Size and Attenuation Measurements for Urinary Stone on Virtual Monoenergetic Images Derived From Dual Layer Spectral CT: A Phantom Study

OBJECTIVE: This study aimed to investigate the optimal energy level for virtual monoenergetic (VM) images with specific window settings for stone size and attenuation measurements, and to determine the effects of radiation dose and reconstructed slice thickness on these measurements. METHODS: Forty-eight urinary calcium oxalate stones (<1 cm in diameter) were placed within a 32-cm phantom, and CT scans were performed on a dual-layer detector scanner at 3 radiation doses (volumetric computed tomography dose index=5, 10, and 15 mGy). Images were reconstructed as conventional 120-kVp and VM images ranging from 40 to 100 keV in 10-keV increments with a slice thickness of 1 mm. In addition, 70-keV images were reconstructed at slice thicknesses of 2 and 5 mm. Length, width, and attenuation were measured at bone and soft tissue window settings by 2 radiologists. The reference stone size was measured with a digital caliper, and the reference attenuation was measured on 120 kVp images. Statistical analyses were performed using the paired t test, the Wilcoxon test, and repeated-measures one-way analysis of variance. RESULTS: The stone size and attenuation decreased with increasing energy level of the VM images, irrespective of window settings. The stone size on the VM 70- to 90-keV images under the soft tissue window and on the VM 40-keV images under the bone window was closest to the true size. Attenuation values measured on the VM 70-keV images were similar to those obtained on the conventional 120-kVp images across both window settings. In addition, measurements of stone size (length and width) and attenuation on VM images with thinner slices were closer to the true value. CONCLUSIONS: For VM images used to measure urinary stone size and attenuation, the optimal energy level required may differ based on the window settings. In addition, reconstructed thinner slice thickness can improve measurement accuracy.