Language models / Published · 2025
LLMCARE & clinical language
From language to cognitive signals
Transformer models and synthetic language data for cognitive impairment detection.
01 / Inside the method
How it works.
Select a stage to explore the workflow.
Conceptual schematic · Simplified method overview
Language samples provide the input for studying patterns associated with cognitive impairment.
LLM-generated synthetic language data expand the training material in a setting with limited data.
Transformer representations capture contextual relationships within language for the detection task.
The study evaluates cognitive impairment detection. This schematic summarizes the learning workflow rather than a clinical diagnostic service.
02 / Context & sources
The research.
Learning from clinical language
LLMCARE explores early detection of cognitive impairment using transformer models enhanced with LLM-generated synthetic data. The study examines how language models can help address limited training data in this research setting.
I contributed as a coauthor to this collaborative study of language-based cognitive impairment detection.
Publication
LLMCARE: early detection of cognitive impairment via transformer models enhanced by LLM-generated synthetic data. Frontiers in Artificial Intelligence, 8, 1669896 (2025).