基于医学图像内容检索的计算机辅助乳腺X线影像诊断技术  

Computer-aided Diagnosis Technique on Mammograms Using Content-based Medical Image Retrieval

作　　者：郝欣[1] 曹颖[1] 夏顺仁[1] 

机构地区：[1]浙江大学生物医学工程教育部重点实验室,杭州310027

出　　处：《中国生物医学工程学报》2009年第6期922-930,共9页Chinese Journal of Biomedical Engineering

基　　金：国家自然科学基金资助项目(60772092)

摘　　要：乳腺癌是女性中高发的恶性肿瘤疾病。近年来,其发病率呈增高趋势。早期发现、早期诊断和早期治疗是降低乳腺癌患者死亡率的关键。计算机辅助诊断(CAD)技术能够有效提高早期诊断的准确性,而基于内容医学图像检索(CBMIR)技术的引入,为乳腺癌的诊断提供了有效的决策支持。文中就近年来基于医学图像内容检索的计算机辅助乳腺X线影像诊断关键技术进行了较为详尽的综述,包括微钙化和肿块检测、特征提取、相似性测度和相关反馈技术等,同时对该领域的发展趋势进行了展望。Breast cancer is a malignant tumor disease with high incidence among women. In recent years, the disease incidence of breast cancer continues to rise. Early detection, diagnosis and treatment are essential to reducing the mortality. Computer-aided diagnosis （CAD） technique improves the accuracy of early diagnosis effectively. Currently, the introduction of content-based medical image retrieval （CBMIR） techniques provides an effective decision support for breast cancer diagnosis procedure. This review presented a detailed overview of recent advances in the key techniques of CBMIR based CAD system, including the detection of microcalcification and mass, feature extraction, similarity measure as well as relevance feedback technique. At the same time, the prospect of the related research fields was discussed.

关 键 词：乳腺癌 计算机辅助诊断(CAD) 乳腺X线影像 基于内容的医学图像检索(CBMIR) 

分 类 号：R318[医药卫生—生物医学工程]