Joseph D. Janizek (@joejanizek) 's Twitter Profile
Joseph D. Janizek

@joejanizek

Biomedical AI research (and rads residency) @stanfordradres // Previously: Senior Bioethicist @JiankuiHeLab; MD/PhD @UWCSE, @UW_MSTP

ID: 879027708805632001

linkhttps://jjanizek.github.io/ calendar_today25-06-2017 17:25:12

2,2K Tweet

1,1K Takipçi

2,2K Takip Edilen

rosemarie (@rosemarieho_) 's Twitter Profile Photo

when i was 19, i went to office hours with a prof in comparative historical poetics - the kind of guy obsessed w/ Pindar and dactyls - for advice. he told me to pick up as many languages as i could, because that was the gift of being at a place like chicago. it makes me so sad

yuyin zhou@ICLR'25 (@yuyinzhou_cs) 's Twitter Profile Photo

🚨 Google’s MedGemma & OpenAI’s GPT-4o are impressive, but their openness is limited—either fully closed-source or releasing only weights without data/training code. 🔥 Meet MedVLThinker — a fully open multimodal medical reasoning recipe that matches their performance. Simple.

🚨 Google’s MedGemma & OpenAI’s GPT-4o are impressive, but their openness is limited—either fully closed-source or releasing only weights without data/training code.

🔥 Meet MedVLThinker — a fully open multimodal medical reasoning recipe that matches their performance.
Simple.
Joseph D. Janizek (@joejanizek) 's Twitter Profile Photo

My for you page keeps showing me this guy tweeting this, but I don’t really understand what’s going on. You can literally buy a real time pulse ox over the counter from CVS? I don’t think this is like, some fda ban on pulse oximetry??

My for you page keeps showing me this guy tweeting this, but I don’t really understand what’s going on. You can literally buy a real time pulse ox over the counter from CVS? I don’t think this is like, some fda ban on pulse oximetry??
owl (@owl_posting) 's Twitter Profile Photo

average american bio-ml paper: we have created a very useful tool average european bio-ml paper: we have conclusive proof that the tool is useless average chinese bio-ml paper: we have used the tool to create a first-in-class therapeutic, here is n=58 proof of its efficacy

Joseph D. Janizek (@joejanizek) 's Twitter Profile Photo

Excellent substack post on why the amyloid hypothesis for Alzheimer’s still is likely correct: open.substack.com/pub/astralcode…

Excellent substack post on why the amyloid hypothesis for Alzheimer’s still is likely correct: open.substack.com/pub/astralcode…
Anshul Kundaje (anshulkundaje@bluesky) (@anshulkundaje) 's Twitter Profile Photo

Great write up and pretty solid rationale for the design of Arc's gene expression perturbation prediction challenge. The challenge outreach has been top notch as well. Good model for other challenges to follow.