The simulated CT images were generated with the XCAT anthropomorphic digital phantom (Duke University, Durham, North Carolina, U.S.A.) at several inspiration levels and heart positions. This digital phantom provides a virtual model of patient anatomy and physiology including pulmonary vasculature and bronchopulmonary structures up to the terminal branches [1, 2]. The phantom allows for the specification of more than 250 parameters, including contrast administration, and  respiratory and cardiac phase. This allows us to manipulate the voxel values representing different tissues with enhanced attenuation due to the presence of intravenous iodinated contrast agent. In this way, CT images could be simulated representing a patient being scanned at different respiratory and cardiac phases with and without contrast.

100 clinical subtraction CT pulmonary angiography (CTPA) cases from our institution were retrospectively evaluated to study clinical-relevant inspiration differences by measuring the difference in diaphragm positions between unenhanced and enhanced CT scans. It was found that the mean difference was 5.7 mm, with 20 mm being the 98th percentile.Therefore for the test cases three different diaphragm positions were used: 3 mm (small difference), 8 mm (approximately average clinical difference and closest available to 5.7 mm) and 20 mm (large difference). Our institution scans patients for the unenhanced CT scan with a higher dose (lower radiation) compared to the enhanced CT scan (higher radiation). To test for systematic registration errors, first a pair of images were generated at two identical respiratory phases, but different noise level: one containing intravenous contrast and one without intravenous contrast. For the training set there are three other set of scans with similar diaphragm position (2 mm, 8 mm, and 19 mm).

If you would like to work with the XCAT phantom, we would like to refer to the next two papers:

1.            Segars, W.P., et al., 4D XCAT phantom for multimodality imaging research. Med Phys, 2010. 37(9): p. 4902-15.

2.            Abadi, E., et al., Modeling Lung Architecture in the XCAT Series of Phantoms: Physiologically Based Airways, Arteries and Veins. IEEE Transactions on Medical Imaging, 2017. PP(99): p. 1-1.