Breast cancer is one of the most common diseases for women. Early detection of breast cancer may decrease the cost of treatments and improve the quality of life of the patients. While screening mammography plays an important role in detecting cancers at an early stage, ultrasonography is employed as a noninvasive diagnostic examination.
Breast cancer screening using mammography has been introduced in Japan for over-50-year-old women by governmental recommendation. Examinations with ultrasound images also being discussed for over-40-year-old women whose breasts tend to be dense, because small nonpalpable masses hidden by normal structures, such as mammary glands, may be shown on ultrasound images.
In Japan, mammographic breast cancer screening for women aged over 50 was introduced in 2003 with governmental recommendation. In addition, examination using ultrasound images for women over 40 has been discussed. This is because younger breasts tend to be denser. Small nonpalpable masses can be hidden by normal structures such as mammary glands that are occult on mammograms. Ultrasound imaging may be used to depict these masses. Some regions in Japan have already employed the screening programs with ultrasounds before the governmental recommendation of mammographic examination. Whole breast scanning devices with Octson mechanisms have been used for diagnosis in three decades.
Screening with mammography computer-aided diagnosis (CAD) and ultrasound has been carried out at private hospitals in Japan. Ultrasound images are employed to determine whether the mass findings using CAD systems are true or false shadows.
Although many researchers have reported their work on mammogram CAD systems, there are very few studies concerning breast ultrasound CAD systems K. Drukker, K. Horsch and Giger et al. is have reported fundamental approachs to detect and to classify lesions in breasts. CAD schemes for breast masses using free-hand probes and 3D imaging devices have been developed. The scheme consists of three steps: detection, classification, and visualization of masses. The detection of masses is based on active-contour and balloon models in 2D and 3D spaces. The visualization technique provides virtual B- and C-mode images to confirm the shape and the position of the lesion in the breast. Details of the detection, classification, and visualization techniques are described in the following sections.
Online access to SPIE eBooks is limited to subscribing institutions.