Fiona Kolbinger

This piece is the result of a great collaboration with leaders in the field, led by clinician scientist Matthias Carstens, with Dan Hashimoto, @lena-maier-hein.bsky.social, Annika Reinke, Shubha Vasisht, Scott Zhang & Iulia Barbur.

Key numbers:

▶️ 59% of publications on AI-based surgical scene understanding are related to one procedure (laparoscopic cholecystectomy)
▶️ <20% of works perform external or prospective validation
▶️ Important items are often underreported (data selection, performance errors, limitations, code)

Artificial intelligence for surgical scene understanding: a systematic review and reporting quality meta-analysis - npj Digital Medicine npj Digital Medicine - Artificial intelligence for surgical scene understanding: a systematic review and reporting quality meta-analysis

Despite 10+ years of research, the field of surgical video AI remains detached from clinical needs. Our new systematic review identifies data, validation, and reporting as key bottlenecks: www.nature.com/articles/s41...

#surgicalAI #surgicaldatascience #digitalsurgery

Now out in @natcomputsci.nature.com: "Adaptive validation strategies for real-world clinical artificial intelligence"

It was a pleasure working on this piece with @jnkt.bsky.social, excited to finally see it published!

#medicalAI #clinicalAI #digitalhealth #translationalresearch

The economic effects of federal cuts to US science — in 24 graphs Nature - Is US science facing a recession? Growing evidence points to a looming downturn.

It's not just academia that will be affected by the disruptions to research funding

https://go.nature.com/4noApj5