Open Access

Katharina Preißinger

• Atomic force microscopy (AFM) and fluorescence microscopy allow to reveal diagnostic patterns characteristic to the intra-erythrocytic stages of malaria parasites based on the morphological and fluorescence properties of infected red blood cells (RBCs). These observations provide the basis for a novel neural network (NN)-based scheme that is capable of the high-speed classification of RBCs into the intra-erythrocytic stages with an accuracy as high as 98%. Thus, the approach developed in this work contributes to meeting the rising need for reliable and fast malaria diagnosis and is able to boost NN-based classification of malaria stages in microscopic images. In fact, the universality and the robustness of the NN-based method against imaging platform-specific features naturally grants its applicability for the wide range of light microscopes used worldwide for malaria diagnosis that is a prerequisite for successful in-field applications.