An Approach for Analysis and Prediction of CKD using Deep Learning Architecture

This paper proposes an efficient and novel approach for non-invasive diagnosis of Chronic Kidney Disease (CKD). A novel sensing module for CKD diagnosis is designed and developed in the proposed work. The concentration of urea in the saliva sample is measured for detecting the disease. A Convolutional Neural Network-Support Vector Machine (CNN-SVM) hybrid deep learning model is implemented for making predictions by extracting and classifying the features from the sensor response. Experiments are conducted to test the proposed model with real-time samples and to compare its performance with conventional CNN. The experimental results indicate that the proposed model outperforms traditional data classification techniques. The proposed hybrid model achieved a prediction accuracy of 96.59 %.