A Novel Hybrid Data Mining Technique Using Hybrid Optimized Deep Learning Methods for Chronic Kidney Diseases

  • Dr. Ratna Raju Mukiri, Dr. Bui Thanh Hung, Dr. Prasun Chakrabarti
Keywords: Chronic Kidney Diseases, Machine and Deep learning methods, Saliency Segmentation, Extreme Learning

Abstract

Chronic kidney Diseases (CKD) has become an important issue with an exponential growth.  Since the personcan survive only for eighteen days without kidneys, thrust on predicting the CKD is considered to be real challenge for the physicians and researchers. Data mining has now used in the process of prediction of various diseases. Machine and deeplearning based datamining technique has gained the bright light of research for an effective classification of diseases. This work proposes an ovel hybrid deep learning mining mechanism to predict the CKD or chronic renal diseases (CRD). Theproposed methodology incorporates the data pre-processing technique, saliency segmentation and uses the hybrid deep learning model for better classification. The principle of convolutional neural networks anensemble with optimized extreme learning classifier is used for CKD classification. The comprehensive experimentation has been carried out using the massive datasets which is implemented using Python 3.7 with Keras API. Simulation results shows that the proposed methodology has shown the better performances than the existing classifiers such as Support Vector Machines (SVM), Naïve Bayes (NB), Random Forest(RF), Artificial Neural Networks(ANN), Deep neural network(DNN) and Adaptive Hybrid Convolutional Neuralnetworks (AHCNN).

Published
2022-01-23
How to Cite
Dr. Prasun Chakrabarti, D. R. R. M. D. B. T. H. (2022). A Novel Hybrid Data Mining Technique Using Hybrid Optimized Deep Learning Methods for Chronic Kidney Diseases. Design Engineering, (1), 396-417. Retrieved from http://www.thedesignengineering.com/index.php/DE/article/view/8816
Section
Articles