In a talk for the DataScience SG x ODSC Meetup, Eugene Yan addresses the practical side of applying machine learning in healthcare. The presentation is framed around production: what it takes to put ML systems to work rather than stopping at model development.
The emphasis falls on the engineering and operational context that surrounds a model. In healthcare, that context matters because the systems must fit into real clinical workflows and keep working reliably over time.
With a single source, there is no disagreement to compare. The talk is described as an in-depth sharing, making it a useful reference for practitioners who want a grounded view of the gap between prototype and production in healthcare ML.