Abu Suleiman Amro, Russo Federico, Della Valle Luigi, Ausiello Davide, Bukowska-Olech Ewelina, Iannibelli Vincenzo, Al Droubi M Omar, Sannino Gabriella, Bernardi Marco, Spadafora Luigi
Hull University Teaching Hospitals, Hull HU3 2JZ, UK.
Department of Clinical and Molecular Medicine, Sapienza University of Rome, 00185 Rome, Italy.
J Cardiovasc Dev Dis. 2025 Sep 12;12(9):350. doi: 10.3390/jcdd12090350.
(1) Background: Myocardial bridging (MB) is a congenital coronary anomaly with potential clinical significance. Artificial intelligence (AI) applied to cardiac computed tomography angiography (CCTA), particularly through CT-derived fractional flow reserve (CT-FFR), offers a novel, non-invasive approach for assessing MB. (2) Methods: We conducted a systematic review of the literature focusing on studies investigating AI-enhanced CCTA in the evaluation of MB. (3) Results: Ten studies were included. AI-based models, including radiomics, demonstrated moderate to high accuracy in predicting proximal plaque formation, and motion correction algorithms improved image quality and diagnostic confidence. Other findings were limited by the types of studies included and conflicting findings across studies. (4) Conclusions: AI-enhanced CCTA shows promise for the non-invasive functional assessment of MB and its risk stratification. Further prospective studies and validation are required to establish standardized protocols and confirm clinical utility.
（1）背景：心肌桥（MB）是一种具有潜在临床意义的先天性冠状动脉异常。应用于心脏计算机断层扫描血管造影（CCTA）的人工智能（AI），特别是通过CT衍生的血流储备分数（CT-FFR），为评估心肌桥提供了一种新颖的非侵入性方法。（2）方法：我们对文献进行了系统回顾，重点关注研究人工智能增强型CCTA在心肌桥评估中的应用。（3）结果：纳入了10项研究。基于人工智能的模型，包括放射组学，在预测近端斑块形成方面显示出中等至高的准确性，运动校正算法提高了图像质量和诊断信心。其他研究结果受到所纳入研究类型的限制，且各研究结果相互矛盾。（4）结论：人工智能增强型CCTA在心肌桥的非侵入性功能评估及其风险分层方面显示出前景。需要进一步的前瞻性研究和验证来建立标准化方案并确认临床实用性。
