Automatic analysis of emotional response based on non-linear speech modeling oriented to Alzheimer disease diagnosis

Alzheimer's disease (AD) is the most prevalent form of progressive degenerative dementia. Its diagnosis made by analyzing many biomarkers and test but nowadays a definitive confirmation requires a post-mortem examination of the patients' brain tissue. The purpose of this paper is to examine the potential of applying intelligent algorithms to the results obtained from non-invasive analysis methods on suspected patients in order to contribute to the improvement of both early diagnosis of AD and its degree of severity. This work deals with Emotional Response Automatic Analysis (ERAA) based on classical and new speech features: Emotional Temperature (ET) and Higuchi Fractal Dimension (FD). The method has the great advantage of being, in addition to non-invasive, of low cost and without any side effects. This is a pre-clinic studio oriented to validate future diagnosis tests and biomarkers. ERAA showed very satisfactory and promising results for the definition of features oriented to early diagnosis of AD.