Prediction Model for Chronic Renal Failure Based on Machine Learning and Deep Learning

Alvarez-Fuentes J.; Mejía-Paredes A.; Ticona W. · Lecture Notes in Networks and Systems (2026) · DOI: 10.1007/978-3-032-22236-7_3

Resumen

Chronic kidney disease is a public health problem that affects patients’ quality of life and incurs high costs for the healthcare system. In this context, the growing volume of electronic medical records offers the opportunity to develop predictive models that can help in the early detection of the disease. The objective of this research was to develop a predictive model of chronic kidney disease from medical records using machine learning and deep learning techniques. The methodology is structured in four phases: dataset acquisition, data preprocessing (relevant feature extraction, One-Hot, SMOTE and normalization); Machine Learning (RF, SVM, K-NN, Adaboost and DT) and Deep Learning (GRU and LSTM) model building and model performance evaluation. Superior results were obtained with the GRU model with the Accuracy, Precision, Recall and F1-Score metrics, whose results were 99.08%, 98.45%, 98.42% and 99.42%, respectively. In conclusion, the results demonstrate that Machine Learning and Deep Learning algorithms are effective in predicting chronic renal failure, which can facilitate its diagnosis and timely treatment. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.

Instituciones: Universidad ESAN; Universidad Tecnológica del Perú