If you combine the words "crowd" and "outsourcing," you get "crowdsourcing," a concept that is exactly like what it sounds — getting work done (or obtaining funding) from a large group of people usually found online. Today, crowdsourcing is often the domain of rock groups and retailers, those most inclined toward self-promotion and sales, while being unafraid of mass opinion. Yet, what if the wild and noisy "wisdom of crowds" could be harnessed for a higher purpose?

Along comes Jared Heyman, who developed CrowdMed to help people who are desperate for an answer concerning their undiagnosed medical conditions. His idea for this start-up company is based on his sister’s plight. Carly, a once healthy teen, could barely leave her bed and despite an endless number of doctors and exams, totaling a cool $100,000, she and her parents had found no answers. Finally, the Heyman family lucked into a team of medical experts who properly diagnosed her condition as fragile X-associated primary ovarian insufficiency, or FXPOI. This genetic mutation is extremely rare yet, once diagnosed, the treatment was extremely simple. Within a month of wearing a hormone patch, all of Carly’s symptoms had been virtually eliminated.

Diagnosis Roulette?

Inspired by his sister’s diagnostic odyssey, Heyman created CrowdMed for others suffering from "orphan diseases," those conditions which affect fewer than 200,000 people and so are difficult to diagnose since doctors rarely, if ever, encounter them. “We’re not looking for patients who haven’t seen a doctor yet,” Heyman told Bloomberg. CrowdMed wants those patients who have seen two or three doctors and haven’t received a satisfying diagnosis. Additionally, CrowdMed is modeled on the idea of a team of diagnosticians, yet with a few technological twists.

A patient visiting the site can submit a case by uploading information about her or his condition online and then setting a reward to be given to anyone who can come up with a diagnosis. On average, posted payments range from $165 to $300, with the winnings split between those who give the right answer; CrowdMed itself takes 10 percent. Next, after reading the case, the MDs or medical detectives post diagnoses they believe fit the symptoms. At the same time, the MDs also bet on the probable outcome. In this way, each diagnosis gains a value based on the number of points earned and then an algorithm predicts the probability of each diagnosis being correct. (There’s always an algorithm…)

Since its launch last year, CrowdMed has run over 350 real-world cases with more than half of its patients reporting their CrowdMed results brought them closer to a correct diagnosis or cure. On average, according to the company website, these patients have already been sick for six years and incurred over $55,000 in medical bills while visiting eight doctors.

One important question remains: Who are the MDs, willing to solve difficult patient diagnoses for mere pebbles (compared to a doctor's ordinary salary)? Among the 700 people who currently compose its community of detectives, some are retired doctors, oh so bored with playing golf or making pies for the grandkiddies, while others are medical school students wanting experience with real, though virtual, patients. Schools represented include Northeast Ohio Medical University, Trinity College Dublin, George Washington University, and Stanford School of Medicine.

So, is it worth a spin? I'm sure most of us feel the same: We'd hate to be in a position to find out.