A Machine Learning-Based System for Detecting Leishmaniasis in Microscopic Images

Abstract

Background: Leishmaniasis, a disease caused by a protozoan, causes numerous deaths in humans each year. After malaria, Leishmaniasis is known to be the deadliest parasitic disease globally. Currently, direct visual detection of Leishmania parasite through microscopy is the “gold standard” for the diagnosis of this disease. However, this method is time-consuming and subject to errors. This study was aimed to develop an artificial intelligence-based algorithm and image processing algorithms for the automatic diagnosis of Leishmaniasis.Methods: The Viola-Jones algorithm was used in this study due to its high recognition speed. This algorithm performs in four stages: detection of Haar-like features, integral image creation, Adaboost training, cascade architecture.Results: A 65% recall and 83% precision was concluded in the detection of macrophages infected with the Leishmania parasite. Also, these numbers were 52% and 35%, respectively, related to amastigotes outside of macrophages.Conclusion: The results contain a fairly high sensitivity, with the specificity being less satisfactory. High processing speed, ease of work, and low expenses are advantages of the presented method compared to other procedures. By adding a few adjustments, this method could be considered a viable option.