Parasite and Infected-Erythrocyte Image Segmentation in Stained Blood Smears

Identifying parasite species and their stage is very important in scrutinizing the properties of malaria, and preventing as well as diagnosing it. Unfortunately, it is a labor-intensive and time-consuming task. Moreover, the diagnosis accuracy heavily depends on the experience and skill of the technician, whose training is expensive. Developing a computer-aided system is thus necessary for identifying parasite species and their stage for malaria diagnosis. Extracting the infected erythrocytes and parasites from a blood smear image is essential for such as system. The present study proposes an automatic malaria parasite detector for segmenting parasites and malaria-infected erythrocytes from a blood smear image. This detector can objectively and efficiently help a doctor diagnose malaria. Experimental results show that the proposed detector performs well in segmenting parasites and malaria-infected erythrocytes from a blood smear image taken under a microscope.