap and do not require the use of sophisticated equipment. In this work we propose a deep learning based model for classification of cognitive vectors collected from each patient taking into account the labels corresponding to the disease stage (normal, mild cognitive impairment, and diseased). Experiments on ADNI cohorts shown that our proposal maintained an average accuracy of 0.968 with a standard deviation of 0.01, which is higher than the obtained by the compared methods. The experiments demonstrated the feasibility of the proposed model.