I don't remember offhand if finger-stick blood glucose testing devices are termed significant risk devices or if they are given exemptions. If these devices are given exemptions, then this would likely be a good precedent to follow to execute lab on a chip type devices.
I agree that the lab on a chip analysis represents the most exciting part of the implementation, but I also think that it would be interesting to see what types of associations we can get with usual metrics such as pulse, blood pressure, etc. when the data is more easily obtained.
Further information on significant and non-significant risk devices is available on the FDA website: http://www.fda.gov/MedicalDevices/DeviceRegulationandGuidanc...
http://diabetes.webmd.com/news/20111025/panel-recommends-hep...
I'm for a "no blood" tool at first at least. BGMs are cheap and prevalent.
http://quantifiedself.com/2011/08/nancy-dougherty-on-mindful...
OK, not Non-Intrusive, but quite interesting, I think.
I think the medical tricoder is completely non-invasive on TV, because it's a great way to create the fantasy of the future. As in, the doctor does not even need to touch you!
In the real world labs on chips have come quite far and could go even further. A scratch of skin or tiny droplet of blood could provide an incredible amount of information about your health.
So let's not artificially restrict what the device might be capable of, without good reason.
In fact I think the lab on chip type of analysis is the most interesting part. Measuring pulse, blood pressure, and respiration rate all seem pretty easily achievable.