Utility of Artificial Intelligence in Diagnosis of Malaria and Dengue on Yumizen H 550 Cell Counter

Introduction: Diagnosis of malaria has always been a diagnostic challenge in endemic areas. For many centuries microscopy-based diagnosis has been a gold standard method. Despite the presence of expert microscopists, laboratory misdiagnosis of malaria is still a problem. The HORIBA Medical Haematology Analyzer, Yumizen H550, provides flags for Malaria and Dengue developed through machine learning methods. It can be used as a screening tool for Malaria and Dengue fever in endemic areas. This study aimed to evaluate the sensitivity and specificity of these flags. Methods: The study done in western India from July-2020 to October-2020. Total 335 patients were taken. Among them, 220 were febrile patients and 115 were non febrile patients. Among 220 febrile patients, 116 patients were P.Vivax Positive, 07 were P.Falciparum positive and 32 were Dengue serology positive. Microscopic examination was considered as the confirmatory method for Malaria. Rapid malarial antigen tests were used for additional confirmation. For dengue serology markers like NS1antigen, IgM and IgG by ELISA was used for confirmation. Results: In our study, at cut off point of 0.45, Sensitivity and specificity for P.Vivax was 81.03% and 90.91% respectively. Conclusion: Horiba Yumizen H 550 cell counter gives useful for P.Vivax flag. P.Falciparum and dengue fever flags are not effective.