Abstract

This talk will explore various applications of Large Language Models (LLMs) in the medical domain, focusing on three key topics: structured information extraction from free-text data, Alzheimer’s classification, and the integration of Retrieval-Augmented Generation (RAG) with medical information.

The session will cover how LLMs can be utilized to extract structured insights from unstructured clinical documents, improving the accessibility and usability of medical knowledge. Additionally, we will discuss their role in Alzheimer’s classification, evaluating the effectiveness of fine-tuned LLMs in processing multimodal clinical data for early diagnosis. Finally, we will delve into the integration of RAG to enhance medical information retrieval, reducing hallucinations and improving the reliability of AI-generated responses.

Through this discussion, we will highlight the challenges and opportunities associated with deploying LLMs in healthcare, examining how these models can be optimized for real-world medical applications. The talk will conclude with insights into future research directions and potential advancements in AI-driven medical analysis.

(Cambridge) Harnessing Large Language Models for Medical Data Processing: Structured Information Extraction, Disease Classification, and RAG