Can the Crowd Solve Medical Mysteries?

I'm really impressed by CrowdMed and surprised I hadn't heard of it. I developed my own prototype of a similar system, also in 2013 (undiagnosedpatient.com). I was meaning to work further on it this year, but other work got in the way. Now, I feel like I don't have to - someone else has already done what I was intending with a better model and is much much further along. It's nice when others do your work for you (I mean that seriously, because I felt like it was my moral duty to further develop this site)!

It all seems a bit up in the air - the process seems superficially plausible, but ultimately it comes down to whether or not it actually works in making accurate diagnoses. What is needed is a formal trial of the results.

It does seem to me that a system like this has a lot of potential as an alternative method of diagnosis within the medical profession. At present, on the one hand, you have general practitioners who cover all human health, which is an impossibly large area of expertise, and on the other hand heavily siloed specialists, with relatively poor links between the two. A system like this could encourage more of an ecosystem of expertise - in particular, more of a continuum between generalists and specialists, and more variety of interests and focus between different doctors.