Deep-Learning-Based Research on Cognitive Impairment and Linguistic Features in Older Adults
In recent years,artificial intelligence technologies,particularly deep learning,have emerged as noteworthy new approaches in gerontolinguistics.Utilizing deep learning models for cognitive screening of the older adults based on linguistic markers has become an important international research topic.This method addresses the high sampling costs and invasiveness associated with traditional techniques based on neuroimaging and fluid biomarkers.This paper,starting from the background of deep learning and neural networks,reviews the application of neural network methods in various stages of cognitive impairment detection in older adults using linguistic markers.It introduces the latest advancements in enhancing model performance,the interpretability of embedding features,and large language models for medicine at home and abroad.Furthermore,it proposes future research directions focusing on native Chinese-speaking older adults.In sum,utilizing deep learning technology to empower cognitive screening is an emerging pathway for linguistics to serve human health,which has broad prospects.
NETL
NSTL
万方数据
