Just a niggle of mine but I was considering that EMC protected devices may not be the best devices for fitting RFID tags internally - fitting an RF transceiver inside effectively what's a conductive "box", over certain parts of the RF spectrum, may not be conducive with transmission of useful RF signal. Also that the idea of fitting any passive or active RF transmitter located in close proximity to the processing or measurement electronics inside a medical device might "put the wind up" a manufacturer's legal, regulatory and QA department until the technology is proven - hence my comment regarding "modification".
Perhaps a standard form-factor (std shape/size/specification) data-tag "socket", might be a solution in equipment of the future, i.e. built-in as you say, John. Lots of useful data could be used to track/manage/maintain the device. Before manufacturers will fit these devices I guess there will need to be various issues addressed RE: standards, both in terms of the type and quantity of data stored, transceiver and tag RF characteristics/capabilities, etc. Rigel's idea seems like a promising one to to me.