A Multilayer Perceptron Based Mobile Diagnosis System for Malaria Fever

Abstract

Malaria is a life-threatening disease caused by Plasmodium parasites, which are transmitted to humans through the bite of infected Anopheles mosquitoes. This disease affects millions of people worldwide, particularly in sub-Sahara Africa, and can lead to severe complications such as anemia, cerebral malaria, organ failure and death. Early and accurate diagnosis is critical to effective treatment and control of the disease. Malaria can be detected using a variety of techniques, from clinical evaluation to laboratory analysis. Healthcare professionals use symptoms of malaria to make a clinical diagnosis mostly where laboratory analysis is not available. In an effort to improve the accuracy and efficiency of symptomatic malaria diagnosis, this work presented a Multilayer Perceptron (MLP) model based mobile diagnosis system for malaria fever. The dataset used contained 1733 patients’ records with malaria cases, including the symptoms observed by medical practitioners and complaints made by the patients. The developed MLP model was trained with 70% of the data while 30% was used to test the model. The performance evaluation results of the model showed an accuracy of 100% on the training set and 97.31% on the testing set, also 0.99 F1, 0.99 Recall and 0.99 ROC on the testing set. The mobile diagnosis system provides an easy-to-use interface for users to input their symptoms and receive a diagnosis result. This system poses to be of immense benefits to the health sector and individuals.