AI for Dementia Detection: Open AI Health Diagnostic (OAHD)

Abstract

Alzheimer’s Disease (AD) delivers a heavy burden to many patients and caregivers. This weight is widespread and its toll, carried by many, adds financial and emotional distress to the list of its effects. Diagnosis typically occurs in the later stages of progress, but an earlier diagnosis could delay or, even entirely prevent some of these struggles. This thesis reports on development of an application that could possibly be useful for early-stage detection. Open AI Health Diagnostic application is an artificial intelligence application that uses AI for multi-modal biomarker analysis. In particular, the application extracts emotion features from vocal files to detect presence or absence of AD indicators in vocal audio recordings.