Laboratorio de Procesamiento Digital de Señales, Departamento de Ingeniería Electrónica, DICIS, Universidad de Guanajuato, Carr. Salamanca-Valle de Santiago KM. 3.5 + 1.8 Km., Salamanca 36885, Mexico.
Departamento de Ingeniería Robótica, Universidad Politécnica de Guanajuato, Av. Universidad Norte SN., Comunidad Juan Alonso, Cortazar 38496, Mexico.
J Healthc Eng. 2018 Dec 30;2018:2849567. doi: 10.1155/2018/2849567. eCollection 2018.
In mammograms, a calcification is represented as small but brilliant white region of the digital image. Earlier detection of malignant calcifications in patients provides high expectation of surviving to this disease. Nevertheless, white regions are difficult to see by visual inspection because a mammogram is a gray-scale image of the breast. To help radiologists in detecting abnormal calcification, computer-inspection methods of mammograms have been proposed; however, it remains an open important issue. In this context, we propose a strategy for detecting calcifications in mammograms based on the analysis of the cluster prominence () feature histogram. The highest frequencies of the histogram describe the calcifications on the mammography. Therefore, we obtain a function that models the behaviour of the histogram using the Vandermonde interpolation twice. The first interpolation yields a global representation, and the second models the highest frequencies of the histogram. A weak classifier is used for obtaining a final classification of the mammography, that is, with or without calcifications. Experimental results are compared with real DICOM images and their corresponding diagnosis provided by expert radiologists, showing that the feature is highly discriminative.
在乳房 X 光片中，钙化表现为数字图像中较小但明亮的白色区域。早期检测患者的恶性钙化可提供对这种疾病的高生存预期。然而，由于乳房 X 光片是乳房的灰度图像，因此白色区域很难通过目视检查看到。为了帮助放射科医生检测异常钙化，已经提出了计算机检查乳房 X 光片的方法；然而，这仍然是一个悬而未决的重要问题。在这种情况下，我们提出了一种基于簇突出度 () 特征直方图分析的乳房 X 光片中钙化检测策略。直方图的最高频率描述了乳房 X 光片中的钙化。因此，我们使用 Vandermonde 插值两次获得一个函数来模拟直方图的行为。第一次插值得到一个全局表示，第二次则对直方图的最高频率进行建模。弱分类器用于对乳房 X 光片进行最终分类，即有无钙化。实验结果与真实的 DICOM 图像及其由专家放射科医生提供的相应诊断进行了比较，结果表明特征具有高度的可区分性。
