RPS 1710 - Advanced CT imaging: PCCT, 4D CT, and metal artefacts

Author Block: Y. Zhou, Y. Guo, Z. Wang, L. Lei, X. Guo; Zhengzhou/CN
Purpose: This study aims to investigate the diagnostic value of contrast-enhanced photon-counting CT (PCCT) in differentiating between bone islands and osteoblastic metastases, with the goal of enhancing clinical diagnostic accuracy.
Methods or Background: This prospective study enrolled patients undergoing spectral contrast-enhanced CT between February and May 2025 who were diagnosed with either bone islands or osteoblastic metastases. Based on clinical data and pathological confirmation, 80 cases of bone islands and 80 cases of osteoblastic metastases were included. ROIs were delineated at the lesion sites, and spectral post-processing (SPP) datasets were reconstructed to obtain conventional mixed-energy CT images (CI), virtual monoenergetic images (VMI, 40-150 keV at 10 keV intervals), iodine density maps (ID), and spectral curves. A multiparametric spectral model combining iodine density and VMI (ID-VMI40-150 keV) was established. For each ROI, conventional CT values (CIHU), iodine density values (ID value), and the slope of the spectral curve (λHU) were measured. ROC curves were generated, and the AUC and cutoff values were calculated to compare the diagnostic performance of each spectral parameter.
Results or Findings: Compared with arterial phase CT values (AUC: 0.854; cutoff: 0.69), venous phase CT values (AUC: 0.861; cutoff: 0.702), arterial iodine density (AUC: 0881; cutoff: 0.792), and virtual monoenergetic images at 40 keV (AUC: 0.901; cutoff: 0.735), 50 keV (AUC: 0.899; cutoff: 0.634), and 60 keV (AUC: 0.883; cutoff: 0.725), the spectral multiparametric model ID-VMI60 keV demonstrated the highest diagnostic performance for distinguishing bone islands from osteoblastic metastases (AUC: 0.952; cutoff: 0.934; sensitivity: 97.4%; specificity: 100%).
Conclusion: Contrast PCCT ID-VMI60 keV combined models enhanced the differential diagnosis between BIs and OBMs compared to conventional CT parameters and individual VMI parameters and did not require the use of iodine contrast agents.
Limitations: Not applicable.
Funding for this study: the Key Scientific Research Project of Colleges and Universities in Henan Province (20B320047)
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: This study has been approved by the ethics committee.

Author Block: J. A. Pena¹, F. Thomsen², T. Damm¹, M. Frölich³, S. O. Schönberg³, C-C. Glüer¹, J-B. Hövener¹, M. Kachelrieß⁴, S. Sawall⁴; ¹Kiel/DE, ²Minden/DE, ³Mannheim/DE, ⁴Heidelberg/DE
Purpose: To evaluate the feasibility and accuracy of photon-counting computed tomography (PCCT) for assessing trabecular microstructure in human vertebrae at clinically realistic dose levels, while investigating the effects of radiation dose, spatial resolution, and reconstruction method.
Methods or Background: Seven excised human vertebrae were imaged with a PCCT system (Naeotom Alpha) in ultra-high-resolution mode. To reproduce clinically relevant intersection lengths, the vertebrae were placed inside a semi-anthropomorphic thorax phantom and scanned at 120 kV and dose levels between 6-20 mGy (CTDI32cm). Additional scans were performed without the phantom for high-dose reference measurements. Data were reconstructed with filtered backprojection (FBP) and quantum iterative reconstruction (QIR3) using three kernels of increasing resolution (Br56, Br76, Br89) and 0.2 mm slice thickness. Bone mineral density (BMD), bone volume fraction (BV/TV), trabecular separation (Tb.Sp), and trabecular thickness (Tb.Th) were quantified within standardized volumes of interest. Agreement with high-dose reference scans was evaluated using Wilcoxon signed-rank tests, Bland-Altman analysis, and Lin’s concordance correlation coefficient (rccc).
Results or Findings: BMD and BV/TV were comparable between FBP and QIR3 with no significant differences. Sharper kernels increased noise, particularly for FBP, which compromised microstructural accuracy. Br56 resolution was insufficient to resolve trabecular microstructure. Bland-Altman and rccc analyses showed Br76 with QIR3 achieving the best agreement to the high-dose reference. For Br76, rccc ranged between 0.66-0.94 for Tb.Sp across dose levels, compared with 0.05-0.58 for Br89. Similar trends were observed for other parameters.
Conclusion: PCCT enables reliable estimation of vertebral trabecular microstructure at diagnostic dose levels and clinically relevant intersection lengths. Br76 with QIR3 was the most robust setting across dose levels and offered a good balance between noise suppression and microstructural fidelity.
Limitations: Small sample size of cadaveric vertebrae.
Funding for this study: This study did not receive any funding.
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information:

Author Block: U. Bach, C. Obermüller, D. Cester, F. Ensle; Zürich/CH
Purpose: This study aimed to evaluate the performance of a novel gantry-based, multi-scan CBCT system for spinal imaging with complete anatomic coverage and compare it to energy integrating (EI)CT and photon counting (PC)CT using dose-matched protocols.
Methods or Background: An anthropomorphic torso phantom was used to simulate human anatomy. Gantry-based CBCT scans of the thoracolumbar spine were performed using different presets (low-dose, enhanced, best quality), while EICT and PCCT scans followed dose-matched clinical protocols. Qualitative image analysis was assessed by three blinded readers using a 4-point Likert scale, and quantitative analysis was conducted using global noise level (GNL) measurements.
Results or Findings: CBCT achieved diagnostic-quality imaging for the thoracic and lumbar spine, particularly with "best" and "enhanced" presets. Subjective image quality was highest for PCCT, followed by EICT and CBCT. CBCT demonstrated lower GNL than EICT, nearing PCCT levels. However, high radiation doses (5 mGy) were required for CBCT imaging of the upper thoracic spine (Th1–Th6) due to anatomical complexity, while low doses (0,5 mGy) sufficed for the lower thoracolumbar spine (Th7–S1).
Conclusion: Gantry-based CBCT generated diagnostic-quality images of large spinal regions at relatively low radiation doses, although the upper thoracic spine (above Th6) required higher doses. The overall subjective image quality remained below EICT and PCCT.
Limitations: Limitations of this study include its reliance on a single-sized phantom, which does not fully capture the complexity and variability of human anatomy and physiology. Although the phantom imitates the diverse tissue composition of the human body, it lacks pathological conditions such as fractures, and the dynamic factors present in clinical settings, including motion artifacts.
Funding for this study: None
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information:

Author Block: M. A. Spurek, J. P. Janssen, S. Skornitzke, N. Große Hokamp; Cologne/DE
Purpose: CT evaluation in patients with orthopaedic implants is often complicated by metal artefacts. Although methods such as iterative reconstruction and orthopaedic metal artefact reduction (O-MAR) have shown promising results, evidence on interaction effects with alloy density, acquisition parameters, and with each other, remains limited. This study aimed to systematically assess both main and interaction effects on metal artefact reduction.
Methods or Background: Cylindrical rods of nine implant alloys were scanned using a CT7500 and IQon (Philips) at 100/120/140 kVp and 2/5/10 CTDIvol. Conventional and iterative model reconstructions (IMR) were generated with and without O-MAR. Artefacts were assessed across 20 cross-sections using an objective quantitative approach based on Fourier analysis. Mixed-effects regression was used to quantify the independent and combined effects of physical alloy density, kVp, CTDIvol, reconstruction method, and O-MAR on metal artefacts.
Results or Findings: Artefact levels increased with higher alloy density. O-MAR, IMR, higher kVp, and higher CTDIvol substantially reduced artefacts, with O-MAR exerting the strongest and IMR the second-strongest effect. The artefact-increasing effect of alloy density was particularly reduced by O-MAR and higher kVp, while the interaction effects with both IMR and CTDIvol were small. Combining O-MAR and IMR with higher kVp yielded a small additional artefact reduction, whereas the artefact-reducing effect of higher CTDIvol was diminished by the combined use of O-MAR and IMR.
Conclusion: On standard scanners, O-MAR is the most, and IMR the second-most, effective tool for reducing metal artefacts, and both should be routinely applied. Higher kVp provides additional benefit, particularly in dense alloys, whereas increasing CTDIvol yields little extra improvement and should not be prioritised considering patient radiation exposure. These findings provide practical guidance for optimising CT protocols in patients with orthopaedic implants.
Limitations: Results should be verified in vivo.
Funding for this study: No funding was received for this study.
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information:

Author Block: X. Liu, J. Li; Zhengzhou/CN
Purpose: To explore the potential of combining the iterative metal artifact reduction (iMAR) algorithm with cinematic volume rendering technique (cVRT) in photon-counting CT for assessing hip replacements.
Methods or Background: A retrospective study was conducted on 120 patients who underwent hip arthroplasty exams using the photon-counting CT scanner (NAEOTOM Alpha). Reconstruction of CT images employing conventional methods, volume rendering (VR), and cVRT, both with and without iMAR. Measurements of CT numbers and standard deviations (SDs) in regions of interest (ROIs) were obtained. Objective image quality and subjective scores were assessed using established scales. Statistical analyses included paired T tests, Mann-Whitney U tests, and Kappa tests.
Results or Findings: Compared with the non-iMAR group, the iMAR group showed significantly decreased and increased CT numbers in hyperattenuating and hypoattenuating areas, respectively, as well as lowered artifact and image noise (p<.001). Qualitatively, the iMAR group showed superiority to the non-iMAR group in both image quality and diagnostic confidence, with scores increases of 2.70 and 2.88 points, respectively (P<0.05). iMAR combined with cVRT received the highest subjective score (P<0.05) among the four series of post-processing images, followed by iMAR with VR images(P<0.05), cVRT and VR images in the non-iMAR group both received the lowest scores.
Conclusion: The iMAR algorithm in photon-counting CT effectively reduces artifacts and image noise, enhancing both image quality and diagnostic confidence in post-hip metal replacement assessments. When combined with cVRT, it provides a more intuitive visualization of metal implant stability and the relationship between implants and adjacent tissues.
Limitations: We focused exclusively on patients undergoing unilateral total hip replacement surgery, without differentiation by metal type or specific surgical approaches, leaving the effects of these variables on artifacts unexplored
Funding for this study: Received financial support from Medical Science and Technology project of Henan Province (No.LHGJ20240327).
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: The ethics committee notification can be found under the number 2024-KS-HNSR115

Author Block: Y. J. Jiang, Y. Wu, J. Gao; Zhengzhou/CN
Purpose: This study aimed to evaluate the clinical utility of photon-counting CT (PCCT) virtual noncalcium (VNCa) technology in diagnosing osteoporosis, using quantitative CT (QCT)-measured bone mineral density (BMD) as the reference standard.
Methods or Background: In this cross-sectional study, retrospective data from 445 patients were analyzed. All participants underwent PCCT scans encompassing the L1–L3 vertebral levels, with concurrent QCT BMD measurements obtained using dedicated software. Pearson correlation analysis was performed to assess the relationship between decalcified density and BMD. Multivariable linear regression was used to derive a predictive equation for BMD.
Results or Findings: A calcium ratio of 1.57 was established for decalcified density measurement. Multivariable linear regression revealed a significant model (F = 749.2, p < 0.0001), with calcium density (β = 24.68, p < 0.0001) as a significant positive predictor of BMD, while age (β = -0.4297, p < 0.0001) and BMI (β = -0.4354, p = 0.0341) were negative predictors. Sex (β = -0.4139, p = 0.792) did not significantly influence BMD. The regression equation was:
BMD = 24.68 × Decalcified Density - 0.4297 × Age - 0.4139 × Sex - 0.4354 × BMI + 6.556. The area under the ROC curve (AUC) for decalcified density in diagnosing osteoporosis was 0.9917 (95% CI: 0.9856–0.9997, p < 0.0001). At a decalcified density threshold of <4.685 mg/cm³, sensitivity and specificity were 97.61% and 96.3%, respectively.
Conclusion: Dual-energy CT VNCa technology can serve as an effective alternative for quantifying bone mineral content in osteoporosis diagnosis.
Limitations: The cross-sectional design cannot assess the relationship between dynamic changes in calcium density and the evolution of BMD. manufacturers.
Funding for this study: None
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: This study was a single-center cross-sectional investigation approved by the Ethics Review Committee of the First Affiliated Hospital of Zhengzhou University (Ethical Approval No.: 2021-KY-1222-002).

Author Block: K. Wen¹, Y. Zhu¹, g. fan¹, B. Zhang¹, Y. Ling², X. Liang², B. Li¹, H. Li¹, K. Zhang¹; ¹Shenyang/CN, ²Shanghai/CN
Purpose: To explore the feasibility of evaluating knee trabecular microstructure in vivo using photon-counting CT (PCCT) with automated segmentation and quantitative analysis.
Methods or Background: Changes of knee trabecular microstructure are associated with the onset and progression of bone diseases such as osteoporosis and osteoarthritis. High-resolution peripheral quantitative CT (HR-pQCT) has been used to quantify trabecular microstructure in vivo, but is limited by a relatively small field of view. PCCT enables superior spatial resolution and dose efficiency for whole-body imaging. Seven volunteers (age range 30-69 years) were recruited in this preliminary study for PCCT scans (P10, Neusoft Medical Systems; 104µm @2%MTF and 120 kV) of both knees. Images were reconstructed with a 1024 matrix and slice thickness of 274 μm. Subchondral trabecular bone of femoral and tibial condyle was automatically segmented using a ResUNet-based framework. Trabecular parameters, including bone volume fraction (BV/TV), trabecular thickness (Tb.Th), trabecular separation (Tb.Sp), trabecular number (Tb.N), and synthetic images (SMI), were computed via the NeuSpace platform (Neusoft Medical Systems) and compared with values previously reported in cadaveric PCCT studies.
Results or Findings: The automated pipeline successfully segmented and quantified in vivo subchondral trabecular microstructure. Compared to cadaveric literature of knee PCCT, our in vivo cohort exhibited a 65% higher bone volume fraction (BV/TV: 48.90±8.74 % vs. 29.61±11.06 %), 74% thicker trabeculae (Tb.Th: 0.75±0.08 mm vs. 0.43±0.09 mm), and more plate-like structure (SMI: 0.68±0.26 vs. 1.64±0.74).
Conclusion: This study provides preliminary subchondral trabecular evidence in vivo using PCCT, and demonstrates the feasibility of automated segmentation and quantitative analysis with NeuSpace, suggesting potential applications for osteoporosis and osteoarthritis assessment.
Limitations: This is a preliminary study with a small sample size. Larger-scale studies are needed to confirm the clinical value.
Funding for this study: No funding was received for this study.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: This study was approved by the Institutional Ethical Committee (2025QL004).

Author Block: H. Vries¹, B. van der Heijden¹, S. Hummelink¹, J. G. G. Dobbe², I. Sechopoulos¹, G. J. Streekstra²; ¹Nijmegen/NL, ²Amsterdam/NL
Purpose: To assess how motion artifacts influence registration accuracy across acquisition and reconstruction protocols of different CT systems in 4D CT images of a rotating wrist phantom.
Methods or Background: A rotating wrist phantom with three 3D-printed bones (scaphoid, lunate, capitate) was scanned on four CT systems (two vendors; single- and dual-source). One static 3D scan and multiple axial 4D scans were acquired at different phantom rotation speeds, each lasting 10s. Single-source covered 0.5–3 cycles, and dual-source 1–6 cycles (one cycle represents radial–ulnar–radial; 0.5 cycle represents radial–ulnar). Dose dependence was evaluated on two systems (80 kV/40 mA and 120 kV/100 mA). Images were reconstructed in full and partial modes, segmented, and registered using point-to-image registration. Registration accuracy was calculated as translation and rotation errors of the scaphoid and capitate relative to the lunate, referenced to the static scan, and reported as median with interquartile range per wrist cycle.
Results or Findings: Registration accuracy was unaffected by dose. For one cycle, the errors of the best single-source system were 0.29(0.21–0.36) mm and 1.37(0.82–2.18) degrees for full and 0.21(0.16–0.25) mm and 0.70(0.52–0.95) degrees for partial reconstructions for the capitate; the dual-source system showed the highest accuracy (errors of 0.14(0.12 – 0.17) mm and 0.48(0.39 – 0.62) degrees). For all systems with partial reconstructions, the translation errors remained below the voxel size (0.28mm) for ≤one cycle per 10s.
Conclusion: Dynamic 4DCT of the wrist has shown promise, but the impact of motion artifacts on bone-to-bone measurements needs to be understood. Partial reconstructions kept registration errors below the pixel size for wrist rotations up to one cycle per 10s, while dual-source scans reached similar accuracy at higher speeds.
Limitations: Limited to a phantom with 3D-printed bones and not yet validated in clinical setting.
Funding for this study: Funding was provided by NWO
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information: