Differential diagnosis of dementia using HUMANN-S based ensembles

Abstract

Dementia is one of the most prevalent diseases associated to aging. The two most common variations of this disease are Alzheimer Dementia (AD) type and Vascular Dementia (VD) type, but there are other many forms (OTD): Lewi Body, Subcortical, Parkinson, Trauma, Infectious dementias, etc. All of these forms can be associated with different patterns of anatomical affectation, different risk factors, multiple diagnostic characteristics and multiple profiles of neuropsychological tests, making the Differential Diagnosis of Dementias (DDD) very complex. In this chapter we propose new automatic diagnostic tools based on a data fusion scheme and neural ensemble approach, concretely we have designed HUMANN-S ensemble systems with missing data processing capability. Their ability have been explored using a battery of cognitive and functional/instrumental scales for DDD, among AD, VD and OTD. We carried out a comparative study between theese methods and a clinical expert, reaching these systems a higher level of performance than the expert. Our proposal is an alternative and effective complementary method to assist the diagnosis of dementia both, in specialized care as well as in primary care centres.