Chapter 6:
Image Processing and Pattern Classification Techniques for the Detection of Architectural Distortion in Prior Mammograms of Interval-Cancer Cases
Editor(s): Jasjit S. Suri S. Vinitha Sree Kwan-Hoong Ng Rangaraj M. Rangayyan
Author(s): Banik, Shantanu, Univ. of Calgary Rangayyan, Rangaraj M., Univ. of Calgary Desautels, Joseph Edward Leo, Univ. of Calgary
Published: 2012
DOI: 10.1117/3.899757.ch6

6.1 Introduction

Breast cancer is one of the major health issues among women. It is the most common and second leading cause of cancer death among women. Breast cancer has become a major health problem in both developed and developing countries over the past 50 years. Because only localized cancer is deemed to be treatable and curable, as opposed to metastasized cancer, early detection of breast cancer is of utmost importance. Efficient detection of breast cancer in its early stages can play an important role in reducing the associated mortality rates: localized cancer leads to a five-year survival rate of 97.5%, whereas cancer that has spread to distant organs has a five-year survival rate of only 20.4%.

Mammography is, at present, the best available examination for early detection of signs of breast cancer. Mammograms can reveal pronounced evidence of abnormality, such as masses and calcifications, as well as subtle signs such as bilateral asymmetry and architectural distortion. Mammographic screening has been shown to be effective in reducing breast cancer mortality rates: screening programs have reduced mortality rates by 30% to 70%. Furthermore, accurate detection of malignancies via mammographic screening can result in a reduction in the number of benign biopsies while maintaining a desired sensitivity, which will not only reduce health care costs associated with biopsies but also lessen patient suffering caused by the traumatic experience of biopsy.

However, interpreting screening mammograms is difficult; the sensitivity of screening mammography is affected by image quality and the radiologists' level of expertise. Another factor that affects a radiologist's performance is the high volume of cases examined in a screening program. Bird, Wallace, and Yankaskas estimated the sensitivity of screening mammography to be between 85% and 90%; misinterpretation of breast cancer signs accounted for 52% of the errors, and overlooking signs corresponded to 43% of the missed abnormalities. In a study by van Dijck et al., minimal signs of abnormalities were found to be present on screening mammograms taken previously in many cases of screen-detected cancers. Although double reading of screening mammograms has been shown to provide higher sensitivity than single reading, the resources and expertise required for this purpose render such an approach impractical. Computer-aided diagnosis (CAD) can help in increasing the Image detection sensitivity and accuracy by providing a "second opinion" to the radiologist, and can be almost as effective as double reading.

Online access to SPIE eBooks is limited to subscribing institutions.

Back to Top