Natural language processing‐based analysis of connected speech in prodromal Alzheimer’s disease
Connected speech elicited by scene description tasks reveal low accuracy and sensitivity for automated natural language processing-based detection of prodromal AD. Noun descriptions could emerge as a viable alternative, demonstrated by their improved classification performance and link to brain regions relevant for memory and language.
Authors
- Patrick Dupont (ORCID: https://orcid.org/0000-0003-1980-2540)
- Bastiaan Tamm (ORCID: https://orcid.org/0000-0001-6094-4060)
- Ella Eycken
- Helena Balabin (ORCID: https://orcid.org/0000-0002-6392-9306)
- Rik R. Vandenberghe (ORCID: https://orcid.org/0000-0001-6237-2502)
- Marie‐Francine Moens (ORCID: https://orcid.org/0000-0002-3732-9323)
- Jolien Schaeverbeke (ORCID: https://orcid.org/0000-0003-2257-0568)
- Laure Spruyt (ORCID: https://orcid.org/0000-0002-0618-5005)
- Ines Kabouche
Institutions
- KU Leuven (BE)
Publication Details
- Journal
- Alzheimer s & Dementia
- Published
- 2024-12-01
- DOI
- https://doi.org/10.1002/alz.091238
- Primary Topic
- Dementia and Cognitive Impairment Research
- Type
- article
- Field-Weighted Citation Impact
- 0.00