Chapter 4:
Nonlinear Unsharp Masking for Enhancing Suspicious Regions in Mammograms
Editor(s): Jinshan Tang Sos S. Agaian
Published: 2013
DOI: 10.1117/3.1002311.ch4
Breast cancer is the leading cause of death in women between the ages of 35 and 55. The National Cancer Institute estimates that one out of eight women in the United States will develop breast cancer at some point during her lifetime. The mortality rate of 30% in the U. S. and 45% in Europe has been demonstrated by repeated, randomized, and controlled trials. Currently, there are no effective ways to prevent breast cancer. However, treatments of breast cancer in the early stages are more successful; therefore, early detection is an important and effective method to significantly reduce mortality. There are several imaging techniques for breast examination, including magnetic resonance imaging (MRI), ultrasound imaging, positron emission tomography (PET) imaging, computed tomography (CT) imaging, optical tomography/ spectroscopy, and x-ray imaging. Among them, mammography (x-ray image) is the most-common technique that radiologists use to detect and diagnose breast cancer. Two types of mammography are currently used: film mammography and digital mammography. Digital mammography is preferred by physicians because it has better image quality, requires a lower x-ray dose, provides interpretations with greater confidence in difficult cases, and offers faster diagnosis for routine cases. Due to the limitations of the x-ray hardware systems, screened mammograms - even when using digital mammography - may present low resolution or low contrast, making it difficult to detect tumors at an early stage. Important indicators of early breast cancer, such as irregularly shaped microcalcifications, are very small calcium deposits that appear as bright, granular spots in mammograms. The distinction between the tiny malignant tumors and the benign glandular tissue is not readily discernable; misinterpretation results in unnecessary additional examinations and biopsy. The situation becomes worse when radiologists routinely interpret large numbers of mammograms and can misdiagnose a condition. To improve the visual quality of mammographic images, more image data can be collected at the data acquisition stage, thus improving the image resolution. However, this significantly increases the overall acquisition time, the amount of radiation that a patient is exposed to, and hardware costs. On the other hand, the image visual quality can be enhanced during the postimage-processing stage in medical imaging systems. It utilizes different image enhancement techniques to enhance the contrast of mammograms. In this way, the visual quality of mammograms is improved without affecting the acquisition process or increasing the hardware costs.
Online access to SPIE eBooks is limited to subscribing institutions.

Back to Top