Diagnosis of Smear-Negative Pulmonary Tuberculosis using Ensemble Method: A Preliminary Research

Rusdah Rusdah, Mohammad Syafrullah

Abstract


Indonesia is one of 22 countries with the highest burden of Tuberculosis in the world. According to WHO’s 2015 report, Indonesia was estimated to have one million new tuberculosis (TB) cases per year. Unfortunately, only one-third of new TB cases are detected. Diagnosis of TB is difficult, especially in the case of smear-negative pulmonary tuberculosis (SNPT). The SNPT is diagnosed by TB trained doctors based on physical and laboratory examinations. This study is preliminary research that aims to determine the ensemble method with the highest level of accuracy in the diagnosis model of SNPT. This model is expected to be a reference in the development of the diagnosis of new pulmonary tuberculosis cases using input in the form of symptoms and physical examination in accordance with the guidelines for tuberculosis management in Indonesia. The proposed SNPT diagnosis model can be used as a cost-effective tool in conditions of limited resources. Data were obtained from medical records of tuberculosis patients from the Jakarta Respiratory Center. The results show that the Random Forest has the best accuracy, which is 90.59%, then Adaboost of 90.54% and Bagging of 86.91%.

Keywords


classification; high-tuberculosis burden country; medical data mining; tuberculosis data; Tuberculosis in Indonesia

Full Text: PDF

Refbacks

  • There are currently no refbacks.