[throwanem]

That's not the half of it. The article makes an offhand mention of analyzing stress by way of cortisol production. In order to determine cortisol levels, you need to sample body fluids. Most commonly the fluid in question is blood, although I gather urine and saliva can be used as well.

In entire fairness, I get the impression that was part of a protocol used to develop data to inform the design of this device rather than something the device itself does, but I think it's nonetheless quite a telling example of the degree of invasiveness which the people behind it appear willing to contemplate.