A preliminary dengue fever prediction model based on vital signs and blood profile

A retrospective study was conducted at Hospital Canselor Tuanku Muhriz (HCTM), Pusat Perubatan Universiti Kebangsaan Malaysia, from January 2015 to January 2016, cases registered was included in the study. Linear discriminant model was found to be the most reliable trained model for the 30 dengue fever data with a true positive rate (TPR) of 83.3% and false negative rate (FNR) of 16.7%. The accuracy of the validation was 50% with 50% error.