基于内容图像检索的乳腺肿块诊断技术研究  

A Technology Research for Diagnosis of Mammographic Masses Based on Content-based Image Retrieval

作　　者：王庆岩[1] 宋立新[1] 王立[2] 

机构地区：[1]哈尔滨理工大学电气与电子工程学院,哈尔滨150040 [2]黑龙江省医院,哈尔滨150030

出　　处：《生物医学工程学杂志》2010年第5期999-1003,共5页Journal of Biomedical Engineering

基　　金：黑龙江省自然科学基金资助项目(F200912)

摘　　要：为有效辅助医生对乳腺X线图像中肿块进行诊断,提出一种乳腺肿块诊断方法。该方法对一个待查询的感兴趣区域(ROI)提取22个特征,采用k-最近邻(KNN)算法在图像库中检索,并进一步计算查询ROI与检索结果之间的互信息(MI),以提高检索性能指标。最后,将MI最高的前9幅图像作为最终检索结果。分析比较了三种决策值计算方法,为辅助诊断提供有效的决策信息。实验结果表明,此方法优于单独使用KNN方法并能有效地提高诊断准确度。In order to assist doctors in making the diagnosis of mammographic masses,a method is proposed in this paper.Twenty-two features are extracted from each queried region of interest（ROI）.A k-nearest neighbor（KNN） algorithm is used to retrieve similar images from database,and further calculate the mutual information（MI） between the queried image and the images which are in the retrieval results,so as to improve the retrieval performance.Finally,the scheme takes the first nine images with the highest MI scores as the final retrieval results.With the purpose of providing available decision-making information of diagnostic aids,we compare and analyze three calculating methods of decision index.The experiment results show that this method is better than the method using KNN only,and this method improves the accuracy of diagnosis effectively.

关 键 词：乳腺X线图像 基于内容图像检索 计算机辅助诊断 互信息 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]