Yining Hua

🌍🧠💬 #MentalHealth #AI #LLMs @JohnTorousMD

Check out our scoping review on human-centric LLMs in generative mental healthcare tasks. Key findings call for human-centric study designs, guiding evaluation frameworks, standardized methods, transparent models, reproducible experiments, etc. arxiv.org/abs/2408.11288

"Standardizing and Scaffolding Healthcare AI-Chatbot Evaluation". New preprint from our team led by @yininghua. We reviewed 11 frameworks, refining 271 questions into a structured framework encompassing 3 main constructs, 18 2nd-level, and 60 3rd-level: medrxiv.org/content/10.110…

#AI #Healthcare #Evaluation #LLMs @JohnTorousMD @DBatesSafety @lizhou08

🤖🏥 Healthcare AI chatbots and LLMs have received booming popularity, but lack standard evaluation. We present the first hierarchical framework of relevant constructs, compiled via systematic review and expert discussions. 🧑‍⚕️🧑‍💻 Check it out: doi.org/10.1101/2024.0…

@hjian42 @DBatesSafety @lizhou08 @harvardmed @CCCatMIT @medialab Many thanks to @JieHealthAI for the great feedback on experiment designs. This work will be presented at @AMIAinformatics Annual Symposium 2022 in November.

Please check our new JAMIA paper studying public perception about off-label uses of medications for #COVID19 with @NeneKa31793581, @hjian42, Shixu Lin, Jie Yang, Joseph M. Plasek, @DBatesSafety, @lizhou08. #NLProc #PublicHealth #DrugSafety @harvardmed @CCCatMIT @medialab [1/3]