Want to test-drive a deep learning model to automatically segment & measure the maximal diameter of an abdominal aortic aneurysms (AAAs)? ⚕️

Test Drive 'A4' on Hugging Face
huggingface.co/spaces/AdritRa…
(be patient takes a few minutes to boot server up!)

built by Adrit Rao in…

Brian Hargreaves (Brian Hargreaves), ISMRM Singapore Program Chair, already deserves huge applause for the blood, sweat & tears he has put into the meeting
A HUGE amount of work &, (sorry Brian), lots more in the next few weeks!
Please send👏🏼 to keep him going!
ismrm.org/24/24program.h…

#LargeLanguageModels can outperform medical experts in clinical text summarization, according to research led by AIMI Associate Director Akshay Chaudhari & published by Nature Medicine. go.nature.com/3xkdqQG

Congrats for the work from Dave Van Veen Akshay Chaudhari #LLM can reduce the documentation burden on clinicians with equivalent or superior quality. #GenAI may help clinicians to allocate more time to patient care rather than administrative tasks.
nature.com/articles/s4159…

In this 'Nature' article, #StanDOM 's Neera Ahuja, William Collins, Nidhi Rohatgi, MD, MS, SFHM, Poonam Hosamani MD & colleagues explore how #LargeLanguageModels could help reduce clinician documentation burden. go.nature.com/3PdZUnV

LLMs have made impressive progress on medical benchmarks like MedQA since 2021. Both open and closed medical LLMs have focused a lot on MedQA performance, but this has a number of problems when we think about deploying LLMs in hospitals. /1

Hi all! We have a position open for a clinician scientist to join us in the neuroradiology section Stanford Radiology. Come invent the future of neuroradiology with us! All levels eligible. Please RT and send to friends! Apply here: facultypositions.stanford.edu/en-us/job/4946… stanfordneuroimaging #aslrocks

January MRM Editor's Pick #3: 'Deuterium metabolic imaging for 3D mapping of glucose metabolism in humans with central nervous system lesions at 3T', by Philip M. Adamson (Philip M Adamson) et al. at StanfordEE Stanford University. onlinelibrary.wiley.com/doi/10.1002/mr…

'All explanation methods are wrong, but some are useful'

Our latest work on Counterfactual Alignment dissects image features used during clf tasks, in an inherently human-understandable manner. We highlight spurious correlations and provide tools to mitigate their impact!