A Brazilian classified dataset for prognosis of tuberculosis

Data was collected from the Information System for Notifiable Diseases (from Portuguese, Sistema de Informação de Agravos de Notificação, SINAN-TB) from Brazil, between 2001-01 and 2020-04, and contains information about patients who were diagnosed with tuberculosis.
We pre-processed the data to build a dataset focusing on the treatment outcomes of tuberculosis, focused on two classes: CURED and DIED. The dataset comprises 37 categorical attributes (sociodemographic, clinical, and laboratory data) plus the target class (SITUA_ENCE).

There are 927,909 patients classified as CURED and 36,190 classified as DIED, totaling 964,099 records of patients with tuberculosis.