Neural network ultrasonographic diagnosis system of cirrhosis using DWT for preprocessing

In this paper, a pyramid neural network is applied to characterize the hepatic parenchymal diseases in ultrasonic B-scan texture. The cirrhotic parenchymal diseases are classified into 4 types according to the size of hypoechoic nodular lesions. The B-mode patterns are wavelet transformed, and then the compressed data are fed into a pyramid neural network to diagnose the type of cirrhotic disease. Compared with the 3-layer neural networks, the performance of the proposed pyramid neural network is improved by utilizing the lower layer effectively. The simulation results show that the proposed system is suitable for diagnosis of cirrhosis.