Xinhui Duan PhD, Abstract Co-Author: Nothing to Disclose

Mingliang Qu MD, Abstract Co-Author: Nothing to Disclose

Guozhi Zhang, Abstract Co-Author: Nothing to Disclose

Zhoubo Li, Presenter: Nothing to Disclose

Lifeng Yu PhD, Abstract Co-Author: Nothing to Disclose

Shuai Leng PhD, Abstract Co-Author: Nothing to Disclose

Ahmed Halaweish PhD, Abstract Co-Author: Employee, Siemens AG

Joel Garland Fletcher MD, Abstract Co-Author: Grant, Siemens AG

Cynthia H. McCollough PhD, Abstract Co-Author: Research Grant, Siemens AG

To assess the impact of increased spectral separation on urinary stone classification in phantoms mimicking a range of patient sizes.

87 human urinary stones of 5 common types (uric acid, cystine, calcium oxalate, brushite, calcium apatite) were scanned in 35, 40, 45, and 50 cm wide anthropomorphic phantoms. Data were acquired using 2 dual-source CT scanners. On scanner A (SOMATOM Force, Siemens), we used 150 kV and 0.6 mm tin on the high-energy tube and 70, 80, 90 or 100 kV on the low-energy tube. On scanner B (Definition Flash, Siemens), we used 140 kV and 0.5 mm tin on the high-energy tube and 80 or 100 kV on the low-energy tube. The volume CT dose index (CTDIvol) for all kV combinations matched clinical values (e.g. 13.5 mGy for 35 cm phantom). A decrease in the mean CT number of water > 10 HU was used as objective evidence of photon starvation; images meeting this criterion were removed from the analyses. Ratios of the CT numbers in the low and high energy images were calculated and used to differentiate uric acid (UA) from all types of non-UA stones, and oxalate from apatite stones. The ability to accurately classify stone type was assessed using receiver operating characteristic (ROC) analysis.

The 100/150 Sn kV combination on Scanner A was the only combination to avoid photon starvation artifacts across all phantom sizes; 70 kV combinations avoided artifact only for the 35 cm phantom. For scanner B, no kV combinations avoided starvation artifacts for the 45 and 50 cm phantoms; 80 kV combinations avoided artifact only for phantoms ≤ 40 cm. Spectral separation increased the area under the ROC curve (AUC) for differentiation of UA and non-UA stones for large phantoms, e.g., 0.95 (Scanner A, 100/150 kV) vs. 0.63 (Scanner B, 80/140 kV) for the 50 cm phantom. At each phantom size, the AUC for oxalate vs. apatite stones was higher for all kV combinations on Scanner A than any kV combination on Scanner B.

Increased spectral separation between the low and high energy x-ray beams resulted in decreased photon starvation artifact in large phantoms, and improved classification of urinary stones, particularly for oxalate and apatite stones.

Increased spectral separation in dual-energy CT improves the ability to characterize renal stone types, especially for large patients.

Duan, X, Qu, M, Zhang, G, Li, Z, Yu, L, Leng, S, Halaweish, A, Fletcher, J, McCollough, C, Urinary Stone Composition Analysis: Impact of Increased Spectral Separation in Dual-energy CT on Urinary Stone Material Classification.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14006619.html