Kernel Convolutional Neural Network Deep Learning Algorithm To Classify LIVER Disease
Pardeep Kaur1, Harinder Kaur2

1Pardeep Kaur, Department of CSE, Sachdeva Engineering College for Girls, Gharuan (Mohali), Punjab, India.
2Harinder Kaur, Department of CSE, Sachdeva Engineering College for Girls, Gharuan (Mohali), Punjab, India.
Manuscript received on July 20, 2019. | Revised Manuscript received on August 10, 2019. | Manuscript published on August 30, 2019. | PP: 2002-2017 | Volume-8 Issue-6, August 2019. | Retrieval Number:E7826068519/2019©BEIESP| DOI: 10.35940/ijeat.E7826.088619
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (

Abstract: Now a day, liver disease is common disease due to the bad eating habits among individuals. Some disturbance in the functioning of the liver may cause liver sickness. Liver is responsible for overall functioning of the body. Hence, it becomes necessary to diagnosis the liver disease at an early stage. In advanced world of technology, various methods has been been developed to diagnosis and detect the disease includes data mining. This is novel concept to determine the data by extracting features and recognize indications of liver disease by medical experts. The existing technique has implemented optimize the rules released from Boosted classification with a genetic algorithm, to enhance the LDD (Liver Disease Diagnosis) interval of time and accuracy level. Hence, GA is utilized for enhancing and enhancing directions of another method. In this research work, defines a novel method ECNN (Enhanced CNN) of LDD and enable medical specialists to recognize sign of disease and optimization is done for maximum period, decrease the death rate. Clustering and Feature extraction phase to extract the unique feature based on Kernel method and divide the data into a group or cluster-based using FCM algorithm. Implement CNN method to predict or detect the liver disease to improve the performance and classification of rules set. The proposed method has implemented to achieve better performance and compared with existing methods. The simulation tool used in this research works MATLAB 2016a and calculates the performance is Accuracy achieved 96 % ad existing GA accuracy rate 92.9 % achieved in our work.
Keywords: LDD (liver Disease Diagnosis), CNN (Convolutional Neural Network), KPCA (Kernel Principal Component Analysis) and FCM (Fuzzy C-means Clustering) method.