A Disease Detection Method of Liver Based on Improved Back Propagation Neural Network

The computer-aided diagnosis of liver disease plays an important role in timely detection and treatment of liver. Conventional identification methods based on back propagation neural network (BPNN) get problems of overlong training time and local optimum. This paper proposes an application of improved BPNN on liver disease identification and the improvements of BPNN are based on self-adaptive learning rate and momentum in training. Noise suppression and feature extraction of liver CT image which obtained from a hospital in Hangzhou are processed first. A comparison between standard BPNN and improved BPNN shows that the latter gets less training time and better accuracy.