Related Experiment Video
Updated: Sep 12, 2025
03:14
Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
Published on: December 6, 2024
682
Human in the Loop: Embedding Medical Expert Input in Large Language Models for Clinical Applications
Pedram Golnari1, Katrina Prantzalos1, Dipak Upadhyaya1
1Department of Population and Quantitative Health Sciences, Case Western Reserve University, Cleveland, OH 44106 USA.
Studies in Health Technology and Informatics
|August 8, 2025
Summary
Biomedical ontologies enhance large language models (LLMs) for medical NLP. Integrating expert knowledge, like an epilepsy ontology for Dravet syndrome, improves LLM accuracy in complex medical applications.
Area of Science:
- Biomedical informatics
- Artificial intelligence in medicine
Background:
- Large language models (LLMs) show promise in medical natural language processing (NLP).
- Optimizing LLMs requires effective integration of human medical expertise.
- Current applications often highlight the need for improved accuracy and consistency.
Purpose of the Study:
- To introduce a novel approach using biomedical ontologies to enhance LLM performance in biomedical NLP.
- To demonstrate the effectiveness of a specialized epilepsy ontology for improving LLM accuracy in rare pediatric epilepsy, Dravet syndrome.
- To establish a new method for integrating human expertise into LLMs for high-accuracy medical applications.
Main Methods:
- Development and application of a unique epilepsy ontology.
- Integration of the ontology as a knowledge model for LLMs.
- Focus on biomedical NLP tasks, specifically related to Dravet syndrome.
Main Results:
- Demonstrated significant improvement in LLM performance using the biomedical ontology.
- Showcased enhanced accuracy in processing information related to Dravet syndrome.
- Validated the ontology-based approach for optimizing LLMs in specialized medical domains.
Conclusions:
- Biomedical ontologies offer a powerful mechanism for integrating human expertise into LLMs.
- This approach leads to more accurate and consistent results in medical NLP applications.
- The study paves the way for advanced, expert-informed LLMs in healthcare.
