There's no easy calculation. Q/t (heat flow through the kapton film is caused by convection from the surface plus conduction through the thermocouple wires. It depends on many factors, e.g. ambient temperature, wire cross section, air flow.
A reasonable approach is to keep the heat dissipation as low as possible, by attaching a piece of styrofoam as thermal insulation on top of the temperature probe and using smallest available thermocouple gauge.
As for switching noise affecting the thermocouple measurement, that's just a matter of appropriate filtering.