Lower Respiratory Tract Infection Clinical Diagnostic System Driven by Reduced Error Pruning Tree (REP Tree)

Olayemi Olufunke Catherine

Abstract

Combating Lower Respiratory Tract Infections (LRTIs) in Peadiatrics has prompted the development of this new diagnostic model. It was confirmed from several literature reviews that the Lower respiratory tract infections accounts for over a million children illness and death yearly as the effect of lack of prompt diagnosis, or no diagnosis due to shortages of medical experts and medical facilities in our localities. This new diagnostic model was built by applying Reduced Error Pruning Tree (REP Tree) Algorithm on the LRTIs data sets collected from a Federal Medical Center Owo, in Ondo State. When the model was tested, it presented 100% detection proportion on the training cases and 95.7142% success proportion on the testing cases. It is sure that the full implementation of this presented model (rules generated from the REP Tree) on any platform will decrease the high death rate associated with respiratory infections in peadiatrics.