I think this is a situation where when you get a shiny new hammer everything looks like a nail. The article has some interesting ideas but are they the most economically viable?

The patient routing functions described in the article could more easily and cheaply be done using RFID tag technology. Basic data collection could be done by the patient walking across a scale in a doorway, and placing a hand in a (new) device that could take temperature, and blood pressure, as well as blood and sweat samples. This all needs to be connected to a network but it doesn't require more expensive visual recognition technology.

As for recognizing what a doctor is doing by watching I find that more wishful thinking than practical. Sure it would be cool but I don't see it being the most economically feasible given the complexities involved