How To Help Malaria Sufferers Without Using Patents: Crowdsourcing Diagnosis
from the working-together dept
A little while back we wrote about Nathan Myhrvold’s sniffy comment that if you’re not doing anything to help people suffering from malaria, you have no right to criticize his patent troll operation, Intellectual Ventures. As we also noted, this argument is rather undermined by the fact that his research involves such deeply impractical solutions as “photonic fences” and using magnets to make mosquitoes explode.
If lives are to be saved here and now, and not in some patent-encumbered fantasy world tomorrow, what we need is a rather different approach that works with resources that are available and cheap today. Perhaps a crowdsourced solution like this:
Background: There are 600,000 new malaria cases daily worldwide. The gold standard for estimating the parasite burden and the corresponding severity of the disease consists in manually counting the number of parasites in blood smears through a microscope, a process that can take more than 20 minutes of an expert microscopist’s time.
Objective: This research tests the feasibility of a crowdsourced approach to malaria image analysis. In particular, we investigated whether anonymous volunteers with no prior experience would be able to count malaria parasites in digitized images of thick blood smears by playing a Web-based game.
Digitized blood sample images were placed on a Web site, and then people were invited to count the parasites in each. A special algorithm was used to combine the analyses from several visitors to produce a better collective detection rate. It seems to work:
Results: Over 1 month, anonymous players from 95 countries played more than 12,000 games and generated a database of more than 270,000 clicks on the test images. Results revealed that combining 22 games from nonexpert players achieved a parasite counting accuracy higher than 99%. This performance could be obtained also by combining 13 games from players trained for 1 minute.
That’s pretty impressive. And unlike bonkers ideas such as “photonic fences”, this crowdsourced approach requires little beyond bandwidth for distributing images and enough people participating. Putting the two together potentially allows huge numbers of blood samples to be checked for the presence of malaria infection with high accuracy once the system has been refined to include additional factors like parasite species and growth stages. That makes this approach scalable — crucially important when there are over half a million new cases of malaria each year. The same can hardly said about using magnets to make mosquitoes explode.