Fig 1.
Stationary Digital Chest tomosynthesis system, CNT array retrofitted into a portable X-ray system.
Fig 2.
Reader score sheet provided to radiologist for both Chest X-ray and Stationary Digital Chest Tomosynthesis.
Fig 3.
Representative coronal slices obtained on the s-DCT system (a-c) and corresponding CXR from the same patient (d).
There is improved in-plane visibility of the left port-a-cath and line, proximal airways, and vasculature. The retrocardiac lung and osseous structures (ribs and spine) are better visualized in-plane given lack of superimposed structures.
Fig 4.
Comparison of single slices of s-DCT (a) with corresponding CXR (b). Comparison single coronal slice CT images are also provided (c).
Row 1: Improved visualization of a 1.4 cm solid pulmonary nodule in the medial right middle lobe. The overlying right atrium limits evaluation on the CXR. Row 2: S-DCT and CXR show a dense left perihilar consolidation. Tomosynthesis demonstrates ground glass opacities in the retrocardiac left lower lobe, not seen on CXR, which are confirmed on chest CT. Row 3: S-DCT demonstrates a ground glass nodular opacity in the subpleural right upper lobe. Not well seen on CXR due to the overlying osseous structures. Findings are confirmed on chest CT.
Table 1.
Difference in confidence of reader assessments of s-DCT versus CXR, by 10 separate criteria and overall, p-values and confidence limits. Positive calculated differences favor s-DCT. LUL = lower confidence limit, UCL = upper confidence limit.
Table 2.
Predicted probabilities of the outcomes 1) Reader perceived additional information was provided by s-DCT and 2) CT recommended as a follow up to s-DCT compared to low, median, and high confidence intervals of reader overall confidence in s-DCT evaluation.
Fig 5.
Average difference in overall reader confidence in s-DCT vs. CXR compared to disease processes.
Each bar represents an average difference in confidence for each study, among three readers. A negative value favors confidence in CXR over s-DCT.