I've certainly done similar things. If you have an instrument,
then you can create a scanner calibration chart on the printer,
create the reference file on the instrument, and then
create the scanner profile in the usual way.
The result is that the scanner is much closer to reading
colorimetric values for that print device. This is only
useful of course, if the resulting scanner profile
is distributed to those without the instrument, but have
the same printer and scanner.
