Speaker
Description
Introduction: Low-dose CT is essential for lung cancer screening, but reducing the dose increases noise and may impair the detection of ground-glass opacity (GGO) nodules [1,2]. Ag and Cu filters modify the beam spectrum, affecting the dose–noise relationship, yet their impact on the spatial distribution of noise texture is poorly understood [3]. The noise power spectrum (NPS) describes this distribution as a function of spatial frequency [4]. This study compared the NPS of two low-dose CT protocols using Ag and Cu filters in an anthropomorphic chest phantom containing a GGO nodule. The hypothesis was that the lower-dose Ag-filter protocol increases the NPS magnitude without substantially altering its spectral shape, thereby preserving noise texture.
Materials and methods:An anthropomorphic chest phantom with a GGO nodule was scanned on an Aquilion One CT using two low-dose protocols: Ag (120 kV, 60 mAs, CTDIvol 0.5 mGy) and Cu (120 kV, 84 mAs, CTDIvol 1.2 mGy), both with AICE LUNG reconstruction. Images were processed in IMQuest.[6] Four 32×32-pixel ROIs were placed adjacent to the nodule, with 10 acquisitions per protocol. Noise magnitude and NPS were calculated via 2D FFT with radial averaging. Values were expressed as median ± SD, and compared using the paired t-test (p < 0.05).
Results: The Ag-filter protocol presented significantly higher mean noise than the Cu protocol (36.28 ± 0.86 vs. 25.52 ± 0.93 HU; p < 0.001), representing a 42.2% noise increase for a 58.3% dose reduction (CTDIvol 0.5 vs. 1.2 mGy). Peak NPS was 103.8% higher in the Ag protocol (1,302.68 vs. 639.17 HU²·mm²). However, peak frequency (p = 0.482) and mean frequency (p = 0.505) did not differ significantly, suggesting preservation of noise texture.
Conclusions: The lower-dose Ag-filter protocol (0.5 mGy) increased noise magnitude and NPS amplitude compared with the Cu-filter protocol (1.2 mGy), while preserving the spectral distribution of the NPS, as evidenced by the lack of significant differences in peak and mean frequencies (p > 0.05). This suggests that the noise increase occurs in a similar manner across the frequency spectrum, likely related to the reduced photon fluence associated with the lower-dose protocol, rather than to changes in noise texture induced by the filter material. These findings indicate that the Ag filter may represent a potential alternative for dose reduction in low-dose CT protocols for GGO nodule evaluation, although further studies with clinical images and larger nodule samples are necessary to confirm its diagnostic applicability.
Figure 1: Median NPS curves for Cu (1.2 mGy) and Ag (0.5 mGy) filters. Error bars represent the interquartile range (IQR) of the 10 measurements.
[1] Godoy MC, Naidich DP. Subsolid pulmonary nodules and the spectrum of peripheral adenocarcinomas of the lung: recommended interim guidelines for assessment and management. Radiology. 2009;253(3):606-622.
[2] Henschke CI, et al. CT screening for lung cancer:frequency and significance of part-solid and nonsolid nodules. AJR Am J Roentgenol. 2002;178(5):1053-1057.
[3] Travis WD, et al. International Association for the Study of Lung Cancer/American Thoracic Society/European Respiratory Society international multidisciplinary classification of lung adenocarcinoma. J Thorac Oncol. 2011;6(2):244-285.
[4] National Lung Screening Trial Research Team. Reduced lung-cancer mortality with low-dose computed tomographic screening. N Engl J Med. 2011;365(5):395-409.
[5] Costa PR, Boiset GR, Pimenta EB, et al. Hybrid phantom for lung CT: design and validation. Med Phys. 2025;52(8):e17990. doi:10.1002/mp.17990.
[6] Solomon J, et al. IMQuest [computer program]. Version 7.3. Durham (NC): Duke University Clinical Imaging Physics Group; 2023. Available from:railabs / SameiResearchGroup / imquest · GitLab
| Keywords | Low-dose CT, Anthropomorphic chest phantom,GGO nodule, NPS |
|---|---|
| Proceedings Publication Intent | Yes |
| Student Award Eligibility | yes |
| Financial Support Request | no |