This study is related to the PhD study entitled "Study of a prediction model for tuberculosis treatment abandonment" was financed by the São Paulo Research Foundation (FAPESP), grant #2018/23963-2, and to the e-Science Program: Digital Human Health, grant #2020/01975-9. The study main goal is to research and develop a model for TB treatment outcomes. This model would allow better decision making process in TB patients healthcare and improve the healthcare data management and quality. We will apply the Knowledge Discovery in Databases (KDD) method to the public datasets that provide information about the whole patient clinical history during the TB treatment. Then, a data mining technique will identify clusters in clinical pathways done through patient TB treatment and correlation will be analyzed between these clinical pathways and the treatment outcome. The term clinical pathways is defined as the sequence of clinical decisions done through the TB patient treatment. After data description of the clinic pathways the predictive model for the TB patient treatment outcome will be developed using machine learning algorithms to create a classifier for bad outcomes (death, treatment loss to followup, hospitalization, resistance development).
News published in Agência FAPESP Newsletter about the scholarship: