Skip to content

Crowdsourcing malaria diagnosis with games

5 Mar, 2013

The malaria biogame developed by UCLA.

Disclaimer: This project is not funded or affiliated with the Wellcome Trust.

Can you turn gamers into armchair pathologists? At the recent AAAS conference in Boston, Dr Aydogan Ozcan presented an online game his lab is developing to identify red blood cells that are infected with the malaria parasite.

Looking at blood samples under a microscope remains the most common way of diagnosing the disease. Although more modern techniques such as PCR are available, these are hard to implement in areas where the disease is endemic because access to equipment and reagents can be few and far between. But signs that a red blood cell is infected with malaria can be easy to miss and extensive training is required to be able to accurately interpret and diagnose from microscope images. Advances in microscope technology mean that it’s getting cheaper and easier to get the equipment into developing countries, but what about analysing the samples? There simply aren’t enough trained professionals to go around.

Could gaming be the answer? Ozcan and his team at University of California, Los Angeles, have developed an online computer game with a simple premise. Players are shown images of red blood cells and have to decide if they are infected or not. Using mathematical wizardry (a dark art to me), they can model the error rate of each player and adjust the results to get pretty close to the diagnostic accuracy of trained professionals. In a study published last year in PLoS ONE, they reported an accuracy of within 1.25 per cent of a professional using only non-expert gamers.

The malaria biogame developed by UCLA.

With such high stakes, no one is arguing that the game could replace a trained professional. But online games are already being used to identify galaxies, solve protein structure problems, and analyse cancer cells. What if the players are scientists themselves? In his talk, Ozcan described a controlled experiment using only professional pathologists. The result was a digital ‘super pathologist’ with 99.5 per cent accuracy in identifying infected cells, that’s compared with 85.1 per cent accuracy for the best pathologist in the group.

According to the World Health Organization, around half a billion people are infected with malaria each year. That’s a lot of samples to analyse. If nothing else, this could be a useful training platform for early-career medical professionals. But who knows, maybe one day we could all be helping to diagnose this devastating disease from the comfort of our living rooms or maybe even on the bus. Why not have a go? Warning – it’s quite addictive!

Jen Middleton

Jen Middleton is Senior Media Officer at the Wellcome Trust

2 Comments leave one →
  1. 5 Mar, 2013 11:00 pm

    Really interesting Jen, thanks!

    Given your interest, I think that you (and the other readers here) would be really interested in some recent research that I have come across that theorizes about crowds and citizen science.

    It’s called “The Theory of Crowd Capital” and you can download it here if you’re interested.

    In my view it provides a powerful, yet simple model, getting to the heart of the matter.

  2. 6 Mar, 2013 10:21 am

    Reblogged this on Belledelettres's Blog and commented:
    what do you think?

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: