Clinical malaria diagnosis: rule-based classification statistical prototype.

In this study, we identified predictors of malaria, developed data mining, statistically enhanced rule-based classification to diagnose malaria and developed an automated system to incorporate the rules and statistical models. The aim of the study was to develop a statistical prototype to perform clinical diagnosis of malaria given its adverse effects on the overall healthcare, yet its treatment remains very expensive for the majority of the patients to afford. Model validation was performed using records from two hospitals (training and predictive datasets) to evaluate system sensitivity, specificity and accuracy. The overall sensitivity of the rule-based classification obtained from the predictive dataset was 70 % [68-74; 95 % CI] with a specificity of 58 % [54-66; 95 % CI]. The values for both sensitivity and specificity varied by age, generally showing better performance for the data mining classification rules for the adult patients. In summary, the proposed system of data mining classification rules provides better performance for persons aged at least 18 years. However, with further modelling, this system of classification rules can provide better sensitivity, specificity and accuracy levels. In conclusion, using the system provides a preliminary test before confirmatory diagnosis is conducted in laboratories.