Virtual Event
How to Evaluate if the Model Actually Works
July 22 @ 12:00 pm - 1:00 pm
Virtual Event
Every AI tool arrives with an impressive number attached. The number is usually real but whether it means anything is another matter.
This 60-minute session teaches life-sciences researchers and managers how to evaluate a performance claim for AI models. We open with a model that scored well in its own hospital but fell apart in the next one. You’ll leave with a five-question checklist that exposes a weak claim in a few minutes.
The session ends with a hands-on exercise where we examine two models and decide which to deploy.
Who this is for: Life science professionals, managers, researchers, and anyone in life sciences who is looking to evaluate AI tools.
Hosted by:
