
Dereck Paul MD
@dereckwpaul
CEO @GlassHealthHQ | @BrighamWomens @HarvardMed via @UCSFMedicine @JohnsHopkins | Equity & Innovation | Son of Immigrants 🇹🇹🇬🇾🇺🇸
ID: 951925479983890432
http://glass.health 12-01-2018 21:14:55
8,8K Tweet
12,12K Followers
3,3K Following

Honored and humbled to accept a new role as Division Chief for Hematology & Oncology at Georgetown starting next month. Excited for all that the future holds for Georgetown University Georgetown Lombardi with the best colleagues one could ask for.



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…


Today is my first clinic as attending physician at MD Anderson Cancer Center #endcancer #sarcoma


Exciting new feature launching from Glass Health soon


At Glass Health, we're building the ultimate AI assistant for clinicians. You can now tailor AI CDS and documentation to your personal preferences, medical specialty (EM, critical care, neuro, etc), and insitution. Try Glass now at glass.health!

The future of medicine is almost certainly caring for patients collaboratively with agentic AI. We're at the very beginning of figuring out exactly what this will look like in healthcare. At Glass Health, we're bullish on a IDE-like interface with deep EHR integration.


We want Glass Health to operate on the EHR in the way that Cursor/Claude Code operates on a GitHub repo.

In medicine, customization of your AI assistant is incredibly important. Without it, you're likely to spend significant time editing AI-generated documentation drafts before they're usable in your practice/EHR. At Glass Health, we've made it extremely easy to teach our AI your




One of the reasons our AI assistant for clinicians Glass Health is so powerful is its strong and flexible agentic architecture. When you submit a query to Glass, an orchestrator agent analyzes your input. It determines which sub-agents must be called upon to generate a


This study, conducted at Penda Health's clinics in Nairobi, Kenya, found significant reductions in clinical decision errors with LLM-based AI clinical decision support (CDS). But clinicians using the AI CDS spent more time on each patient compared to those in the control group,