Computer-Aided Detection and Diagnosis for 3D X-Ray Based Breast Imaging

Computer-Aided Detection and Diagnosis for 3D X-Ray Based Breast Imaging

Gautam S. Muralidhar (The University of Texas at Austin, USA), Alan C. Bovik (The University of Texas at Austin, USA) and Mia K. Markey (The University of Texas at Austin, USA)
DOI: 10.4018/978-1-4666-0059-1.ch003
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

The last 15 years has seen the advent of a variety of powerful 3D x-ray based breast imaging modalities such as digital breast tomosynthesis, digital breast computed tomography, and stereo mammography. These modalities promise to herald a new and exciting future for early detection and diagnosis of breast cancer. In this chapter, the authors review some of the recent developments in 3D x-ray based breast imaging. They also review some of the initial work in the area of computer-aided detection and diagnosis for 3D x-ray based breast imaging. The chapter concludes by discussing future research directions in 3D computer-aided detection.
Chapter Preview
Top

Introduction

One of the major developments in radiology over the last 15 years has been the advent of new 3D x-ray based breast imaging modalities such as digital breast tomosynthesis (Niklason, et al., 1997), digital breast CT (Boone, et al., 2006; Boone, Nelson, Lindfors, & Seibert, 2001), and stereo mammography (Getty, 2003, 2004; Getty, D’Orsi, & Pickett, 2008). These developments are significant as they promise to alleviate the effects of a key limitation of mammography - anatomical noise caused by overlapping out of plane tissue structures on mammographic projections of the breast. Anatomical noise often hinders accurate detection and diagnosis of breast lesions in mammography. The problem posed by anatomical noise is exacerbated in mammograms of women with dense breasts. Anatomical noise can also mislead radiologists into believing that suspicious findings are present on a mammogram when in reality there are none. Such false positive diagnoses often result in additional imaging tests and biopsy procedures that not only inflate the monetary costs of a screening program, but are also stressful for the women undergoing these procedures. Unfortunately, the positive predictive value of mammography for routine screening of asymptomatic women is quite low (10-30%) (Karssemeijer, et al., 2009; Skaane, 2009; Vinnicombe, et al., 2009).

Computer-aided detection (CAD) systems have been developed to assist radiologists in interpreting mammograms (e.g., (Nishikawa, Giger, Doi, Vyborny, & Schmidt, 1993; Yin, et al., 1991). Commercial CAD systems for mammography have been in use for more than a decade now and recent clinical work (Gilbert, et al., 2008; Gromet, 2008; Skaane, Kshirsagar, Stapleton, Young, & Castellino, 2007), as well as a long history of encouraging laboratory studies demonstrate the high potential of these systems in assisting radiologists to detect early cancer. However, both computer systems and radiologists are constrained by the fact that mammography is a 2D imaging modality. Both human and computer vision are hindered by anatomical noise, which impacts the detection and diagnosis of breast lesions on mammography. This limitation of mammography has spurred the radiology community to develop alternate and adjuvant 3D x-ray based breast imaging modalities for early detection of breast cancer.

In this chapter, we review some of the recent developments in 3D x-ray based breast imaging. We also provide a review of the current strategies adopted in the development of CAD systems for the new breast imaging modalities. We discuss the limitations of the current strategies adopted for the development of CAD systems for 3D x-ray based breast imaging and provide suggestions for future research on the development of these systems.

Complete Chapter List

Search this Book:
Reset