Peder Larson

@pezlarson.bsky.social

Professor of Radiology and Biomedical Imaging at UCSF Mostly #MRI methodology research, including metabolic imaging, lung imaging and myelin imaging

This is a great analysis, especially I like the assessment of different variables on performance in detecting aggressive prostate cancer from MRI important for ML Researchers to see work like this where real world AUCs ~0.7, showing often fundamental challenges in biomedical problems

Radiology: Artificial Intelligence@radiology-ai.bsky.social · 11mo ago

#ThrowbackThursday#DeepLearning models predicted #prostate cancer aggressiveness on #MRI (Apr 2025) https://doi.org/10.1148/ryai.230555 #AI #MachineLearning #ArtificialIntelligence

Image from article in Radiology: Artificial Intelligence

Happy to walk anyone through this. Our grants are proposed work. We often budget for things like salaries, student tuition, and equipment and consumables (in my field). We work on tight margins - we're often asked to "trim" or modify budgets because program officers try to spread the money.

Maya Sen@mayasen.bsky.social · last yr.

Ok so yeah, this has quickly become the #1 misunderstanding about the canceled grants the grants are not “subsidies” or “entitlements” to Harvard or Princeton or whatever they aren’t going into universities’ endowments they are competitive contracts won by these universities to do research

Two new papers on lung structure and ventilation #MRI with 3D UTE methods, inc patient results My impressions - UTE methods (indirect) not as good as HP Xe (direct, but requires inhaled and hyperpolarized gas) - sequences maturing, processing is critical - great potential for lung dz evaluation

BildBild

I think this #MRI award from @brukercorporation.bsky.social is a really nice way to recognize novel preclinical sequence work! Past winners include hot topics of ZTE/UTE, hyperpolarized, CEST, MRF, deuterium

Magnetic Resonance News@nmr900.bsky.social · last yr.

RT @brukercorporation.bsky.social The finalists for this year's Bruker MRI Award have been chosen! We've narrowed it down to the top 3 finalists. It's your last chance to vote for your favorite! Cast your vote: www.bruker.com/en/landingpa... Voting will remain open until April 18 #NMRchat #MRISky 🧲

Impressive work by @mathieuboudreau.bsky.social et al.! Here is the related NeuroLibre preprint that makes replication easy preprint.neurolibre.org/10.55458/neu...

Stability of qMRI metrics

preprint.neurolibre.org

🌙 Lune Bellec@lune-bellec.bsky.social · last yr.

🧠📊 Just out: a deeply sample structural brain and spine dataset from the cneuromod.ca team, now published in Imaging Neuroscience! led by Mathieu Boudreau and Julien Cohen-Adad. 📰 Paper here: lnkd.in/ewXA3dED 1/🧵

The CNeuroMod longitudinal brain and spine structural protocol