Leon Chen

Co-founder @md_ai_ - AI-powered reporting and annotation for radiology. Software/ML/AI engineer, physician, hobbyist electronic musician.

[Leon Chen @transcranial]

@rasbt @karpathy Andy Serkis' LotR is the pinnacle of audiobooks. Been reading and listening at the same time for maximal immersion. Absolute joy.

[Leon Chen @transcranial]

@JanWitowski @Ataraxis_AI @kjgeras Congrats Jan! Knew you were cooking up something exciting.

[Leon Chen @transcranial]

@xtbot That'll be 0.01 BTC to wake up again.

[Leon Chen @transcranial]

@xtbot @EIFY @docmilanfar Here's a fun collection of these: https://t.co/lOHOOE6V3p

[Josh Wolfe @wolfejosh]

1/ For years, a friend (“RD”) and I trade examples we find of
“Nominative Determinism”—

—when someone’s name possibly maybe influenced what they ended up doing in their career (or otherwise and most probably just hilarious coincidence)...

Like...
https://t.co/gASdGZCQkD

[Leon Chen @transcranial]

@nsthorat @lilac_ai @databricks Congrats!

[OpenAI @OpenAI]

Announcing GPT-4, a large multimodal model, with our best-ever results on capabilities and alignment: https://t.co/TwLFssyALF

[Andrew Beam @AndrewLBeam]

New preprint!

Q: How accurate is GPT-3 at predicting Dx and triage advice compared to Harvard docs and to an average person Googling their symptoms?

For Dx, GPT-3 is much more accurate than your average person and almost as accurate as the Harvard docs! For triage, not so much.

[MD.ai @md_ai_]

Great work! Glad we could contribute in a small way to this massive effort.

[Hari Trivedi @HariTrivediMD]

Excited to announce the @hiti_lab's release of the EMory BrEast Imaging Dataset (EMBED) on @AWSOpenData containing 20,000 women and over 500,000 mammographic images. @judywawira @ImonBanerjee6 @EmoryRadiology #breastcancer #AI
https://t.co/kjHssxhdJ6

[Satya Nadella @satyanadella]

ChatGPT is coming soon to the Azure OpenAI Service, which is now generally available, as we help customers apply the world’s most advanced AI models to their own business imperatives. https://t.co/kQwydRWWnZ

[Percy Liang @percyliang]

📣 CRFM announces PubMedGPT, a new 2.7B language model that achieves a new SOTA on the US medical licensing exam. The recipe is simple: a standard Transformer trained from scratch on PubMed (from The Pile) using @mosaicml on the MosaicML Cloud, then fine-tuned for the QA task.

[Leon Chen @transcranial]

Fascinating: glioblastomas form scale-free networks, with small populations of activating periodic tumor cells in hubs.

Target the network hub cells; bring down the network.

[Leon Chen @transcranial]

@xtbot Congrats!

[Leon Chen @transcranial]

@MiraRuido Love these. Evokes feelings of what farcaster portal houses in Hyperion would look like.

[Leon Chen @transcranial]

@alex_teichman Me neither. Probably in the category of domains where a little fine-tuning will go a long way.

[Leon Chen @transcranial]

Pretty impressive ability to interpret clinical shorthand.

[Aashay Sanghvi @aashaysanghvi_]

[new post]

"4 questions on AI"

https://t.co/RA7Csnjgeq

[Leon Chen @transcranial]

Wherein ChatGPT very confidently hallucinates non-existent academic references.
If we could hook up PubMed to ChatGPT and have it produce actual references, this could be incredibly useful.

[Ron Alfa @Ronalfa]

Asked ChatGPT an interesting onc clinical question. To my surprise received a definitive answer but when I asked for the reference: it manufactured a fake JAMA Oncology citation and Phase 3 trial. 😂

[Lena Maier-Hein New: @lena-maier-hein.bsky.social @lena_maierhein]

🥂🥂🥂IT’S OUT!🥂🥂🥂

Pitfalls of image analysis metrics 2.0: https://t.co/WZ3wzDj1P5

More than 50 illustrations compiled by image analysis experts from over 60 institutions. What a team effort!

Initiated by #HelmholtzImaging, @ProjectMONAI and @MICCAI_Society
and… (1/n)

[Leon Chen @transcranial]

@DrLaurenOR @Denniston_Ophth Phenomenal and truly important work!