CBS - The 17th Left Main & Coronary Bifurcation Summit, CBS2025

HOME | SCIENTIFIC LIBRARY | Cardio-Oncology

Cardio-Oncology

Abnormalities in 3-Dimensional Left Ventricular Mechanics With Anthracycline Chemotherapy Are Associated With Systolic and Diastolic Dysfunction Cardiac Metastases in Patients with Neuroendocrine Tumours: Clinical Features, Therapy Outcomes, and Prognostic Implications Cardiotoxicity and Cardiac Monitoring Among Chemotherapy-Treated Breast Cancer Patients Longitudinal Assessment of Vascular Function With Sunitinib in Patients With Metastatic Renal Cell Carcinoma Apixaban to Prevent Venous Thromboembolism in Patients with Cancer Cardiovascular effects of radiation therapy Cardio-Oncology Services: rationale, organization, and implementation: A report from the ESC Cardio-Oncology council ACS Management in a Patient in Need of Noncardiac Surgery

Sort by hits |<< 1 2 3 4 5 >>|

Original Research

JOURNAL：CBSMD Article Link

人工智能助力介入心脏病学临床研究

CBSMD

Pre-reading

人工智能(Artificial Intelligence, AI)是用来表示运用数学算法赋予机器以推理能力，履行解决问题、识别事物/文字和辅助决策等认知功能的统称。AI涉及机器学习(machine learning，ML)、深度学习(deep learning， DL)、自然语言处理(natural language processing，NLP)、认知运算（cognitive computing）、计算机视觉(computer vision)和机器人（robotics)等各个分支领域。发表于2019年7月JACC Cardiovascular Interventions7月刊的“Impact of Artificial Intelligence on Interventional Cardiology: From Decision-Making Aid to Advanced Interventional Procedure Assistance”解读了从AI在介入心脏病学临床操作中从辅助决策到辅助介入术的萌芽及未来趋势。本文就其中涵盖的目前已开展的AI在介入心脏病学临床研究中的应用展开介绍。

AI为临床研究提供新的统计学思路及可能

“Incidence of contrast-induced acute kidney injury in a large cohort of all-comers undergoing percutaneous coronary intervention: Comparison of five contrast media”既是利用ML方法广义增强回归(generalized boosted regression）在大型复杂、异质性数据中验证对比剂的种类与PCI后对比剂相关急性肾损伤的相关性。而传统的统计学方法经常受数据本身的复杂性和异质性掣肘。

“Phenomapping for Novel Classification of Heart Failure With Preserved Ejection Fraction”通过深度学习聚合式阶层分群法（agglomerative hierarchical clustering）进一步指导HFpEF的表型分类。

AI辅助导管室日常检测

TCT 2018大会上的多项研究着眼于AI在心血管影像学中的应用，介绍了使用IVUS影像数据实现管腔面积、斑块负荷的自动运算。“Machine Learning Approaches in Cardiovascular Imaging”总结了AI在导管室用以识别包括狭窄直径、钙化、血栓及夹层等病变性质的应用。

“Real-world clinical utility and impact on clinical decision-making of coronary computed tomography angiography-derived fractional flow reserve: lessons from the ADVANCE Registry”说明了采用DL非介入影像学实现评估冠脉狭窄解剖学和功能学的可能性。

FAST-FFR “Accuracy of Fractional Flow Reserve Derived From Coronary Angiography”研究是一项有关冠脉造影模拟功能学与传统FFR测量准确性对比的前瞻性研究，FFRangio是直接将冠脉造影结果通过3D重建，并结合软件分析计算出FFR值的方法。研究结果显示与FFR相比，FFRangio的特异性和敏感性分别高达91%（95%CI，86-95％）和94%（95%CI，88-97％），诊断准确度高达92%，在FFR值介于0.75-0.85之间的病变，FFRangio的诊断准确度依然较高（为87％）。FFRangio和FFR的线性相关性表明了两者的一致性（r=0.80，P<0.001）

除此之外，ACC2018大会上也有研究指出AI或可在急诊为急性心肌梗死提供筛查提供准确度高的决策依据。

Abstract

Recommended Article

人工智能助力介入心脏病学临床研究

Pre-reading