Evaluation of Differential Analytic Model as a Tool for the Diagnosis of Malaria Infection
James Ajigasokoa Ndako, Surajudeen Alim Junaid, Ogechukwu Y. Ozoadibe, Victor T. Dojumo, Akinyomade Oladipo Owolabi · Journal of Applied Sciences · 2023
Background and Objective: Malaria is a life-threatening parasitic disease that is entirely preventable and curable but considered as one of the main diagnoses of acute febrile illness in the tropics.Alteration of various haematological parameters has been observed among patients with malaria parasite infection.Analytic models are used to provide an explicit framework for understanding accurate malaria diagnosis in the human population.It is therefore necessary to make a critical assessment of the predictive differential model for accurate diagnosis.Materials and Methods: Blood samples were collected from volunteer subjects using the standard method.These were analyzed and evaluated using an automated haematology analyzer.Results: The Fisherʼs linear discriminant function for gender classification of haematological properties of malaria patients obtained by subtraction is presented as: z = -0.0030PCV-0.0315WBC-0.0557NEUT-0.0367LYMP-0.2227MONO+0.0018PLT.The sample mean of the discriminant function scores was obtained by the constant coefficients as, z G = 6.354.The threshold against which a paediatric patient discriminating score is evaluated as 0.0000, this is obtained by taking the average of the centroid function.This implies that any new patient with discriminant scores greater than 0.0000 would be diagnosed to be a male, otherwise, a female.These data analysis hypotheses efficiently the categorization of the genders with 59.3% accuracy based on the retribution estimate method and 53.6% accuracy based on the leave one out method.Conclusion: This model could be useful as preliminary data to adopt better strategies for precise diagnosis of Malaria infection.