Abstrait

Research on hepatitis virus identification based on improved BP neural network

Wang Zunfu, Wu Jianlin, Jiang Zhihua, Zhao Caiyun, Yu Bingxue

Hepatitis virus identification plays a key role in clinical diagnosis and is one of the difficulties and hot research fields for the researchers related. The paper takes hepatitis B virus for example and presents a newmodel for hepatitis virus identification based on BP neural network and ant colony algorithm. First, the flow chart of hepatitis virus identification is designed based on the hepatitis virus image processing; Second, aiming at the shortages of the existing BP neural network algorithm of data-mining for hepatitis virus identification, ant colony and BP neural network algorithm are integrated and some improvements are advanced to speed up the convergence and simplify the structure and to improve identification accuracy of the original BPmodel. Finally, themodel is realized by the data fromthree hospitals to carry out comprehensive hepatitis virus identification and the experimental results indicate that the model has favorable hepatitis virus identification results.