A Deep-Learning-Based System for Automated Sensing of Chronic Kidney Disease

Abstract

In this article, we propose a new sensing technique for the automated detection of kidney disease. The salivary urea concentration is monitored to detect the disease. A new sensing approach is introduced to monitor the urea levels in the saliva sample. Furthermore, for analyzing the raw signals obtained from the sensor, we have implemented a one-dimensional deep learning convolutional neural network (CNN) algorithm, which is incorporated with a support vector machine (SVM) classifier. The use of CNN-SVM integrated network enhanced the classification accuracy of the model. The proposed model successfully classified the samples with an accuracy of 98.04%.