A new proposal for early stage diagnosis of urinary tract infection using computers aid systems

Abstract

Hundreds of newborns everyday are affected by urinary tract infection worldwide. Urinary tract infection can cause serious illness over the long term. Early diagnosis is crucial for the treatment of the disease and the health of the newborn baby. In this study, a decision support system was established for the preliminary diagnosis of whether the newborn who was infected with urinary tract. For this purpose, artificial neural network methods and other bioinformatics techniques for comparison were used. Tests conducted with artificial neural networks resulted in: The probabilistic neural network method gave the best result for the test with 91.4251 ratio, whereas the multilayer neural network method showed the best result with 98.9130 ratio in training. Thus, it has been shown that the classification process with accuracy rate that can be considered important in the use of flexible computation and bioinformatics techniques in diagnosing urinary tract infections in newborn infants is successful.

Keywords

• urinary tract infection
• new-born
• artificial neural network
• medical diagnosis
• classification