A Review: Analyzing Risk Factors and Prediction for Chronic Kidney Disease using Machine and Deep Learning Techniques

Chronic nephritic illness is another name for chronic kidney disease (CKD). It outlines limitations that impact your kidneys and lower your chances of maintaining good health. Numerous complications, such as elevated blood levels, weak bones, anemia (low blood count), and nerve damage, will be a problem. It is usually possible to prevent chronic uropathy from worsening by early identification and treatment. Between 4.902 and 7.083 million people worldwide are expected to have chronic kidney disease (CKD), requiring renal replacement treatment. The cost-effectiveness of disease prevention measures should be assessed in light of the resources and economic development of the local community. This study covers the research done over the last five years (2019–2023) on machine learning (ML) and deep learning (DL)-based techniques for exact CKD predictions. The study concludes that quick advancements in sensing technologies, machine learning, and deep learning will offer comprehensive, sensible choices for more accurate estimations of CKD prediction. According to projections, CKD will soon require increasingly specialized applications of sensor platforms, ML, and DL techniques, the integration of various sensor modalities with expert knowledge, and the development of hybrid models that combine various ML and DL techniques. This review demonstrates how knowledge-based CKD prediction can sustainably improve products’ productivity and quality.