Analysis of Urinary Tract Infections Using Machine Learning Based Prediction and Classification

Infections of the urinary system are the most prevalent kind in humans. An infection of the urinary tract, encompassing the ureters, the bladder, kidneys, and urethra, is known as a UTI. The area of the urinary tract most commonly infected with a UTI is the lower part of the urinary tract, which includes the bladder and urethra. Although the bladder infection is inconvenient and uncomfortable, if it spreads to the kidneys, catastrophic consequences could result. Women are physiologically more susceptible to urine infection than males are. In this study, the causes and consequences of urinary tract infections data sets are examined and evaluated, as well as the efficacy of machine learning in predicting urinary disease. Additionally, the research recommended utilising machine learning to predict urine infections, predict the outcomes of planned therapies using real-time data, and test those predictions. To obtain high accuracy in predicting UTIs, this paper utilizes a confusion matrix, machine learning classification, and an artificial neural network (ANN) approach. Certain parameters were selected using an analysis of variance technique. A naive Bayes classifier, J48 decision tree algorithm, and ANN were employed to predict urine infections. The proposed model achieved an accuracy of approximately 85.6% for UTI prediction.