Machine learning models' performance in predicting sputum conversion after two months of intensive phase of anti-tuberculosis medication in Taraba State, Nigeria

TB is a major cause of hospitalization and death across the world, especially in sub-Saharan Africa with mode of transmission basically through cough affecting males, females, young and the old. 
WHO has developed strategies aiming to end surge of the tuberculosis, one of the strategies is aimed at achieving a 100% sputum conversion after completion of intensive phase (ie absence of the bacilli in sputum after 2 months of commencement of anti-TB medication). The study aimed at using machine learning models to develop predictive models and comparing performance of the models.
The study was carried out at the facilities that provide TB care services between May and October 2023 
From the study 4.16% of respondents failed to become sputum negative for the tuberculosis bacilli after two months. The benefits from the models will help in profiling clients at commencement of treatment, a better outcome of the study would be a longitudinal study than the cross-sectional study.
The use of models is not in anyway replacing clinical practice and experience