A survey of machine learning in kidney disease diagnosis
Jaber Qezelbash-Chamak, Saeid Badamchizadeh, Kourosh Eshghi, Yasaman Asadi
Abstract
Applications of Machine learning (ML) in health informatics have gained increasing attention. The timely diagnosis of kidney disease and the subsequent immediate response to it are of the cases that shed light on the substantial role of ML diagnostic algorithms. ML in Kidney Disease Diagnosis (MLKDD) is an active research topic that aims at assisting physicians with computer-aided systems. Various investigations have tried to test the feasibility, applicability, and superiority of different ML methods over each other. However, lacking a holistic survey for this literature has always been a noticeable shortcoming. Hence, this paper provides a comprehensive literature review of ML utilizations in kidney disease diagnosis by introducing two different frameworks, one for MLs, classifying various aspects of kidney disease diagnosis, and the other is the framework of medical sub-fields related to MLKDD. In addition, research gaps are discovered, and future study directions are discussed.