基于SVM的医学图像分类器的设计  

Design of Medical Image Classifier Based on SVM

作　　者：华翔[1] 孙蕾[2] 

机构地区：[1]西安工业大学电子信息工程学院,陕西西安710032 [2]西安电子科技大学,陕西西安710071

出　　处：《微电子学与计算机》2011年第6期171-175,共5页Microelectronics & Computer

摘　　要：提出了一个基于支持向量机的医学图像分类器.能提取形状和纹理特征作为分类算法的特征输入,进行计算机辅助诊断.提出了一种支持向量机新算法,解决了当两类中的样本数量差别较大时,支持向量机的分类能力将会下降的问题.实验表明,在小样本、两类样本数量严重不均衡的情况下,该算法有着较强的分类能力,可以极大地提高医学图像分类的效率和准确性.A medical image classifier is proposed in this paper.Gradient Vector Flow（GVF） is used to segment tumor area,and the texture and shape features are the inputs to classifier.A novel SVM algorithm is presented in this paper,which solves the problem that SVM has a poor performance when the two-class problem samples are very unbalanced.The experimental results indicate that the modified algorithm has a strong capability of classification for the unbalanced samples,and the efficiency and accuracy in the medical images are greatly improved.

关 键 词：支持向量机 医学图像分类 特征提取 图像分割 梯度矢量流 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]