Artificial Intelligence in Vascular Surgery: A Literature Review Focusing on Current Applications, Imaging Advances and Future Prospects.
Authors
Affiliations (8)
Affiliations (8)
- Faculty of Medicine, Medical University of Plovdiv, 4002 Plovdiv, Bulgaria.
- Department of Anatomy, Histology and Cytology, Faculty of Medicine, Medical University of Plovdiv, 4002 Plovdiv, Bulgaria.
- Neurological Surgery, Pulmed University Hospital, 4000 Plovdiv, Bulgaria.
- Department of General and Clinical Pathology, Faculty of Medicine, Medical University of Plovdiv, 4002 Plovdiv, Bulgaria.
- Department of Clinical Pathology, University Hospital "Pulmed", 4002 Plovdiv, Bulgaria.
- Department of Thoracic Surgery, University Hospital "Kaspela", 4002 Plovdiv, Bulgaria.
- Department of Cardiovascular Surgery, Medical University of Plovdiv, 4002 Plovdiv, Bulgaria.
- Clinical and Experimental Morphology Division, Research Institute at Medical University of Plovdiv, Medical University of Plovdiv, 4002 Plovdiv, Bulgaria.
Abstract
<b>Background/Objectives:</b> Artificial intelligence (AI) is increasingly being integrated into vascular surgery, particularly in diagnostic imaging, perioperative planning, intraoperative guidance, and postoperative surveillance. This literature review evaluates the current applications of artificial intelligence in vascular surgery and endovascular practice, with a particular focus on imaging technologies and their role in improving diagnostic precision, workflow efficiency, and patient outcomes. In addition, the review examines emerging AI applications in operative workflow optimization, endovascular navigation, postoperative surveillance, training platforms, and AI-assisted clinical decision support. <b>Methods:</b> A literature review was conducted using PubMed and Scopus with the search terms: (artificial intelligence OR AI OR neural network) AND (vascular surgery) AND (diagnosis OR treatment). Reference lists of included studies were manually screened, and additional recent studies were identified from relevant journals. Articles published in English up to April 2026 were included. Studies were assessed for their applications in vascular diagnostics, plaque characterization, endovascular workflow optimization, and postoperative surveillance. <b>Results:</b> AI demonstrated strong diagnostic performance across multiple imaging modalities. Deep learning systems achieved a sensitivity of 91.3% and specificity of 95.2% in peripheral arterial stenosis classification, while plaque characterization models showed accuracies up to 96% and substantial agreement with expert imaging interpretation. AI-assisted operative systems improved procedural efficiency through reductions in operative duration, radiation exposure, and contrast utilization. However, many studies were retrospective, single-center, and based on relatively small cohorts with heterogeneous endpoints. <b>Conclusions:</b> AI has significant potential to improve vascular surgical practice through enhanced image interpretation, procedural guidance, and individualized treatment planning. Despite promising outcomes, current evidence remains limited by methodological heterogeneity and insufficient external validation. Prospective multicenter studies and standardized evaluation frameworks are required before widespread clinical implementation can be achieved.