Automatic Detection of Malarial Parasite Using Microscopic Blood Images

Objective: Currently, malaria parasite such as Plasmodium falciparum (Pf) and Plasmodium vivax (Pv) detection relies mainly on the manual microscopic examination of Giemsa-stained blood films and it is very long, tedious, susceptible to error and prone to technician dependent variations. Machine based malarial parasite analysis and recognition has opened a new area for early malaria detection that showed potential to overcome the drawbacks of manual strategies. The aim of this paper is to review, analyze, categorize and address the recent developments of machine assisted malarial parasite recognition using machine learning approach based on microscopic images of peripheral blood smears. Materials and Methods: Although, malaria cell segmentation and morphological analysis is a challenging problem due to both the complex cell nature uncertainty in microscopic videos. To improve the performance of malaria parasite segmentation and classification, many researchers have presented camera based approach. We intend to bring the current state of art, future directions and summarize the open problems. Results and Discussion: The literature is reviewed in an explicit segmentation and machine learning framework, the components of which are preprocessing, feature extraction and selection, classification, and evaluation. Conclusion: The accuracy can be improved by using hyperspectral imaging and adding the expert knowledge through neuro-fuzzy.