Braydon Dymm, MD
@BraydonDymm
Stroke Neurologist @CAMCNeuroIAM | Telestroke @Duke_Neurology | AI-in-medicine educator 🤖🧠👨🏫 | Follow for threads + memes #MedTwitter | Not med advice.
What if learning about Neurology was as fun as playing your favorite game? In collaboration with master game designer @zach_london, we made that a reality! 🤩 Gorgeous neuroanatomy artwork 🎮 Unique co-op gameplay with real medical stakes Learn more: neurdgames.com/circle-of-will…
How you use your phone can predict psychiatric illness. Great screening tool or creepy invasion of privacy? 🤔
Introducing our latest milestone on conversational diagnostic AI g-AMIE arxiv.org/abs/2507.15743 To harness AI's abundant potential in diagnosis and treatment, we need ways for licensed professionals to oversee and own responsibility for safety-critical decisions. In clinical…
Finally! A real-world, randomized trial showing that LLMs can work together with clinicians to provide better care.
We saw significant relative error reductions: history-taking (-32%), investigations (-10%), diagnostic (-16%), treatment (-13%) for clinicians with vs. without AI. At Penda alone, AI Consult would avert diagnostic errors in 22k visits and treatment errors in 29k visits yearly.
📣 Excited to share our real-world study of an LLM clinical copilot, a collab between @OpenAI and @PendaHealth. Across 39,849 live patient visits, clinicians with AI had a 16% relative reduction in diagnostic errors and a 13% reduction in treatment errors vs. those without. 🧵
We’re always looking for ways to build connection and joy in our program :) excited to bring in Neurdgames for our next team-building session! @slusom @zach_london
My patient, in the middle of a stroke code (mild, non-disabling), offered me this joke: Have you ever diagnosed someone with yellow canary disease? 🐤 No? Okay good. Because it’s untweetable
I love the idea of this project. Trained exclusively on 50 books from history, soon 500-600 books.
a guy created a dataset of 50 books from London 1800-1850 for LLM training. no modern bias. it’s actually super cool to see what can be trained on it!