Early Detection of Dementia Using a Large Language Model-Powered Chatbot
Sidheswar Routray, Durga Samir, Gandhi Pushya, Jyoti Srivastava · 2024
Artificial Intelligence has grown widely in different areas and provided various extensions to healthcare systems like automation techniques for remote monitoring and early disease detection. This work presents a chatbot-driven Large Language Model (LLM) that can identify dementia at an early stage. Dementia is a syndrome characterized by progressive cognitive decline. An AI-powered conversational robot is used to identify linguistic patterns while user conversation suggests the cognitive decline level. This work was tested on 500 participants aged between 60 to 80. The outcomes show that the LLM-based chatbot outperformed conventional screening techniques in detecting early indications of dementia, achieving a sensitivity of 85% and specificity of 88%. After analyzing conversational data, certain linguistic signals were shown to be strongly associated with cognitive impairment. This work demonstrates the potential of conversational agents driven by artificial intelligence for early dementia monitoring and detection that potentially transform elderly care screening procedures.