Research Presentation Session: Cardiac

RPS 1203 - Spectral CT: new developments and clinical applications (part 1)

March 1, 08:00 - 09:00 CET

7 min
Iodine quantification with a CdZnTe clinical prototype photon-counting scanner at reduced radiation dose: initial cardiac phantom results and reproducibility analysis
Amir Pourmorteza, Atlanta, GA / United States
Author Block: T. W. Holmes1, S. Sharma2, S. Ross2, T. Labno2, R. Zhang2, S. Wu2, Z. Yu2, R. Thompson2, A. Pourmorteza1; 1Atlanta, GA/US, 2Vernon Hills, IL/US
Purpose: This study aimed to assess effectiveness of a clinical prototype CT scanner with CdZnTe-based photon-counting detector (PCD) CT technology in spectral mode, specifically regarding the measurement of iodinated contrast agents at both standard and reduced radiation dosage levels.
Methods or Background: We 3D-printed a series of phantoms mimicking coronary arteries with lumen diameter of 3.5 mm with stents, soft plaque, and hard plaques placed in them. The phantoms were placed inside two water tanks with 12 cm diameters. The water tanks were scanned on the prototype scanner at 120 kVp and 200, 100, and 50 mAs exposures in 6-bin spectral mode. The 50 mAs scan was repeated twice to assess reproducibility. We measured accuracy, stability with respect to radiation dose, and reproducibility of iodine through Bland-Altman analysis of values measured in circular regions of interest (ROIs) ranging from 2.2 mm to 10 mm in diameter.
Results or Findings: We measured very small bias of -0.15 and good 95% confidence interval (CI) [-1.52 1.21] for concentrations of iodine ranging from 0 to 35 compared to calibrated values (all values are reported as mg I/mL). At 50% radiation dose reduction we observed bias of 0.35 with CI=[-0.32 1.04] compared to the high-dose (200 mAs) measurements. At 75% dose reduction the bias and CI were 0.45 and [-0.49 1.44]. The 75% dose-reduced iodine maps showed good reproducibility with bias of 0.01 mg I/mL and confidence interval of [-0.66 0.69].
Conclusion: Accurate quantification of iodine is an important tool in characterisation of coronary artery plaques as well as for making detailed measurements of myocardial perfusion. PCD-CT shows potential for making accurate iodine measurements in dose-reduced settings.
Limitations: Limitations of this study were: this was a limited pilot study with 2 phantoms; more experiments mimicking different patient sizes are warranted.
Funding for this study: Funding was received through a sponsored research agreement with Canon Medical Research USA.
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information: No information provided by the submitter.
7 min
Impact of photon-counting detector CT-based calcium removal algorithm on coronary stenosis grading and CT-FFR values
Zsofia Jokkel, Budapest / Hungary
Author Block: Z. Jokkel1, B. Vattay1, M. Boussoussou1, M. Vecsey-Nagy1, M. Kolossvary1, M. Kiss2, B. Sipos1, P. Maurovich-Horvat1, B. Szilveszter1; 1Budapest/HU, 2Forcheim/DE
Purpose: CT-based fractional flow reserve (CT-FFR) provides non-invasive evaluation of lesion specific ischaemia. However, blooming artefacts from heavily calcified plaques could hinder the assessment of luminal stenosis and alter CT-FFR values. The objective of this study was to investigate the effect of a calcium removal algorithm on the grading of coronary stenoses and CT-FFR values in comparison to measurements based on the standard image reconstruction used for clinical evaluation.
Methods or Background: We enrolled consecutive patients with calcified plaques and intermediate coronary artery stenosis (30-90%) who underwent a dual-source photon counting CT (NAEOTOM Alpha, Siemens Healthineers) for suspected coronary artery disease. A total of 54 lesions were analysed and two reconstructions were used (standard: 0.4 mm slice thickness, Bv40, 70keV; PureLumen calcium removal algorithm: 0.8mm slice thickness, Qr40, 70keV). Anatomical landmarks were used as fiducial markers to ensure that the same location was used for CT-FFR assessment. Area-stenosis and CT-FFR values distal to the plaques were measured using syngo.via Frontier CT-cFFR software (Siemens Healthineers). Comparison of stenosis and CT-FFR values was performed using Wilcoxon signed rank test.
Results or Findings: A total of 25 patients (mean age 64.5±8.1years, 24% female) were included in the study. Median calcium score and Q1-Q3: 258 (154.35-401.75). Mean area stenosis of the 54 lesions was 50.5±22.5% on standard reconstruction compared to 43.5±20.9% on PureLumen images (p<0.001). CT-FFR values were higher for PureLumen reconstructed lesions compared to standard reconstruction: 0.87±0.11 vs 0.82±0.15 (p<0.001).
Conclusion: PureLumen calcium removal technique might allow for more accurate assessment of the degree of stenosis by removing calcium blooming artefacts. In patients with intermediate stenosis, this resulted in lower stenosis grades and higher CT-FFR values, which may affect further therapeutic decisions.
Limitations: Correlation with invasive coronarography is warranted and will be available in a subpopulation.
Funding for this study: Funding was provided by the János Bolyai Research Scholarship of the Hungarian Academy of Sciences. Project number RRF-2.3.1-21-2022-00003 has been implemented with the support provided by the European Union.
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: This study was approved by the Scientific and Research Ethics Committee of the Hungarian Medical Research Council (IV/665-3/2022/EKU) and was carried out in accordance with the tenets of the Declaration of Helsinki.
7 min
Stenosis quantification on virtual non-calcium images from photon-counting detector CT
Victor Mergen, Zurich / Switzerland
Author Block: V. Mergen1, S. Rusek2, F. Civaia2, P. Rossi2, R. Rajagopal3, R. Manka1, A. Candreva1, M. Eberhard1, H. Alkadhi1; 1Zurich/CH, 2Monaco/MC, 3Jodhpur/IN
Purpose: This study aimed to evaluate, in patients with known or suspected coronary artery disease, the feasibility and accuracy of quantification of calcified coronary stenoses using virtual non-calcium (VNCa) images in CCTA with photon-counting detector (PCD) CT compared with quantitative coronary angiography (QCA).
Methods or Background: This retrospective, institutional review board approved study included consecutive patients with calcified coronary artery plaques undergoing CCTA with PCD-CT and invasive coronary angiography between July and December 2022. Virtual monoenergetic images (VMI) and VNCa images were reconstructed. Diameter stenoses were quantified on VMI and VNCa images by two readers. Stenosis measurements from 3D-QCA served as the standard of reference. Measurements were compared using Bland-Altman analyses, Wilcoxon tests, intraclass correlation coefficients (ICC), and weighted-Kappa analysis.
Results or Findings: Thirty patients (mean age, 64 years ± 8 [standard deviation]; 26 men) with 81 coronary stenoses from calcified plaques were included. Ten of the 81 stenoses (12%) had to be excluded because of erroneous plaque subtraction on VNCa images. Median diameter stenosis determined on 3D-QCA was 22% (interquartile range, 11-35%; total range, 4-88%). As compared with 3D-QCA, VMI overestimated diameter stenoses (mean differences -7%, p<.001, ICC:.84 and -10%, p<.001, ICC:.87 for reader 1 and 2, respectively), whereas VNCa images showed similar diameter stenoses (mean differences 1%, p=.07, ICC:.93 and 0%, p=.68, ICC:.94 for reader 1 and 2, respectively).
Conclusion: This preliminary experience suggests that virtual calcium removal in CCTA with PCD-CT is feasible in a high proportion of calcified, minimal to moderate stenoses and has the potential to improve the quantification of coronary stenoses.
Limitations: First, this single centre retrospective study included only a limited number of patients and stenoses. Second, only a limited number of moderate and severe stenoses were present.
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: No information provided by the submitter.
7 min
Improvement of the visualisation of coronary stents by ultra-high-resolution photon-counting detector CT
Le Qin, Shanghai / China
Author Block: L. Qin, S. Zhou, F. Yan, W. Yang; Shanghai/CN
Purpose: This study aimed to investigate the quantitative and qualitative characteristics of coronary stents of ultra-high-resolution coronary CT angiography (CCTA) with different reconstruction kernels on photon counting detector-CT (PCD-CT) in patients after percutaneous coronary intervention (PCI).
Methods or Background: From July 2023 to September 2023, 68 patients (132 stents) after PCI were enrolled to undergo CCTA on a dual-source PCD-CT system. Standard resolution images with thickness of 0.6 mm, increment of 0.4 mm and kernel of Bv48, and UHR images with thickness and increment of 0.2 mm and kernel of Bv48, Bv56, Bv60, Bv64, Bv72 and Bv76 were reconstructed. CT attenuation and noise were measured in the aorta root and within the stents. Stent and in-stent diameters were also assessed. Subjective image quality was evaluated by a Likert-5 point scale.
Results or Findings: Image noise significantly increased with the reduction of image thickness and elevation of kernels (2.6±5.7~136.3±18.5 HU, P<0.001). UHR images with Bv72 and Bv76 had the smallest differences between aorta and in-stent CT values (14.8±68.3 HU, 6.2±73.9 HU, P<0.001), and differences between stent and in-stent diameters (1.2±0.2 mm, 1.2±0.2 mm, P<0.001). UHR images with Bv72 had the largest in-stent lumen diameter (2.3±0.5 mm, P<0.001) and the smallest differences between stent diameter and nominal diameter (0.4±0.3 mm, P<0.001) compared to other images. Subjective analysis showed that images with Bv72 had the most superior effect of blooming artefact reduction (5 [5, 5], P<0.001), and in-stent lumen and stent demonstration (5 [5, 5], P<0.001). Bv72 also had the highest diagnostic confidence (5 [5, 5], P<0.001).
Conclusion: UHR CCTA on PCD-CT results in the significantly improved visualisation of coronary stents and Bv72 is the optimal reconstruction kernel to show the stent struts and in-stent lumen.
Limitations: An identified limitation was that the in-stent diameter and diagnostic accuracy for in-stent re-stenosis were not compared with the quantitative invasive coronary angiography.
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 Ruijin hospital ethics committee, Shanghai Jiaotong University School of Medicine.
7 min
Photon-counting CT angiography in pre-TAVR aortic annulus sizing: high-pitch spectral vs low-pitch ultra-high-resolution CT-angiography
Muhammad Taha Hagar, Freiburg im Breisgau / Germany
Author Block: M. T. Hagar1, T. Jennerjahn1, D. Westermann1, S. Faby2, C. Von Zur Mühlen1, C. L. Schlett1, F. Bamberg1, T. Krauß1, M. Soschynski1; 1Freiburg im Breisgau/DE, 2Forchheim/DE
Purpose: The study aimed to assess the diagnostic efficacy of low-pitch, retrospective ultra-high-resolution CT-angiography (UHR-CTA) and high-pitch, prospective spectral CTA (HPS-CTA) utilising a first-generation, dual-source photon-counting detector CT scanner in preprocedural planning for transcatheter aortic valve replacement (TAVR).
Methods or Background: Consecutive patients with severe aortic valve stenosis were referred clinically, underwent UHR-CTA (collimation: 120 x 0.2 mm) and HPS-CTA (144 x 0.4 mm with full spectral information) for TAVR planning and were retrospectively enrolled. Radiation doses were extracted from CT reports. Two radiologists independently evaluated UHR-CTA and HPS-CTA datasets, analysing aortic annulus image quality (on a 4-point scale) and measuring the aortic annulus area (AAA) and perimeter (AAP). These measurements then informed hypothetical valve prosthesis sizing according to vendor recommendations.
Results or Findings: Among a total of 64 patients (mean age 81, ±7 SD; 28 women), HPS-CTA was associated with lower radiation exposure (4.1 mSv) than UHR-CTA (12.6 mSv, p<0.001, while UHR-CTA exhibited superior image quality (median score: 1, IQR: 1–2 vs. 2, IQR: 2–3; p<0.001). The AAA and AAP assessments from both CTA datasets were strongly correlated (Pearson r2 = 0.857), leading to consistent valve prosthesis sizing in 89% of patients. However, those with lower image quality on HPS-CTA (score ≥3) more frequently encountered varied sizing recommendations in hypothetical aortic valve prosthesis selection.
Conclusion: UHR-CTA and HPS-CTA by photon-counting CT technology provide reliable aortic annular evaluations for TAVR planning. While UHR-CTA provides enhanced image quality, HPS-CTA features lower radiation exposure. However, impaired image quality in HPS-CTA may necessitate additional UHR-CTA to avoid valve sizing discrepancies.
Limitations: Identified limitations were: (1) the study's generalisability may be limited by its focus on primarily older patients, (2) this was a single-centre study with a retrospective design, and (3) hypothetical valve prosthesis sizing based on aortic annular measurements might overlook other patient-specific factors in real-world decision-making.
Funding for this study: Funding was received from the Ministerium für Wirtschaft, Arbeit und Wohnungsbau Baden-Württemberg (35-4223.10/20).
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Informed written consent was obtained from all patients. This study was approved with the approval code: 21–2469; approval date: 21-09-2021.
7 min
First clinical experience with a novel reconstruction algorithm to reduce stair-step artefacts in step-and-shoot cardiac photon-counting detector CT
Lukas Jakob Moser, Zürich / Switzerland
Author Block: L. J. Moser1, V. Mergen1, M. Eberhard2, H. Alkadhi1; 1Zurich/CH, 2Unterseen/CH
Purpose: One of the most common cardiac CT scan modes is prospective ECG-triggering (or step-and-shoot (SAS)). Due to the stack-wise acquisition over usually several heart cycles, SAS may be associated with "stair-step" artefacts in the transitional areas, where stacks are misaligned or overlapping due to irregular heart rates. This study aimed to evaluate a novel cardiac imaging reconstruction algorithm designed for photon-counting detector CT to reduce the occurrence and severity of such artefacts.
Methods or Background: In this clinical study, we included 50 consecutive patients who underwent cardiac CT on a clinical dual-source photon-counting detector CT scanner in the SAS mode. Each scan was reconstructed without and with the ZeeFree algorithm, which uses overlapping information to perform a non-rigid registration between the borders of two adjacent sub-volumes respectively, to minimise potentially occurring stair-step artefacts. The presence or absence and, if present, the extent of stair-step artefacts were rated on a visual analogue four-point scale.
Results or Findings: Forty of the total 800 coronary segments (5%) had stair-step artefacts, from which 12 (30%) led to a non-diagnostic image quality (1.5% of all segments). The novel reconstruction algorithm significantly reduced the number and extent of these artefacts compared to standard reconstructions (median score ZeeFree: 1; median score standard reconstruction: 3; p<0.001). From the initially rated 12 non-diagnostic segments in standard reconstruction, 9 (75%) improved to a diagnostic image quality using the algorithm.
Conclusion: Our results demonstrate the feasibility and effectiveness of a novel reconstruction algorithm, which significantly reduces stair-step artefacts in coronary CT angiography with dual-source photon-counting detector CT acquired in the SAS mode.
Limitations: Identified limitations were (1) that this was a single centre experience and (2) the effect on diagnostic accuracy was not evaluated.
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: This was a retrospective study design and a waiver of written informed consent was obtained.

This session will not be streamed, nor will it be available on-demand!