Ryoichi Shimamoto1, Jun-Ichi Suzuki2, Tadashi Yamazaki1, Taeko Tsuji1, Yuki Ohmoto1, Toshihiro Morita1, Hiroshi Yamashita1, Junko Honye1, Ryozo Nagai1, Masaaki Akahane3, Kuni Ohtomo3. 1. Department of Cardiovascular Medicine, Faculty of Medicine and Graduate School of Medicine, University of Tokyo, Hongo 7-3-1, Bunkyo-ku, Tokyo 1138655, Japan. 2. Department of Cardiovascular Medicine, Faculty of Medicine and Graduate School of Medicine, University of Tokyo, Hongo 7-3-1, Bunkyo-ku, Tokyo 1138655, Japan. Electronic address: suzukij-2im@h.u-tokyo.ac.jp. 3. Department of Radiology, Faculty of Medicine and Graduate School of Medicine, University of Tokyo, Hongo 7-3-1, Bunkyo-ku, Tokyo 1138655, Japan.
Abstract
PURPOSE: Coronary artery vascular edge recognition on computed tomography (CT) angiograms is influenced by window parameters. A noninvasive method for vascular edge recognition independent of window setting with use of multi-detector row CT was contrived and its feasibility and accuracy were estimated by intravascular ultrasound (IVUS). METHODS: Multi-detector row CT was performed to obtain 29 CT spatial profile curves by setting a line cursor across short-axis coronary angiograms processed by multi-planar reconstruction. IVUS was also performed to determine the reference coronary diameter. IVUS diameter was fitted horizontally between two points on the upward and downward slopes of the profile curves and Hounsfield number was measured at the fitted level to test seven candidate indexes for definition of intravascular coronary diameter. The best index from the curves should show the best agreement with IVUS diameter. RESULTS: Of the seven candidates the agreement was the best (agreement: 16±11%) when the two ratios of Hounsfield number at the level of IVUS diameter over that at the peak on the profile curves were used with water and with fat as the background tissue. These edge definitions were achieved by cutting the horizontal distance by the curves at the level defined by the ratio of 0.41 for water background and 0.57 for fat background. CONCLUSIONS: Vascular edge recognition of the coronary artery with CT spatial profile curves was feasible and the contrived method could define the coronary diameter with reasonable agreement.
PURPOSE: Coronary artery vascular edge recognition on computed tomography (CT) angiograms is influenced by window parameters. A noninvasive method for vascular edge recognition independent of window setting with use of multi-detector row CT was contrived and its feasibility and accuracy were estimated by intravascular ultrasound (IVUS). METHODS: Multi-detector row CT was performed to obtain 29 CT spatial profile curves by setting a line cursor across short-axis coronary angiograms processed by multi-planar reconstruction. IVUS was also performed to determine the reference coronary diameter. IVUS diameter was fitted horizontally between two points on the upward and downward slopes of the profile curves and Hounsfield number was measured at the fitted level to test seven candidate indexes for definition of intravascular coronary diameter. The best index from the curves should show the best agreement with IVUS diameter. RESULTS: Of the seven candidates the agreement was the best (agreement: 16±11%) when the two ratios of Hounsfield number at the level of IVUS diameter over that at the peak on the profile curves were used with water and with fat as the background tissue. These edge definitions were achieved by cutting the horizontal distance by the curves at the level defined by the ratio of 0.41 for water background and 0.57 for fat background. CONCLUSIONS: Vascular edge recognition of the coronary artery with CT spatial profile curves was feasible and the contrived method could define the coronary diameter with reasonable agreement.