Almost a million people a year — mostly kids — die because the parasite that causes malaria is tough to detect. The symptoms of malaria aren’t specific to the disease, and microscopic signs of the Plasmodium parasite looks different at different parts of it’s lifecycle. So detection requires a lab with a good quality microscope, and the time of someone who knows what to look for amid slides containing millions of cells.
There’s got to be a better way to spot malaria in time to treat patients who have it, while avoiding unnecessarily treating people who don’t.
An international group of researchers is putting automation to work to develop a cheaper, faster way of spotting malaria. In their technique a drop of blood is pumped through a disposable microfluidic chip (no more stained slides), and a cheap CMOS chip captures images of laser light scattered off the blood cells. Then two very different methods (fuzzy logic and principal component analysis) are used to crunch numbers to see if the images are statistically similar to infected cells. In the preliminary test of 25 samples, the technique worked very well — and worked fast, cutting time to diagnosis dramatically.
Learn more via the article I wrote for Optics & Photonics News: Spotting Malaria Fast.
Or read the original paper (freely available on the web): Toward fast malaria detection by secondary speckle sensing microscopy in Biomedical Optics Express.