Comparison of Two MCDA Classification Methods over the Diagnosis of Alzheimer's Disease

摘要

In the present study, we introduce, compare and apply an approach developed upon two Multicriteria Decision Aid (MCDA) classification methods to assist clinicians and researchers in the diagnosis of Alzheimer's disease (AD). Trying to leverage the classifiers' performances, two techniques, one based on ELECTRE IV methodology and the other on a customized genetic algorithm, are employed in order to select the prototypes and calibrate the control parameters automatically. Various experiments were performed over a novel dataset that takes as reference both the functional and cognitive recommendations of the Brazilian Academy of Neurology and a neuropsychological battery of exams made available by the well-known Consortium to Establish a Registry for Alzheimer's Disease (CERAD).