The Stanford Institute for Human-Centered Artificial Intelligence (HAI) and the Center for Digital Health (CDH) recently awarded the AIDE lab a research grant to leverage large language models (LLMs) and develop an automated framework to generate and evaluate patient-friendly clinical reports.  

The framework first comprises innovating new methods to generate patient-friendly and easily accessible clinical reports, which often contain specialized terminology and are difficult to interpret. Next, the lab plans to establish robust evaluation and monitoring strategies for the LLM-generated clinical reports, helping ensure their safety, quality, and effectiveness. Lastly, the AIDE lab aims to validate the automatically generated reports in the clinic by engaging with all relevant stakeholders, including the patients themselves.  

The HAI Seed Research Grants support ambitious ideas with the goal of obtaining initial results. All submitted proposals are subject to rigorous review by HAI-affiliated faculty and senior fellows, along with an ethics and society review. The research proposal “Empowering Patients with AI: LLM-Based Summarization of Clinical Reports for Patient Experience Improvement”, will begin in February of 2025, with a study period of one year. 

Read more here: https://hai.stanford.edu/seed-grant-recipients-2024