Research On Application Of Support Vector Machine In Liver B Ultrasonic Images Classification

Support Vector Machine(SVM)is a powerful machine learning method based in the framework of Statistical Learning Theory.SVM has become a new active area in the field of classification because of its excellent learning performance.SVM as a new machine learning method is brought into medical image classification.The purpose of this thesis is to utilize the predominance of SVM in informational process for the recognition of liver B ultrasound image classification.Combining with computer technology,biomedicine technology and pattern recognition,we propose liver B ultrasound classification using SVM.Our research works are summarized as follows:At the process of feature extraction of B ultrasound image,we put the image into gray scale,and choose the interested region from the image firstly,then we extract the features of the image including texture statistical moments and co-occurrence matrices texture feature.According to the data of this paper,the feature data have good and rotating invariants and conform to the classification parameter of B ultrasound image.We use 120 samples(30 samples each) which are from the same doctor and the same ultrasound equipment to distinguish these images to interested regions of 200~*200 pixels.And we extract the features from the interested regions.Considering the great difference of data forms,we unit all data in order to apply to class test with the standardization of feature data.An overview of both theoretical basis and principle of Support Vector Machine is given.The classification algorithms are concerned,and theirs advantages and disadvantages and application scope are showed.This thesis discusses the application of the support vector machine in classification.The performance influence of different kernel functions and different texture features is analyzed.The classification results show that RBF kernel can give the best performance in most classification group.Finally,we compare the results of the support vector machine algorithm and the neural network algorithm.The results show that the support vector machine algorithm is superior to the neural network algorithm. At last,we have used Visual C++6.0 to develop a B ultrasound image classification based on SVM.It can identify B ultrasound image of normal liver, hepatic adipose infiltration,schistosome liver and liver cancer.