Abstract
Hepatocellular carcinoma is one of the most common types of liver cancer in adults. In patients having this disease, prediction of survival is very strenuous. Through this eminent experiment, the authors have proposed a newimproved classification approach using DNN (deep neural network) for predicting survival of patients with hepatocellular carcinoma. The dataset was obtained at a University Hospital in Portugal and contains several demographic, risk factors, laboratory and overall survival features of 165 real patients diagnosed with HCC. Authors have selected 15 risk factors out of 49 risk factors which are significantly responsible for HCC in this proposed method. The outcome of this experiment has proved to be of significant increase in accuracy of the prediction of survival over the conventional methods like multivariable Cox model or unsupervised classification.