Discrimination of benign and malignant microcalcifications using dual-energy imaging

Abstract

Screening mammography has a high detection efficiency for microcalcifications. In a mammogram, morphological characteristics are used to distinguish benign and potentially malignant changes. In this study, we propose a new method to distinguish malignant and benign microcalcifications using only mammographic images based on the dual-energy method. A photon-counting spectral mammography system was simulated using the Geant4 Application for Tomographic Emission (GATE) simulation tool. The dual-energy images were obtained using two energy bins. Microcalcifications of type I (calcium oxalate) and type II (calcium hydroxyapatite) were studied. For statistical analysis, the microcalcifications were classified as type I or type II based on a score calculated using the dual-energy images. The score values were calculated using the ratio of values at low energy and high energy because the attenuation difference was small in the high-energy region and large in the low-energy region. In other words, the classification of microcalcifications associated with pathogenesis was performed using the attenuation ratio as a discrimination criterion. For the appropriate dual-energy images, the effects on the energy spectra of microcalcifications were evaluated. We confirmed that the contrast and the noise were affected because the classification method used in this study is based on the pixel values of the images. In addition, we suggested the possibility of automatic classification for malignant microcalcifications using segmentation methods and the minimum and maximum thresholds of score values.