Explaining Prediction Models Using Generative AI in Preventive Healthcare
Multiple studies have already demonstrated the effectiveness of electronic health record data in building Type 2 Diabetes Mellitus screening and prediction models. Through collaboration with family medicine physicians, nurse practitioners, patients, and computer science experts, the team at the University of Maribor aims to bring the T2DM prediction models closer to healthcare experts.
Professor or Parrot: Is Medical Knowledge an Emergent Phenomenon of Large Language Models?
Join the Natural Language Processing Working Group for an energetic discussion of the article Performance of ChatGPT on USMLE: Potential for AI-assisted medical education using large language models by Tiffany Kung et al.
Achieving Equitable Impact with Biomedical NLP: Needs for Translational Research
New advances are constantly made in biomedical and health natural language processing, but very few of these advances translate to measurable impact in medicine. When new AI and NLP methodologies are used in practice, they often magnify social biases or exhibit other undesirable behaviors. These failures of translation and AI-related injustices stem from a common source.
Towards Effective and Efficient Interpretation of Deep Neural Networks: Algorithms and Applications
Dr. Xia (Ben) Hu presents a systematic framework from modeling and application perspectives for generating DNN interpretability, aiming at dealing with main technical challenges in interpretable machine learning, i.e., faithfulness, understandability and the efficiency of interpretability.
Tasks As Needs: Reframing the Paradigm of Clinical Natural Language Processing Research for Real-World Decision Support
In this presentation, Karin Verspoor introduces ideas presented in a recent JAMIA Perspective paper, grounding them in some recent examples of information extraction systems developed with clinical partners.