We’re excited to share that QoLEAD PhD candidate Baran Polat has had his abstract accepted for a poster presentation at the upcoming 35th Alzheimer Europe Conference, taking place from 6 to 8 October 2025 in Bologna, Italy.
Baran’s research explores how artificial intelligence (AI) can be used to classify stages of Alzheimer’s disease based on speech characteristics in Dutch-speaking individuals. While speech is increasingly recognized as a promising biomarker for Alzheimer’s, there is still relatively little research focusing on Dutch speakers. This study contributes to a growing body of work exploring how Dutch-specific linguistic features—such as diminutives, grammatical gender, and compound word complexity—might help detect early cognitive decline.
The study has now entered the analysis phase, following the completion of data collection. In total, 30 participants were recruited: 10 with mild Alzheimer’s, 10 with mild cognitive impairment (MCI), and 10 healthy controls. Each participant engaged in open conversations and completed the Cookie Theft Picture Task. These recordings are being analyzed using Natural Language Processing (NLP), Automatic Speech Recognition (ASR), and acoustic tools like PRAAT.
At the conference, Baran will present his findings on differences in speech characteristics between the groups, with a focus on the unique properties of Dutch. He will also explain how these insights will be used to develop AI-based Virtual Agents that better align with individual needs.
We look forward to seeing Baran’s work featured at the conference and congratulate him on this exciting milestone in his PhD journey!
