Skin lesion AI misses its uncertainty targets
A new benchmark reveals that even calibrated image-recognition models struggle to safely hand off difficult skin cancer cases to human specialists.
“We will develop health and technology together.
In this way, we have broken our tails and determined our route, and we are a comet that shines brightly and we will illuminate each other. “
Yusuf Yesil, M.D.
Founder of Yesil Science
Would you like to get closer to our young team with a wide background from engineering to life science and art ?Join us in our dream of combining medicine, art, and technology. Let's work online together.

It is a community open to everyone who wants to be together with us in our dream of bringing together medicine, technology and art, and to be in active communication with Yesil Science in our ecosystem.
A new benchmark reveals that even calibrated image-recognition models struggle to safely hand off difficult skin cancer cases to human specialists.
Silicon Valley is betting billions that AI agents can run clinical workflows without human intervention, but the messy reality of hospital…
A new machine learning model proves that how we walk, rather than how much we walk, reveals the true severity of…