An Intelligent Approach to Cost-effective Management of Typhoid Fever in Children

Typhoid fever poses a significant health concern among children, due to its potential for severe complications and high treatment costs. This paper proposes an intelligent approach to modelling the prognosis and management of paediatric typhoid infections leveraging machine learning and data analytics to optimize healthcare resources and improve patient outcomes. The dataset was sourced from hospitals in Edo State, Nigeria. The dataset comprises laboratory tests of blood samples of children, from age zero (0) to five (5) years. Six (6) attributes that exhibit the highest information gain: widal test, malaria count, monocyte, platelets, HB, eosinophils, cost of treatment respectively, were selected and ranked thus: 0.6759, 0.4936, 0.3147, 0.2843, 0.2416, 0.2127, and 0.20. Five different classification algorithms (J48_Consolidated, LMT, RepTree, MultiBoost Decision Stump) were employed. and Random Forest) and their results were compared. Performance analysis on the five (5) classifiers shows that Multiboost Decision Stump exhibited the best accuracy of 94% and least MAE of 0.0604. The finding shows that the developed model when integrated into automated disease diagnostic workflows will provide an enhanced and flexible solution for quick prognosis of typhoid fever in children in the tropics with significant cost savings with respect to hospitalization. The finding presupposes that we can enhance the efficiency of typhoid fever management while ensuring the delivery of quality care to children in resource-limited settings.