Centro Cardiologico Monzino, IRCCS, Milan, Italy.
Department of Electronics, Information and Biomedical Engineering, Politecnico di Milano, 20133, Milan, Italy.
Eur J Nucl Med Mol Imaging. 2022 Jul;49(9):3119-3128. doi: 10.1007/s00259-022-05732-w. Epub 2022 Feb 23.
To evaluate the diagnostic accuracy of a deep learning (DL) algorithm predicting hemodynamically significant coronary artery disease (CAD) by using a rest dataset of myocardial computed tomography perfusion (CTP) as compared to invasive evaluation.
One hundred and twelve consecutive symptomatic patients scheduled for clinically indicated invasive coronary angiography (ICA) underwent CCTA plus static stress CTP and ICA with invasive fractional flow reserve (FFR) for stenoses ranging between 30 and 80%. Subsequently, a DL algorithm for the prediction of significant CAD by using the rest dataset (CTP-DL) and stress dataset (CTP-DL) was developed. The diagnostic accuracy for identification of significant CAD using CCTA, CCTA + CTP stress, CCTA + CTP-DL, and CCTA + CTP-DL was measured and compared. The time of analysis for CTP stress, CTP-DL, and CTP-DL was recorded.
Patient-specific sensitivity, specificity, NPV, PPV, accuracy, and area under the curve (AUC) of CCTA alone and CCTA + CTP were 100%, 33%, 100%, 54%, 63%, 67% and 86%, 89%, 89%, 86%, 88%, 87%, respectively. Patient-specific sensitivity, specificity, NPV, PPV, accuracy, and AUC of CCTA + DL and CCTA + DL were 100%, 72%, 100%, 74%, 84%, 96% and 93%, 83%, 94%, 81%, 88%, 98%, respectively. All CCTA + CTP stress, CCTA + CTP-DL, and CCTA + CTP-DL significantly improved detection of hemodynamically significant CAD compared to CCTA alone (p < 0.01). Time of CTP-DL was significantly lower as compared to human analysis (39.2 ± 3.2 vs. 379.6 ± 68.0 s, p < 0.001).
Evaluation of myocardial ischemia using a DL approach on rest CTP datasets is feasible and accurate. This approach may be a useful gatekeeper prior to CTP stress.
通过使用静息心肌 CT 灌注（CTP）数据集评估深度学习（DL）算法预测血流动力学意义上的冠状动脉疾病（CAD）的诊断准确性，并与有创评估进行比较。
112 例连续的有症状患者，计划行有创性临床指征的冠状动脉造影（ICA），均行 CCTA 加静息负荷 CTP 和 ICA 以评估狭窄程度 30%至 80%的狭窄。随后，开发了一种基于静息数据集（CTP-DL）和负荷数据集（CTP-DL）预测有意义 CAD 的 DL 算法。测量并比较了 CCTA、CCTA+CTP 负荷、CCTA+CTP-DL 和 CCTA+CTP-DL 用于识别有意义 CAD 的诊断准确性。记录 CTP 负荷、CTP-DL 和 CTP-DL 的分析时间。
单纯 CCTA 的患者特异性敏感性、特异性、NPV、PPV、准确性和曲线下面积（AUC）分别为 100%、33%、100%、54%、63%和 67%，CCTA+CTP 分别为 86%、89%、89%、86%、88%和 87%。CCTA+DL 和 CCTA+DL 的患者特异性敏感性、特异性、NPV、PPV、准确性和 AUC 分别为 100%、72%、100%、74%、84%和 96%，CCTA+CTP-DL 分别为 93%、83%、94%、81%、88%和 98%。与单纯 CCTA 相比，CCTA+CTP 负荷、CCTA+CTP-DL 和 CCTA+CTP-DL 均显著提高了对血流动力学意义上的 CAD 的检测（p<0.01）。与人工分析相比，CTP-DL 的时间明显缩短（39.2±3.2 与 379.6±68.0 s，p<0.001）。
使用静息 CTP 数据集的 DL 方法评估心肌缺血是可行和准确的。该方法可能是 CTP 负荷之前的有用筛选方法。
