Toshiaki Nishii1, Yoshinori Funama2, Shingo Kato3, Tae Iwasawa4, Naofumi Yasuda4, Yoichiro Ota4, Hironori Kawagoe4, Seitaro Oda5, Takashi Tsutsumi6, Daisuke Utsunomiya7. 1. Department of Diagnostic Radiology, Yokohama City University Medical Center, Yokohama City University Graduate School of Medicine, Yokohama, Japan. 2. Department of Medical Radiation Sciences, Faculty of Life Sciences, Kumamoto University, Kumamoto, Japan. 3. Department of Diagnostic Radiology, Yokohama City University Graduate School of Medicine, 3-9, Fukuura, Kanazawa-ku, Yokohama-shi, Kanagawa, Japan. sk513@yokohama-cu.ac.jp. 4. Department of Radiology, Kanagawa Cardiovascular and Respiratory Center, Kanagawa, Japan. 5. Department of Diagnostic Radiology, Faculty of Life Sciences, Kumamoto University, Kumamoto, Japan. 6. Research and Development Center, Canon Medical Systems Corporation, Tochigi, Japan. 7. Department of Diagnostic Radiology, Yokohama City University Graduate School of Medicine, 3-9, Fukuura, Kanazawa-ku, Yokohama-shi, Kanagawa, Japan.
Abstract
PURPOSE: The purposes of this experimental study were to compare the quantitative and qualitative visibility of in-stent restenosis between conventional-resolution CT (CRCT) and ultra-high-resolution CT (U-HRCT) and to investigate the effects of the image reconstruction techniques on the visualization of in-stent restenosis. MATERIALS AND METHODS: A vessel tube with non-calcified plaque in a 3.0-mm stent was scanned by using CRCT and U-HRCT at 4 stent directions (0, 30, 60, and 90 degrees) to the through-plane direction. Hybrid iterative reconstruction (HIR); model-based iterative reconstruction (MBIR); deep-learning-based reconstruction (DLR) were used as reconstruction methods. The lumen size was assessed using the full width at half maximum method, and image quality was visually evaluated using 4-point scale. RESULTS: U-HRCT had the significantly wider lumen sizes and narrower stent strut thickness than CRCT in three types of the reconstruction methods (P < 0.01). The lumen sizes for U-HRCT with 90 degrees were narrower than those with the other angle directions regardless of the reconstruction methods. Visual score was significantly higher for U-HRCT than CRCT (3.2 ± 0.7 vs 2.0 ± 0.4, P < 0.001). CONCLUSIONS: U-HRCT quantitatively and qualitatively provided better visualization of in-stent restenosis compared to CRCT. Image quality of U-HRCT may be affected by stent angle.
PURPOSE: The purposes of this experimental study were to compare the quantitative and qualitative visibility of in-stent restenosis between conventional-resolution CT (CRCT) and ultra-high-resolution CT (U-HRCT) and to investigate the effects of the image reconstruction techniques on the visualization of in-stent restenosis. MATERIALS AND METHODS: A vessel tube with non-calcified plaque in a 3.0-mm stent was scanned by using CRCT and U-HRCT at 4 stent directions (0, 30, 60, and 90 degrees) to the through-plane direction. Hybrid iterative reconstruction (HIR); model-based iterative reconstruction (MBIR); deep-learning-based reconstruction (DLR) were used as reconstruction methods. The lumen size was assessed using the full width at half maximum method, and image quality was visually evaluated using 4-point scale. RESULTS: U-HRCT had the significantly wider lumen sizes and narrower stent strut thickness than CRCT in three types of the reconstruction methods (P < 0.01). The lumen sizes for U-HRCT with 90 degrees were narrower than those with the other angle directions regardless of the reconstruction methods. Visual score was significantly higher for U-HRCT than CRCT (3.2 ± 0.7 vs 2.0 ± 0.4, P < 0.001). CONCLUSIONS: U-HRCT quantitatively and qualitatively provided better visualization of in-stent restenosis compared to CRCT. Image quality of U-HRCT may be affected by stent angle.
Authors: Jagat Narula; Y Chandrashekhar; Amir Ahmadi; Suhny Abbara; Daniel S Berman; Ron Blankstein; Jonathon Leipsic; David Newby; Edward D Nicol; Koen Nieman; Leslee Shaw; Todd C Villines; Michelle Williams; Harvey S Hecht Journal: J Cardiovasc Comput Tomogr Date: 2020-11-20