In many developing countries, the barriers to building rapidly a medical personnel and infrastructure are so huge that millions of children die every year due to complications from easily treated illnesses. One of those illnesses is pneumonia, which actually kills more people than HIV in Africa each year. And the sad thing about it is, a simple diagnoses of chest complications is all that is needed to distribute penicillin to the patient. The problem there is twofold: sometimes field doctors don't have the right equipment. And worse, sometimes they don't have any training.
So a team of doctors, researchers and hackers came up with StethoCloud, which takes a regular stethoscope and hooks it up to a smartphone so that a simple set of code can diagnose the patient quickly. The system runs on the Windows Azure cloud.
But here is what is interesting about it. With the cloud, it would make it entirely easy to treat the disease as a living, breathing data set. Uploading to the cloud data from millions of diagnostic tests would enable doctors and researchers to see how the disease behaves in the community. Inherent to this, I think, is a switch from treating the disease at every individual source - which is kind of reactive -- to understanding the disease as an organism. It's kind of like every Aspen tree is really one single Aspen tree. Disease moves through communities the way ideas do, passing from each susceptible source to each able spreader.
Right now, medical infrastructure, even in the developed world, masses huge resources against treating the individual. It spreads out a ratio of, say, 5000 possible patients to each doctor, and creates a workflow that is reactionary rather than predictive. That's really because all a doctor can do is apply her basic knowledge against a series of presentable symptoms. What if you reversed that? What if, by using smartphones, you were able to build, at scale, a platform of disease data?
What I think happens then is that you can leverage the huge power of the cloud to make a predictive diagnosis of where the disease will be next.
We don't often see disease as a collecdtive organism, because we don't treat it collectively. StethoCloud will be one of the first steps to understand that behavior, and I would bet that with more of this kind of smart phone technology, we might also see a huge decrease in the presentation of pneumonia, and, who knows, malaria, HIV, tuberculosis and more.