Researchers have developed RAVEN, a new generative pretraining strategy for sequential electronic health record (EHR) data. This model uses a Recurrence-Aware next-Visit EveNt prediction approach and was trained on over one million patient records. RAVEN learns to generate clinical events for future visits based on a patient's history and demonstrates strong zero-shot performance in forecasting diseases, outperforming Transformer models and prompted LLMs. AI
IMPACT This model could improve clinical forecasting and disease prediction by leveraging large-scale EHR data.
RANK_REASON Publication of a research paper detailing a new model and methodology. [lever_c_demoted from research: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
