Transitioning AI from the lab to real-life applications
summary
SAS embeds AI into its analytics software to automate the full data lifecycle, claiming to automate full calculation that lets Amsterdam UMC’s computer vision mimic radiologists on CT scans to measure chemotherapy response and trigger surgical or treatment decisions. Work includes model deployment tooling, efforts to generate natural language explainability, and a failed internal face ID use case blocked by opt-in rules. Key limits are messy training data, bias, ethics and the need to democratize model-building.
AI-written summary · about
chapters (5)
- 0:00Introduction
- 0:46Examples of AI in practice
- 1:26How does the human AI interface work
- 2:15Moving AI out of the lab
- 3:22Challenges
related talks
01h 20mSAS an open ecosystem for Artifical Intelligence
23mAI in the Life Science Sector
36mResponsible AI
11mUsing AI to track wildlife patterns
26mFrom Pixel to Patient: Making Medical AI Work in the Real World
35mHow may AI help you? - healthcare
46mExplainable AI for Biomedical Image Processing
27mEnterprise AI at Scale