Scholarlink

The application of artificial intelligence (AI) in cardiac computed tomography (CT) imaging represents a significant advancement in the management of coronary artery disease (CAD). AI enhances image quality, accelerates image analysis, and assists in identifying myocardial ischemia and hemodynamically significant stenosis. Convolutional neural networks (CNNs) have been employed to mitigate image noise in low-dose non-contrast cardiac CT scans, facilitating accurate coronary artery calcium (CAC) scoring. Additionally, CNN-based denoising algorithms applied to coronary computed tomography angiography (CCTA) improve CAD detection at reduced radiation doses. Furthermore, AI algorithms are applied for automatic segmentation of cardiac structures, essential for precise stenosis evaluation and coronary plaque characterization. Advanced AI methodologies, including deep learning and radiomics, enhance the assessment of plaque morphology and myocardial ischemia, with emerging applications in CT-derived fractional flow reserve (CTFFR), showing increased diagnostic accuracy. AI’s role in detecting pulmonary embolism (PE) further exemplifies its diagnostic potential, with AI algorithms demonstrating high sensitivity and specificity. The integration of AI in cardiac CT imaging is poised to optimize diagnostic workflows, enhance clinical decision-making, and improve patient outcomes.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Improving Cardiovascular Diagnosis in Computed Tomography Imaging with the Use of Artificial Intelligence

  • 章节

  • 发表日期：

    2025年

查看全文
  • Costanza Lisi,
  • Federica Catapano,
  • Marco Francone

摘要

The application of artificial intelligence (AI) in cardiac computed tomography (CT) imaging represents a significant advancement in the management of coronary artery disease (CAD). AI enhances image quality, accelerates image analysis, and assists in identifying myocardial ischemia and hemodynamically significant stenosis. Convolutional neural networks (CNNs) have been employed to mitigate image noise in low-dose non-contrast cardiac CT scans, facilitating accurate coronary artery calcium (CAC) scoring. Additionally, CNN-based denoising algorithms applied to coronary computed tomography angiography (CCTA) improve CAD detection at reduced radiation doses. Furthermore, AI algorithms are applied for automatic segmentation of cardiac structures, essential for precise stenosis evaluation and coronary plaque characterization. Advanced AI methodologies, including deep learning and radiomics, enhance the assessment of plaque morphology and myocardial ischemia, with emerging applications in CT-derived fractional flow reserve (CTFFR), showing increased diagnostic accuracy. AI’s role in detecting pulmonary embolism (PE) further exemplifies its diagnostic potential, with AI algorithms demonstrating high sensitivity and specificity. The integration of AI in cardiac CT imaging is poised to optimize diagnostic workflows, enhance clinical decision-making, and improve patient outcomes.