Identification of Malaria-Infected Red Blood Cells Via Digital Shearing Interferometry and Statistical Inference

Malaria is one of the most widespread diseases, particularly in Asia and Africa. Correct diagnosis of malaria is necessary for its proper treatment. A compact automated tool for malaria identification will greatly benefit healthcare professionals in these regions. We propose a method that has the potential to automatically detect malaria-infected red blood cells (RBCs). This method combines the simplicity and robustness of lateral shearing interferometry with the flexibility of statistical methods to achieve the classification of diseased RBCs. Shearing interferograms generated using a glass plate in a common path setup were Fourier analyzed to retrieve the gradient phase and amplitude information of the cell. Then, multiple features based on the complex amplitude information of the cells are measured automatically and used to differentiate healthy and malaria-infected cells. Multivariate statistical inference algorithm of the experimental data shows that there is a difference between the populations of healthy and malaria-infected RBCs by using the measured RBC features.