Significant symptoms and nonsymptom-related factors for malaria diagnosis in endemic regions of Indonesia

ObjectivesThis study aims to identify significant symptoms and nonsymptom-related factors for malaria diagnosis in endemic regions of Indonesia.MethodsMedical records are collected from patients suffering other febrile diseases public hospitals Indonesia. Interviews with eight Indonesian medical doctors conducted. Feature selection machine learning techniques used develop classifiers identifying factors.ResultsSeven (duration fever, headache, nausea vomiting, heartburn, severe symptom, dizziness, joint pain) patients’ history as a factor contribute most diagnosis. As fever duration is more than temperature or distinguishing diseases. Shivering, sweating (known indicate presence Indonesia) shown be less regions.ConclusionsThree suitable have been developed the features that can predict distinct With extensive experiments on classifiers, identified help clinical raise awareness at early stages.