Scholarlink

Objectives <p>This study aimed to assess the impact of the photon-counting detector (PCD)-CT-based virtual monoenergetic image (VMI) reconstruction keV levels on CT-based fractional flow reserve (CT-FFR), compared to the energy-integrating detector (EID)-CT.</p> Methods <p>Patients undergoing clinically indicated coronary CT angiography (CCTA) on an EID-CT were prospectively enrolled for a research CCTA on a PCD-CT within 30 days. PCD-CT datasets were reconstructed at VMI levels of 45, 55, 70, and 90 keV. CT-FFR was obtained semiautomatically using an on-site machine learning algorithm by two readers. CT-FFR ≤ 0.80 was considered hemodynamically significant.</p> Results <p>A total of 20 patients (63.3 ± 8.8 years; 13 men (65%) were included. Median CT-FFR values in the per-vessel analysis for PCD-CT scans were 0.86 (0.81–0.92) for 45 keV, 0.87 (0.80–0.93) for 55 keV, 0.85 (0.79–0.92) for 70 keV and 0.82 (0.76–0.89) for 90 keV, and 0.86 (0.71–0.93) for EID-CT. Comparison among different VMIs showed significant differences only for 45 vs. 90 keV (<i>p</i> &lt; 0.001), and 55 vs. 90 keV (<i>p</i> &lt; 0.001). No significant differences were found in the pairwise comparison between any VMI and EID-CT (all <i>p</i> &gt; 0.05). PCD-CT at 70 keV showed the highest correlation (<i>r</i> = 0.83, <i>p</i> &lt; 0.001), agreement (ICC: 0.90 (0.84–0.94)), and the lowest bias (mean bias −0.01; limits of agreement, 0.84/0.94) when compared to EID-CT.</p> Conclusion <p>VMI reconstructions showed significant influence on CT-FFR values only at the extreme levels of the spectrum, while no significant differences were found in comparison with EID-CT. VMI at 70 keV demonstrates the highest correlation and agreement, with the lowest bias compared to EID-CT.</p> Critical relevance statement <p>Evidence on novel spectral photon-counting detector (PCD)-CT’s impact on CT-fractional flow reserve (FFR) is limited; our results demonstrate the feasibility of CT-FFR using PCD-CT, showing no significant differences between various virtual monoenergetic images and energy-integrating detector (EID)-CT values</p> Key Points <p><UnorderedList Mark="Bullet"> <ItemContent> <p>The impact of spectral photon-counting detector (PCD)-CT on CT-derived fractional flow reserve (CT-FFR) is unclear.</p> </ItemContent> <ItemContent> <p>Spectral PCD-CT-based CT-FFR is feasible, differing only at extreme virtual monoenergetic image levels.</p> </ItemContent> <ItemContent> <p>CT-FFR from PCD-CT at 70 keV showed the strongest correlation with energy-integrating detector-CT.</p> </ItemContent> </UnorderedList></p> Graphical Abstract <p></p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Influence of virtual monoenergetic reconstructions on coronary CT angiography-based fractional flow reserve with photon-counting detector CT: intra-individual comparison with energy-integrating detector CT

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年2月17日

查看全文
  • Giuseppe Tremamunno,
  • Daniel Pinos,
  • Emese Zsarnoczay,
  • U. Joseph Schoepf,
  • Milan Vecsey-Nagy,
  • Chiara Gnasso,
  • Nicola Fink,
  • Dmitrij Kravchenko,
  • Muhammad Taha Hagar,
  • Joseph Griffith III,
  • Jim O’Doherty,
  • Andrea Laghi,
  • Tilman Emrich,
  • Akos Varga-Szemes

摘要

Objectives

This study aimed to assess the impact of the photon-counting detector (PCD)-CT-based virtual monoenergetic image (VMI) reconstruction keV levels on CT-based fractional flow reserve (CT-FFR), compared to the energy-integrating detector (EID)-CT.

Methods

Patients undergoing clinically indicated coronary CT angiography (CCTA) on an EID-CT were prospectively enrolled for a research CCTA on a PCD-CT within 30 days. PCD-CT datasets were reconstructed at VMI levels of 45, 55, 70, and 90 keV. CT-FFR was obtained semiautomatically using an on-site machine learning algorithm by two readers. CT-FFR ≤ 0.80 was considered hemodynamically significant.

Results

A total of 20 patients (63.3 ± 8.8 years; 13 men (65%) were included. Median CT-FFR values in the per-vessel analysis for PCD-CT scans were 0.86 (0.81–0.92) for 45 keV, 0.87 (0.80–0.93) for 55 keV, 0.85 (0.79–0.92) for 70 keV and 0.82 (0.76–0.89) for 90 keV, and 0.86 (0.71–0.93) for EID-CT. Comparison among different VMIs showed significant differences only for 45 vs. 90 keV (p < 0.001), and 55 vs. 90 keV (p < 0.001). No significant differences were found in the pairwise comparison between any VMI and EID-CT (all p > 0.05). PCD-CT at 70 keV showed the highest correlation (r = 0.83, p < 0.001), agreement (ICC: 0.90 (0.84–0.94)), and the lowest bias (mean bias −0.01; limits of agreement, 0.84/0.94) when compared to EID-CT.

Conclusion

VMI reconstructions showed significant influence on CT-FFR values only at the extreme levels of the spectrum, while no significant differences were found in comparison with EID-CT. VMI at 70 keV demonstrates the highest correlation and agreement, with the lowest bias compared to EID-CT.

Critical relevance statement

Evidence on novel spectral photon-counting detector (PCD)-CT’s impact on CT-fractional flow reserve (FFR) is limited; our results demonstrate the feasibility of CT-FFR using PCD-CT, showing no significant differences between various virtual monoenergetic images and energy-integrating detector (EID)-CT values

Key Points

The impact of spectral photon-counting detector (PCD)-CT on CT-derived fractional flow reserve (CT-FFR) is unclear.

Spectral PCD-CT-based CT-FFR is feasible, differing only at extreme virtual monoenergetic image levels.

CT-FFR from PCD-CT at 70 keV showed the strongest correlation with energy-integrating detector-CT.

Graphical Abstract