AI-Based Automatic Dementia Detection

This project aims to advance early dementia detection by developing AI-based diagnostic tools that integrate multiple biomarkers, with a primary focus on speech and language analysis. Dementia, particularly Alzheimer’s disease, affects millions worldwide, and early detection remains a significant challenge due to limitations in current diagnostic methods, such as the high cost and limited accessibility of neuroimaging-based screening.

Possible biomarkers should include speech and language biomarkers, neuroimaging (MRI, PET), genetic biomarkers (e.g., APOE4), blood biomarkers (amyloid, tau), cognitive assessment data, and longitudinal clinical datasets.

Key partners include SMARTLABS (Budapest University of Technology and Economics, Hungary) and PSI (KU Leuven, Belgium), with expertise in speech technology, pathological speech processing, ASR, and spoken language understanding. We are searching for consortium partners who can contribute clinical datasets and other biomarker expertise. Expected outcomes include a multilingual AI dementia detection tool, a demonstrator prototype, publications, and a roadmap for clinical integration.