The major tool for diagnosing malaria is microscopy. Hundreds of millions of blood slides are inspected under the microscope every year for parasites causing malaria. This is a tedious and error-prone manual process, whose success depends on the experience of a microscopist.
To reduce the diagnostic burden, the National Library of Medicine (NLM), part of the National Institutes of Health, has developed software that can count parasite-infected and uninfected red blood cells using automatic image analysis and machine learning algorithms developed by computational scientists at NLM and the University of Missouri. With funding from the HHS Ventures Fund, this software has been ported to Android smartphones. Running on a camera-equipped smartphone attached to a microscope by an adapter, the software screens the field of view for malaria parasites and reports the level of parasitemia to the microscopist.
The software has been trained with more than 200,000 blood cell images, which the NLM team acquired in Bangladesh and have been annotated by an expert microscopist.
The team is now field testing the software in Bangladesh and Thailand, following which the software will be made widely available to other sites.