Journal of Research and Health

en Artificial Intelligence in Coronary Angiography: A Systematic Review of Automated Detection, Segmentation, 3D Reconstruction, and Risk Prediction ● Artificial Intelligence مقاله مروری Review Article <div> <div id="ftn1">Background: This review analyzes AI's transformative role across the full spectrum of coronary imaging (CCTA<a href="#_ftn1" name="_ftnref1" title="">[1]</a> and XCA<a href="#_ftn2" name="_ftnref2" title="">[2]</a>). AI is addressing critical limitations by enhancing image quality, enabling personalized interpretation, and improving workflow efficiency. Methods: In this study, we conducted a comprehensive search of the scientific literature for studies on the application of AI in CCTA or XCA to date. Searches were performed in PubMed/MEDLINE, Embase, and IEEE Xplore. Two reviewers then independently screened the selected articles for data extraction, which a third reviewer validated. The findings were ultimately organized into five topic areas: the role of AI in coronary CTA, coronary artery segmentation, automated coronary artery detection and classification, 3D image reconstruction of coronary arteries, and risk prediction assessment. Results: From the 98 records identified, 62 studies met the inclusion criteria. In CCTA, AI applications extend from anatomical assessment to automated plaque quantification, functional assessment via FFR<a href="#_ftn3" name="_ftnref3" title="">[3]</a>-CT, and the extraction of novel biomarkers. For XCA, AI facilitates automated stenosis quantification, 3D reconstruction, and real-time guidance. Key innovations include advanced architectures like DR-LCT-U-Net, benchmark datasets such as CADICA, and specialized systems like CAG-VLM for diagnostic support. Ultimately, AI models significantly outperform conventional methods in risk stratification. Conclusion: While AI offers significant potential to bridge CCTA and XCA, its clinical implementation faces hurdles in interpretability, integration, and validation. Realizing personalized coronary artery disease management will require prospective trials and explainable AI frameworks. <div>  <hr align="left" size="1" width="33%" > <div id="ftn1"><a href="#_ftnref1" name="_ftn1" title="">[1]</a> Coronary Computed Tomography Angiography</div> <div id="ftn2"><a href="#_ftnref2" name="_ftn2" title="">[2]</a> X-ray Coronary Angiography</div> <div id="ftn3"><a href="#_ftnref3" name="_ftn3" title="">[3]</a> Fractional Flow Reserve</div> </div> </div> </div> Coronary Angiography, Artificial Intelligence, Automated Detection, Segmentation, 3D Reconstruction, Risk Prediction 7 7 http://jrh.gmu.ac.ir/browse.php?a_code=A-10-2914-1&slc_lang=en&sid=1 Rezvan Monjezi rezvan.monjezi@neyshabur.ac.ir 1 No Department of Intelligent & Cognitive Technologies, Faculty of Novel Interdisciplinary Technologies, University of Neyshabur, Iran. Mahdieh Ghasemi ghasemi.biomed@gmail.com 2 Yes Department of Intelligent & Cognitive Technologies, Faculty of Novel Interdisciplinary Technologies, University of Neyshabur, Iran. Alireza Rowhanimanesh rowhanimanesh@neyshabur.ac.ir 3 No Department of Intelligent & Cognitive Technologies, Faculty of Novel Interdisciplinary Technologies, University of Neyshabur, Iran. Seyedeh Samaneh Tabaee Tabaees@nums.ac.ir 4 No Noncommunicable Disease Research Center, Neyshabur University of Medical Sciences, Iran.