Yi Tang, Qingyue Li, Yanxuan Wu, Johannes Tröger, Nicklas Linz, Stefan Teipel, Alexandra König
*Poster presented at AAIC 2026
Introduction: Background: Digital speech biomarkers are increasingly recognized as scalable, low-burden tools for early detection, staging, and monitoring of Alzheimer’s disease (AD).
State of the art: Current validation efforts focus largely on Western populations. Emerging approaches such as domain adaptation and multilingual feature representations show promise but lack systematic cross-cultural benchmarking.
Problems: Differences in language structure, communication norms, education, and sociocultural context substantially influence speech-derived features, limiting global applicability and equitable clinical deployment.
