Automated quantification of coronary artery-related myocardial territory on cardiac CT

2017 Fiscal Year Final Research Report

• Project/Area Number: 15K09892
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Radiation science
• Research Institution: Ehime University
• Principal Investigator: Kurata Akira 愛媛大学, 医学部附属病院, 講師 (40423438)
• Co-Investigator(Kenkyū-buntansha): 上谷 晃由 愛媛大学, 医学部附属病院, 講師 (00423450) ; 城戸 輝仁 愛媛大学, 医学系研究科, 講師 (50403837) ; 望月 輝一 愛媛大学, 医学系研究科, 教授 (80145094)
• Project Period (FY): 2015-04-01 – 2018-03-31
• Keywords: 画像診断 / 冠動脈疾患 / computed tomography / 支配かん流域 / 定量評価

Outline of Final Research Achievements

We have developed the Voronoi algorithm-based myocardial segmentation on CT according to coronary artery stenosis reliably estimated SPECT based myocardial ischemic territories.
In this study, we assessed the relationship between the CTA-based myocardial area at risk (MAAR) according to the infarction-culprit arteries (IRA) and the infarct size quantified by the late gadolinium enhancement in cardiac MR (LGE-MRI) in patients with acute myocardial infarction (AMI). Significant correlation between the two assessments were observed in 15 patients with initial onset of AMI (r=0.625, p<0.05; Spearman test). In most cases, the LGE area were located in the core of the IRA-related CT-MAAR, and smaller than them. This study demonstrated that the Voronoi diagram based myocardial segmentation could objectively estimate the maximum myocardial area at risk would result in myocardial infarction, and might aid in the diagnosis and therapeutic decision-making in patients with coronary artery disease.

Free Research Field

放射線科