According to the World Health Organization approximately 216 million cases of malaria occur in the world and the disease kills about 655,000 people. Malaria is a treatable infectious disease transmitted by mosquitoes. Malaria is caused by a parasite called Plasmodium, which is transmitted via the bites of infected mosquitoes. In the human body, the parasites multiply in the liver, and then infect red blood cells.
Currently, "the gold standard" for malaria diagnosis worldwide consists of first detecting parasites and then counting the number of parasites in blood smears, manually through a microscope- usually the more the parasites more severe is the infection. An average of 100 images has to be visually checked by a specialist; a process that can take up to 30 minutes. Confirmation of a negative diagnosis is ultimately dependent on the technician's expertise. In addition, over time, as malaria prevalence decreases, microscopy technician skills may be redeployed elsewhere.
Thus, we need scalable, fast, ubiquitous and accurate screening systems (a priority in the agenda for malaria eradication). Mobile phone coverage is reaching every corner of the planet and we see the global connectivity as an opportunity to distribute the images all over the internet.
We believe that there is so much on-line talent out there that can be used to analyze malaria images (while you play a game!). To start with, we have developed the MalariaSpot game. During the game, we will log all the player clicks (scores) - both parasites accurately hunted and mistakes - so the analysis of all the games played will allow us to learn (a) how fast and accurate is the parasite counting of non-expert microscopy players, (b) how to combine the analysis of different players to obtain accurate results as good as the ones provided by expert microscopists.
The Malariaspot game is a kick-off proof-of-concept experiment, which is part of a larger vision devoted to:
- - The establishment of a global specialized task force of remote gamers/workers able to perform on-line malaria diagnosis (and potentially other diseases).
- - The development of new on-line games powered by artificial intelligence engines able to diagnose and minimize the time required get a perfect parasite count.
- - Developing a microscopy-in-a-mobile-phone system for telediagnosis, allowing data transfer directly from field workers and health centers to the Malariaspot platform for rapid diagnosis.
The images are Giemsa-stained thick blood films infected with malaria (Plasmodium falciparum parasite) acquired with optical microscopy.
They have been kindly provided by Dr. John Frean from the National Institute for Communicable Diseases in Johannesburg (South Africa), and have been previously used to assess the possibility of using automated image processing systems to count the number of parasites in the research article "Reliable enumeration of malaria parasites in thick blood films using digital image analysis" (Malaria Journal 2009, 8:218).
Medium to low parasitemia images have been selected for the game because of the game design (1 minute game) and the fact that discrepancies between automatic counting methodologies and manual expert counting are bigger in low parasitemia cases.
The game score is generated by comparing the user tags with previously analyzed and diagnosed images - although images which have not-yet been diagnosed by professionals will also be introduced in future games to provide diagnosis and to increase the tagged database.
(Check out this video to learn what causes malaria, how it is treated and whether it can be eradicated.)
The MalariaSpot project was conceived of and directed by Dr. Miguel Luengo-Oroz, a researcher from the Universidad Politécnica de Madrid (Biomedical Image Technologies group & CEI Moncla UPM-UCM), Spain. The entrepreneur Asier Arranz is chief developer.
The multidisciplinary team is supported by Jacobo Gómez (design), Enrique Mendoza (web and database), Anoush Tatevossian & Natalia Rodriguez & Maria Fernandez (communications) and a fantastic team of advisors and supporters around the globe (Maria Luengo, Antonio Blanco, Javier Cepa, Jose Luis Rubio, Nati Luengo and Cesar Martin). Game soundtrack by Simón Garcia "A Night in Compostela (Para 6 contrabajos)", album Intersecciones (Escuela de Musica Creativa).