Department of Cardiovascular Surgery, Ruijin Hospital, Shanghai Jiaotong University School of Medicine, Shanghai 200025, China.
Contrast Media Mol Imaging. 2022 Mar 23;2022:8628668. doi: 10.1155/2022/8628668. eCollection 2022.
This research aimed at discussing the diagnosis effect of coronary artery magnetic resonance angiography (MRA) combined with computed tomography (CT) angiography (CTA) based on the back-projection filter reconstruction (BPFR) algorithm in coronary heart disease (CHD), and its role in the diagnosis of coronary artery disease (CAD). Sixty patients with CHD were selected and randomly rolled into group A (undergone MRA examination), group B (undergone CTA examination), and group C (undergone MRA + CTA), with 20 cases in each group. Taking the diagnostic results of coronary angiography as the gold standard, the MRA and CTA images were reconstructed using a BPFR algorithm, and a filter function was added to solve the problem of image sharpness. In addition, the iterative reconstruction algorithm and the Fourier transform analysis method were introduced. As a result, the image clarity and resolution obtained by the BPFR algorithm were better than those obtained by the Fourier transform analytical method and the iterative reconstruction algorithm. The accuracy of group C for the diagnosis of mild coronary stenosis, moderate stenosis, and severe stenosis was 94.02%, 96.13%, and 98.01%, respectively, which was significantly higher than that of group B (87.5%, 90.2%, and 88.4%) and group C (83.4%, 89.1%, and 91.5%) ( < 0.05). The sensitivity and specificity for the diagnosis of noncalcified plaque in group C were 87.9% and 89.2%, respectively, and the sensitivity and specificity for the diagnosis of calcified plaque were 84.5% and 78.4%, respectively, which were significantly higher than those in groups B and C ( < 0.05). In summary, the BPFR algorithm had good denoising and artifact removal effects on coronary MRA and CTA images. The combined detection of reconstructed MRA and CTA images had a high diagnostic value for CHD.
本研究旨在探讨基于反向投影滤波重建（BPFR）算法的冠状动脉磁共振血管造影（MRA）联合计算机断层扫描血管造影（CTA）对冠心病（CHD）的诊断效果及其在冠状动脉疾病（CAD）诊断中的作用。选取 60 例 CHD 患者，采用随机数字表法分为 A 组（行 MRA 检查）、B 组（行 CTA 检查）和 C 组（行 MRA+CTA 检查），每组 20 例。以冠状动脉造影诊断结果为金标准，采用 BPFR 算法对 MRA 和 CTA 图像进行重建，并添加滤波函数以解决图像锐利度问题。同时引入迭代重建算法和傅里叶变换分析方法。结果表明，BPFR 算法重建的图像清晰度和分辨率优于傅里叶变换分析方法和迭代重建算法。C 组对轻度、中度和重度冠状动脉狭窄的诊断准确率分别为 94.02%、96.13%和 98.01%，明显高于 B 组（87.5%、90.2%和 88.4%）和 C 组（83.4%、89.1%和 91.5%）（ < 0.05）。C 组对非钙化斑块的诊断敏感性和特异性分别为 87.9%和 89.2%，对钙化斑块的诊断敏感性和特异性分别为 84.5%和 78.4%，明显高于 B 组和 C 组（ < 0.05）。综上所述，BPFR 算法对冠状动脉 MRA 和 CTA 图像具有良好的降噪和去伪影效果。重建后的 MRA 和 CTA 图像联合检测对 CHD 具有较高的诊断价值。
