Identifying the Type of Chronic Kidney Disease Based on Heavy Metals in Soil using ANN

Within the recent two decades, chronic kidney disease (CKD) has become a reached global threat. In Sri Lanka, CKD is one of the severe health problems because of the rapid development of CKD of unknown etiology (CKDu) in agricultural zones. Agrochemical and toxic metal contaminations of soil and water, quality of the drinking water, and fluoride level of soil are etiologies for the increase CKDu patients within the farming areas. Early detection of the disease form of the CKD (including CKDu) is critical to prevent and manage the disease and its etiologies. Therefore, this paper introduces an Artificial Neural Network (ANN) model to determine the CKD form based on the physicochemical parameters of the soil in farming areas. The results of the Multi-layer Perceptron (MLP) ANN model have been compared with the Decision Tree and Support Vector Machine (SVM) based on the model accuracy, precision, recall, Root Mean Squared Error (RMSE), and Mean Absolute Error (MAE). According to findings, the ANN model presents the best classification and prediction performance for determining the form of the disease.