Originally Aired - Wednesday, March 13 • 1:00 PM - 2:00 PM Eastern Time (US & Canada)
Create or Log in to My Show Planner to see Videos and Resources.
Create or Login to MyHIMSSPlanner to build your schedule and see more information.
{{chatBodyContent}}
Be sure to come back after the session starts to have access to session resources.
Location: W304A
Type: General Education
Session ID: 137
Title: How Natural Language Processing and Large Language Models Unlock Precision Diagnostics in Healthcare
Description: The vast repository of electronic clinical notes remains an underutilized gold mine for precise diagnostics. Clinical Natural Language Processing (NLP) refers to automated methods to enable computers to understand, interpret and generate meaningful and useful human language. Common clinical NLP tasks include information extraction, entity normalization (e.g., mapping diagnosis to ICD-10 CM or HPO codes), medical question answering, etc. Computational algorithms in NLP have been evolving rapidly, from the early rule-based approaches to machine learning and deep learning algorithms. Recently, the latest large language models (LLMs), such as GPT-4 and Med-PaLM, have significantly advanced state-of-the-art performance on many clinical NLP tasks and achieved comparable performance on many medical reasoning tasks to clinician experts. Real-World Evidence (RWE) is becoming increasingly vital in healthcare decision-making. Clinical NLP plays a critical and growing role in harnessing RWE. This presentation will detail an introduction to NLP and LLMs, showcase real-world use cases of NLP for disease screening, and discuss the broader implications of integrating NLP and AI in healthcare for achieving a new level of diagnostic accuracy and personalized care. Speakers include a university professor, an academic medical center chief of computational sciences, and NLP and RWE practitioners from a digital health company.
Level: Intermediate
Format: 60-Minute Lecture
Learning Objective #1: Identify clinical NLP technology, common tasks, evaluation metrics, and its development process
Learning Objective #2: Discuss recent NLP technology advances (ie, large language models), and its strengths, limitations and implications for healthcare
Learning Objective #3: Recognize an NLP-enabled framework that could support scalable and deep RWE insights generation
Learning Objective #4: Plan NLP technologies can be applied for Human Phenotype Ontology term extraction and the early screening for a pediatric genetic rare disease from clinical notes
