Segmentation of the whole breast from low-dose chest CT images

Abstract

The segmentation of whole breast serves as the first step towards automated breast lesion detection. It is also necessary for automatically assessing the breast density, which is considered to be an important risk factor for breast cancer. In this paper we present a fully automated algorithm to segment the whole breast in low-dose chest CT images (LDCT), which has been recommended as an annual lung cancer screening test. The automated whole breast segmentation and potential breast density readings as well as lesion detection in LDCT will provide useful information for women who have received LDCT screening, especially the ones who have not undergone mammographic screening, by providing them additional risk indicators for breast cancer with no additional radiation exposure. The two main challenges to be addressed are significant range of variations in terms of the shape and location of the breast in LDCT and the separation of pectoral muscles from the glandular tissues. The presented algorithm achieves robust whole breast segmentation using an anatomy directed rule-based method. The evaluation is performed on 20 LDCT scans by comparing the segmentation with ground truth manually annotated by a radiologist on one axial slice and two sagittal slices for each scan. The resulting average Dice coefficient is 0.880 with a standard deviation of 0.058, demonstrating that the automated segmentation algorithm achieves results consistent with manual annotations of a radiologist.

Large-scale image region documentation for fully automated image biomarker algorithm development and evaluation

“…The skin surface segmentation 56 provides a separation of the body from other objects in the scanner field of view such as the scanner table. For bone analysis, significant calcified structures are first identified, then ribs 23 , 57 and vertebra 45 , 57 are labeled; finally, the sternum 22 , 57 is labeled, as it is used in the breast analysis 21 , 58 . The cardiac analysis 17 , 18 depends on both the bone and lung segmentations.…”

“…The heart region is localized, and different calcifications (mitral valve, aortic valve, and coronary artery) are identified in this region so that a robust CAC score can be obtained. 18 In more recent work, the vertebral bodies are segmented 45 to identify osteoporosis and compression fractures, and the breast region is segmented 21 , 58 to evaluate breast density and identify any breast abnormalities. The current system embodies 110 individual pixel labels plus a set of group labels.…”

“…The whole breast region as well as the fibroglandular tissue and fat tissue contained in the breast region are segmented using an anatomy-oriented algorithm 21 that separates the pectoral muscles from fibroglandular tissue through the propagation of muscle front along the vertical direction. A coronal view of the breast segmentation as shown in Fig.…”

Automatic anatomy recognition in whole-body PET/CT images

“…Most segmentation methods for PET/CT images described in the literature [18][19][20][21][22][23][24] have confined to mostly one object in one body region and did not consider a whole body region or the whole-body torso. Their goal was different, namely, delineation of a particular object of interest but not object localization.…”