Chronic Kidney Disease Prediction Techniques: A Survey

Abstract

Chronic Kidney Disease (CKD) is predicted by using the CKD dataset collected from different online and offline sources. Several AI techniques available to perform CKD identification use their own sources of information, such as medical images or medical information in a tabulated form with the markers collected from Electronic Health Record (HER). Researchers have designed prediction models by using publicly available standard CKD information from the University of California, Irvine's Machine Learning for enabling the validation of outcomes as well as the comparison against other models. Various systems have been designed in the previous years to perform CKD prediction based on Machine Learning (ML) and Deep Learning (DL). This study discusses about various research works that are reviewed for the CKD prediction in terms of certain parameters. This work employs ML algorithms after pre-processing the data and compares the performance to obtain the accurate result. Here, the effectiveness is computed by using F1 score, precision, accuracy, recall, and AUC score.