Fantastic study (and great research methodology) on the abilities of LLMs to perform evidence appraisal. Certainly something a lot of us have been hoping for.

TL;DR: humans outperform LLMs alone, but human + AI performs quite well in some settings.

Exklusiv: #CumEx -Chefermittlerin Anne #Brorhilker verlässt überraschend die Justiz. Ihre Ermittlungen führten zu ersten Urteilen und brachten sogar Olaf #Scholz in Erklärungsnot. Im Interview mit Massimo Bognanni übt sie Kritik. ardmediathek.de/video/Y3JpZDov…

Ein Thread:

🌐 Exploring the frontier of #DigitalBiomarkers in #healthcare ! Learn how devices like smartphones and wearables are revolutionizing #healthdata collection.

Link to Article: bit.ly/4cWkJhM

Tim Woelfle Lars G. Hemkens #healthtech #smartdevices #wearabledevices

A new scientific article showcases a comprehensive analysis of 48 pragmatic trials spanning from 1967- 2022 shedding light on their characteristics, participant demographics, and the assessment of interventions led by the Pragmatic Evidence Lab Lars G. Hemkens
journals.sagepub.com/doi/10.1177/13…

Activity, deep sleep&heart rate can be derived reliably from #smartwatches & contain independent clinically meaningful information about #MS severity, highlighting potential for continuous passive #monitoring in trials&care. doi.org/10.1016/j.msar…
Tim Woelfle RC2NB Cristina Granziera

We are excited to announce our latest research paper which explores the potential of smartwatches as a valuable tool for monitoring the health of individuals with Multiple Sclerosis (MS) published in the journal of Multiple Sclerosis and Related Disorders. sciencedirect.com/science/articl…

One reason AI research has been progressing so fast is not just because of frequent and early publication of preprint on ArXiv and the exchange of open source code, but also because the ML/AI community has largely freed itself from the stranglehold of for-profit & paywalled

Interesting perspective on the Times vs OpenAI lawsuit: 'The Times suit is marinated in complaints about the costly business of journalism. [The scholarly publishing industry] won’t have that argument to make. It’s hard to call out free-riding when it's your own business model.'

Learn peripheral neurology by playing cards! Looking forward to checking it out.

📢 I'm on the faculty job market for 2024! 📢

Grateful for any RTs and pointers!

My vision is to develop biomedical AI systems that exhibit generalist capabilities, see e.g.
nature.com/articles/s4158…

Below a selected overview of my prior & ongoing work:

We analyzed 1746 trials completed 2009-17 at German university medical centers:
55% changed primary outcome
23% w major change

BUT:
14% of changes only visible in registry history-not when looking at last entry only
1% reported change

#PLOSMedicine
dx.plos.org/10.1371/journa…
1/2