Department of Radiology, Beijing Friendship Hospital, Capital Medical University, No. 95 YongAn Road, Beijing, 100050, People's Republic of China.
Shukun (Beijing) Technology Co., Ltd., Jinhui Bd, Qiyang Rd, Beijing, 100102, People's Republic of China.
BMC Med Imaging. 2022 Feb 17;22(1):28. doi: 10.1186/s12880-022-00756-y.
BACKGROUND: To investigate the influence of artificial intelligence (AI) based on deep learning on the diagnostic performance and consistency of inexperienced cardiovascular radiologists. METHODS: We enrolled 196 patents who had undergone both coronary computed tomography angiography (CCTA) and invasive coronary angiography (ICA) within 6 months. Four readers with less cardiovascular experience (Reader 1-Reader 4) and two cardiovascular radiologists (level II, Reader 5 and Reader 6) evaluated all images for ≥ 50% coronary artery stenosis, with ICA as the gold standard. Reader 3 and Reader 4 interpreted with AI system assistance, and the other four readers interpreted without the AI system. The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and accuracy (area under the receiver operating characteristic curve (AUC)) of the six readers were calculated at the patient and vessel levels. Additionally, we evaluated the interobserver consistency between Reader 1 and Reader 2, Reader 3 and Reader 4, and Reader 5 and Reader 6. RESULTS: The AI system had 94% and 78% sensitivity at the patient and vessel levels, respectively, which were higher than that of Reader 5 and Reader 6. AI-assisted Reader 3 and Reader 4 had higher sensitivity (range + 7.2-+ 16.6% and + 5.9-+ 16.1%, respectively) and NPVs (range + 3.7-+ 13.4% and + 2.7-+ 4.2%, respectively) than Reader 1 and Reader 2 without AI. Good interobserver consistency was found between Reader 3 and Reader 4 in interpreting ≥ 50% stenosis (Kappa value = 0.75 and 0.80 at the patient and vessel levels, respectively). Only Reader 1 and Reader 2 showed poor interobserver consistency (Kappa value = 0.25 and 0.37). Reader 5 and Reader 6 showed moderate agreement (Kappa value = 0.55 and 0.61). CONCLUSIONS: Our study showed that using AI could effectively increase the sensitivity of inexperienced readers and significantly improve the consistency of coronary stenosis diagnosis via CCTA. Trial registration Clinical trial registration number: ChiCTR1900021867. Name of registry: Diagnostic performance of artificial intelligence-assisted coronary computed tomography angiography for the assessment of coronary atherosclerotic stenosis.
背景：本研究旨在探讨基于深度学习的人工智能（AI）对经验不足的心血管放射科医师诊断性能和一致性的影响。
方法：我们纳入了 196 例在 6 个月内行冠状动脉计算机断层扫描血管造影（CCTA）和有创冠状动脉造影（ICA）的患者。4 位心血管经验较少的阅片者（Reader 1-Reader 4）和 2 位心血管放射科医师（二级阅片者，Reader 5 和 Reader 6）对所有图像进行了评估，以≥50%的冠状动脉狭窄为阳性标准，以 ICA 作为金标准。Reader 3 和 Reader 4 在 AI 系统辅助下进行解读，其他 4 位阅片者在无 AI 系统辅助下进行解读。以患者和血管为水平，计算 6 位阅片者的敏感性、特异性、阳性预测值（PPV）、阴性预测值（NPV）和准确性（受试者工作特征曲线下面积（AUC））。此外，我们还评估了 Reader 1 和 Reader 2、Reader 3 和 Reader 4、Reader 5 和 Reader 6 之间的观察者间一致性。
结果：AI 系统在患者和血管水平的敏感性分别为 94%和 78%，均高于 Reader 5 和 Reader 6。在有 AI 系统辅助的情况下，Reader 3 和 Reader 4 的敏感性（分别为+7.2%至+16.6%和+5.9%至+16.1%）和 NPV（分别为+3.7%至+13.4%和+2.7%至+4.2%）均高于无 AI 系统辅助的 Reader 1 和 Reader 2。Reader 3 和 Reader 4 在解读≥50%狭窄方面具有良好的观察者间一致性（患者和血管水平的 Kappa 值分别为 0.75 和 0.80）。仅 Reader 1 和 Reader 2 显示出较差的观察者间一致性（Kappa 值分别为 0.25 和 0.37）。Reader 5 和 Reader 6 显示出中度一致性（Kappa 值分别为 0.55 和 0.61）。
结论：本研究表明，使用 AI 可以有效提高经验不足的阅片者的敏感性，并显著提高 CCTA 评估冠状动脉粥样硬化狭窄的诊断一致性。
临床试验注册：ChiCTR1900021867，名称：人工智能辅助冠状动脉计算机断层扫描血管造影对冠状动脉粥样硬化狭窄评估的诊断性能。
Int J Cardiovasc Imaging. 2024-5
Comput Methods Programs Biomed. 2020-11
Int J Cardiovasc Imaging. 2025-5
Diagnostics (Basel). 2023-11-30
Front Cardiovasc Med. 2023-2-16
Comput Methods Programs Biomed. 2020-11
J Geriatr Cardiol. 2020-1
Zotero 插件
