UK develops national EHR generative AI model for pandemic prediction

A new research paper details the development of Foresight-England, a large-scale generative AI model designed to predict medical events using electronic health records (EHRs) specifically within the context of the COVID-19 pandemic. This transformer decoder model, comprising 243 million parameters, was trained on EHR data from approximately 61 million individuals. While the project aimed to assess the pandemic's effects and generalize to future data, access to the data has been paused by NHS England, preventing the release of quantitative results. The paper therefore serves as a methodological template, outlining the strategies for building population-scale EHR foundation models and highlighting associated challenges. AI

IMPACT This research provides a methodological template for building large-scale EHR foundation models, potentially accelerating future AI applications in healthcare and pandemic response.

RANK_REASON The cluster describes a research paper detailing the development of an AI model for medical event prediction using EHRs. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• Christopher Tomlinson
• COVID-19
• Foresight-England
• ICD-10
• NHS England
• OPCS-4
• SNOMED CT
• Transformer Decoder
• XGBoost

• paper
• model release
• product

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →