X-ray Digital Tomosynthesis Imaging for Pulmonary Nodule Detection

Lung cancer is currently the primary cause of cancer death, and its incidence continues to increase worldwide. Because of its high sensitivity, normal-dose helical computed tomography (CT) is currently considered the gold standard for lung cancer detection. Previous studies have shown that low-dose helical CT can detect early-stage lung cancer, thereby decreasing morbidity (Yankelevitz et al., 2000). CT is advantageous because it is not susceptible to the problem of reduced accuracy due to overlapping anatomy. However, CT has disadvantages, such as higher radiation doses and costs than chest radiography. The advantages of chest radiography include short examination time, low cost, and easy access; however, low sensitivity and specificity are its main disadvantages. In chest radiography, a three-dimensional view of the chest is projected onto a two-dimensional image; therefore, for many analyses, the detection of pathological findings is limited by overlapping anatomy rather than quantum noise. Chest radiography has relatively low sensitivity for the detection of pulmonary nodules. This poor sensitivity precludes its use as a screening method despite its low cost, low dose, and the widespread availability of radiographic devices. Digital tomosynthesis (DT) imaging provides greater contrast than radiography for images of similarly sized nodules. Three recent studies found that the detectability of pulmonary nodules was considerably higher with chest DT than with chest radiography; and in one report, sensitivity was found to be increased, particularly for nodules smaller than 9 mm (Vikgren et al., 2008). Another study reported that DT is an advantageous technique for detecting pulmonary nodules (Zachrisson et al., 2009). DT was also shown to have considerably improved sensitivity for the detection of known small lung nodules in three size groups [< 5, 5 – 10, and > 10 mm] compared with chest radiography (Dobbins et al., 2008). According to these reports, DT is better than radiography for the detection of lung nodules. DT is a limited angle method of image reconstruction. Furthermore, DT provides the benefits of digital imaging (Godwin, 1983; Littleton, 1983; Siegelman et al., 1980, 1986; Fraser et al., 1986; McLendon et al., 1985) and tomographic benefits of CT at decreased radiation doses and costs. In a recent review, chest DT was described on the basis of advantages and limitations, potential applications, and suggested indications, with supporting evidence from the medical literature (Johnsson et al., 2014; Chou et al., 2014). However, as the projection images in DT are acquired over a limited angle, the depth resolution in the reconstructed section images is limited, and complete removal of superimposed tissue cannot be obtained (Johnsson et al., 2014). There are other difficulties associated with the limited depth resolution of chest DT