ABSTRACT
X-ray mammography, either lm screen of full-eld digital mammography (FFDM), is still the main screening method for breast cancer in the general population (Hinz et al. 2011).
Karen Drukker and Charlene A. Sennett
7.1 Introduction 111 7.2 Recent Developments in Breast Imaging 111 7.3 Current Clinical Multimodality Viewing Systems for Breast Imaging 113 7.4 Stand-Alone Performance of Multimodality Breast CAD 115 7.5 Observer Study Performance of Multimodality CAD 120 7.6 Future of Multimodality CAD: Clinical Decision Support Systems 121 References 122
While most people agree that annual screening mammograms have decreased mortality by about 30% (Tabar et al. 2003), the technique is far from perfect with an adjusted sensitivity of 89% in entirely fatty breasts and 62.9% in extremely dense breasts. Likewise, adjusted specicity is 96.9% in entirely fatty breasts, but only 89.1% in extremely dense breasts (Carney et al. 2003). For these reasons, other imaging modalities have gained acceptance as adjuncts to mammography (Karellas and Vedantham 2008), most notably ultrasound (US) (Berg et al. 2008; Corsetti et al. 2011; Youk et al. 2011) and MRI. Each imaging modality has its strengths and weaknesses. For example, X-ray mammography is known to have lower sensitivity for women with dense breast tissue (Carney et al. 2003; Evans et al. 2007; Taylor et al. 2011), handheld US is by denition operator dependent, and MRI is costly. To date, the screening protocol for women at high risk at many institutions includes the use of X-ray mammography, US, and MRI (Boetes 2011a; Dhar et al. 2011; Feig 2011; Houssami and Ciatto 2011; Hutton et al. 2011; Le-Petross et al. 2011). Moreover, it has become general practice to include US in the clinical workup of patients for whom a screening or diagnostic mammogram raises suspicion of breast cancer or is inconclusive (Zanello et al. 2011), thus increasing the specicity over the use of mammography alone. Breast MRI has been invaluable in the evaluation of breast diseases. Diagnostic breast MRI is indicated in the staging of breast cancer, the postoperative assessment of histological positive margins, the assessment of women with known metastatic axillary lymphadenopathy and a negative mammogram and breast US, the assessment of response to neoadjuvant chemotherapy, the evaluation of the integrity of silicone implants, and the evaluation of ambiguous clinical or imaging ndings (Levin 2008). In 2007, the American Cancer Society (ACS) published guidelines for the performance of screening breast MRI including BRCA mutation, untested rst-degree relative of a BRCA carrier, calculated lifetime risk of 20%–25% or greater, chest radiation between the ages of 10 and 30, and certain genetic syndromes (Saslow et al. 2007). e sensitivity of MRI is generally superior to that of X-ray mammography as is its specicity for the latest technology (Boetes 2011b). To further improve diagnostic accuracy, second-look US is often performed to provide additional information for ndings on breast MRI (Linda et al. 2008; Destounis et al. 2009; Abe et al. 2010; Carbognin et al. 2010; Trop et al. 2010; Candelaria and Fornage 2011; Luciani et al. 2011). e high cost of MRI and use of a contrast agent (in the case of dynamic contrastenhanced MRI [DCE-MRI]), however, are drawbacks for the implementation as a general screening modality. In fact, the ACS guidelines specically state that MRI is not indicated as a screening modality in women at average risk of developing breast cancer. All three modalities-mammography, US, and MRI-are also valuable for the assessment of response to neoadjuvant therapy (Schlossbauer et al. 2010).
