多模态乳腺肿瘤图像特征提取与分级诊断的研究进展

Research progress of multimodal image feature extraction and classification diagnosis for breast tumor

乳腺癌是女性最常见的恶性肿瘤之一。影像学检查是诊断乳腺肿瘤的重要方法,其主要包括超声、钼靶、核磁共振成像(MRI)检查等。单一医学影像技术存在诸多的局限性,使用多种影像学技术联合诊断能提高分级诊断效率。临床亟需一个高度智能的乳腺肿瘤诊断系统,有望帮助临床医生提高整体诊断效率,降低漏诊误诊率。而图像特征提取是乳腺肿瘤诊断系统研制中的第一步,也是最关键的一步。本文回顾了超声、钼靶和MRI之间联合使用对乳腺癌进行分级诊断的研究方法,就多模态乳腺肿瘤图像特征提取与分级诊断的研究进展进行综述。

Breast cancer isoneof the most common malignant tumoraffecting women in recent years.Imaging examination is an important method for diagnosing breast tumors.It mainly includes ultrasound,mammography,CT,magnetic resonance imaging(MRI)and so on.The use of a single technique has limitations,and the combined use of multiple imaging techniques can improve the efficiency of graded diagnosis.At this stage,an artificial intelligence breast cancer diagnosis system is expected to help clinicians improve the overall diagnosis efficiency and reduce the rate of missed diagnosis.The most important step in developing an accurate and efficient breast tumor diagnosis system is image feature extraction.This article discusses the practical value of the combined use of ultrasound(US),mammogra⁃phy and magnetic resonance imaging(MRI)for the classification diagnosis of breast cancer.This paper reviews the research advances in feature extraction and hierarchical diagnosis of multimodal breast tumor images.