Development of a graphical user interface software for the prediction of chronic kidney disease

Abstract

Chronic Kidney Disease (CKD) is a severe kidney damage that is difficult to diagnose at the early stages due to the absence of clear symptoms. Late diagnosis of CKD is a common problem in low-income countries and is often associated with lower chances of survival. This study was designed to develop a user-friendly web-based graphical user interface (GUI) software for the prediction of CKD using artificial neural networks (ANNs). The model was developed using Python programming language and trained with 1200 instances of CKD datasets obtained from the University of California Irvine (UCI) machine learning repository. This dataset was split into 80% for training and 20% for testing achieved through an iterative process. A GUI software was developed based on the model using Django, an open-source python web development framework. The model achieved an accuracy of 95.83%, a precision of 100%, a specificity of 100%, and a sensitivity of 89.80%. The GUI software was effectively used to predict CKD and could be of immense benefit as a point of care application for early CKD prediction