基于人工智能的冠状动脉CT血管成像斑块量化:当前认识与未来方向
本文综述了人工智能定量斑块(AI-QP)分析工具的临床意义、各厂商验证现状及对患者护理与结局的影响。斑块负荷与进展是冠心病预后指标,AI-QP可能更准确识别不良事件风险者并影响管理,但其增量预后价值、跨扫描可重复性及斑块进展的临床有意义阈值尚待确立。
要点
- 斑块负荷与进展是冠心病患者的关键预后指标,AI-QP可能更准确识别不良事件风险个体。
- AI-QP相对于冠状动脉钙化积分等既定斑块负荷指标的增量预后价值尚待进一步确立。
- 用于监测冠心病进展时,AI-QP跨CCTA扫描的可重复性需验证,斑块进展的临床有意义阈值需定义。
核心结论
- AI-QP有潜力更准确识别不良事件风险个体并影响患者管理,但需进一步证据确认其临床效用并克服广泛临床实施的障碍。
研究设计与样本
- 综述,文章回顾了斑块量化的临床意义、各AI-QP厂商的验证现状及其对患者护理与结局的意义。
临床应用与阅片要点
- AI-QP可能改善风险分层并指导治疗,但广泛临床实施前需进一步证据确认其临床效用。
局限性
- AI-QP相对于冠状动脉钙化积分等既定斑块负荷指标的增量预后价值必须进一步确立。
- 用于监测冠心病进展时,AI-QP跨CCTA扫描的可重复性必须验证。
- 斑块进展的临床有意义阈值需要定义。
- 需进一步证据确认AI-QP的临床效用并克服广泛临床实施的障碍。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
英文原文(摘要)· 点击展开
Over the past decade, coronary CT angiography (CCTA) has seen major advancements in spatial and temporal resolution, as well as postprocessing software that now integrates artificial intelligence-based quantitative plaque (AI-QP) analysis tools. This review highlights the clinical significance of plaque quantification, examines the validation status of various AI-QP vendors, and discusses implications for patient care and outcomes. As plaque burden and progression are increasingly recognized as key prognostic indicators in patients with coronary artery disease (CAD), AI-QP has the potential to more accurately identify individuals at risk for adverse events and thereby influence patient management. However, several challenges must be addressed. The incremental prognostic value of AI-QP over established measures of plaque burden, such as the coronary artery calcium score, must be further established. For monitoring CAD progression, the reproducibility of AI-QP across CCTA scans must also be validated, and clinically meaningful thresholds for plaque progression need to be defined. Although AI-QP shows promise for improving risk stratification and guiding treatment, further evidence is needed to confirm its clinical utility and overcome barriers to widespread implementation in clinical practice. Keywords: CT-Coronary Angiography, Applications - CT, Artificial Intelligence © RSNA, 2025.
引用
- Jukema RA, Blanke P, Khoo JK, et al. Artificial Intelligence-based Coronary Plaque Quantification Using Coronary CT Angiography: Current Insights and Future Directions. Radiol Cardiothorac Imaging. 2025. PMID: 41165444. doi:10.1148/ryct.240568
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。