Scientists have been looking for a way to more quickly diagnose malaria in underdeveloped regions. Now, Duke University researchers are moving one step forward with a new technique that uses holographic imaging and deep learning to pinpoint the infection.
A computerized method that relies on computers and light-based holographic scans correctly identified malaria-infected cells in a blood sample, scientists said in results published recently in the journal PLOS ONE. The technique doesn’t require any human intervention and could form the basis for a rapid field test for malaria.
“With this technique, the path is there to be able to process thousands of cells per minute,” Adam Wax, a professor of biomedical engineering at Duke who helped pioneer the technology, said in a statement. "That’s a huge improvement to the 40 minutes it currently takes a field technician to stain, prepare and read a slide to personally look for infection."
The scientists’ method is based on quantitative phase spectroscopy, a technology that renders holographic images with information on malarial infections in cells.
Researchers also used deep learning to create their technique. Duke engineers fed data on more than 1,000 healthy and diseased cells into a computer, and a program learned how to tell the difference between the two types of cells. A resulting algorithm could spot malaria 97% to 100% of the time.
Wax and the paper’s first author, Han Sang Park, are rolling with the results. The pair started a company, M2 Photonics Innovations, to develop a diagnostic device that uses the new technology.