I stumbled upon this article, sounds very interesting.
"In an effort to give assistance to food allergy sufferers, a team of UCLA researchers has developed a lightweight smartphone attachment called the iTube, which is designed to detect allergens in food samples using the phone's built-in camera".
http://www.redorbit.com/news/health/1112748170/iphone-attachment-finds-food-allergens-quickly-121312/ (http://www.redorbit.com/news/health/1112748170/iphone-attachment-finds-food-allergens-quickly-121312/)
Not very practical.
Seriously-- this is my background. I have a PhD; my degree is in bioanalytical chemistry.
My personal judgment here is that this is a fairly clever gimmick-- but not much more. Why?
Because the sample prep necessary to reduce allergen determination (as opposed to "detection" which is a far, far SIMPLER problem) to a simple colorimetric assay?
Yeah-- not for: a) the inexperienced, or b) real-world, real-time scenarios.
Besides, there's NO reason to make the determination using any smartphone components. If you can reduce things to a colorimetric assay (which MUST be the case or a smartphone camera wouldn't be picking up the color differentiation, YK?) then the human eye is actually MORE useful, since you can use a comparison strip. This is like doing swimming pool, hot tub, or fishtank 'testing.' Oh, sure-- you COULD use a smartphone for that, too. Why not? Well, because there isn't a compelling reason to add what amounts to a Rube Goldberg variation to things, maybe?
This is NOT going to be a way to know "safe" or "not safe." It just isn't a practical method for most people. Sampling is the LARGEST problem in food allergen determination in real food samples, and I seriously question how well this solves that problem in practical terms. My guess, based on my background and what I know of the challenges of sampling complex matrices for trace components which are non-homogeneously distributed? This probably is useful for.... say... making allergen determinations in... um... colorless solutions which are at least 99% water and have minimal particulates/colloidal suspension formation with the addition of solvent.
It's interesting-- but I find it mostly kind of amusing. This is a tool in search of a problem, basically... and you know what they say about problems when your only tool is a hammer, right? ;)