Harnessing Artificial Intelligence in Interventional Cardiology: A Systematic Review of Current Applications

Abstract

Interventional cardiology has recently advanced with innovations such as percutaneous transluminal coronary angioplasty (PTCA), transcatheter aortic valve replacement (TAVR), and the emergence of artificial intelligence (AI) as a transformative tool. This systematic review explored the current landscape, methodologies, and applications of AI in interventional cardiology. A comprehensive literature search was conducted following preferred reporting guidelines, identifying 20 studies after data extraction and quality assessment. AI-particularly machine learning (ML) and deep learning (DL)-enhances diagnostic accuracy and procedural efficiency. ML aids in arrhythmia detection and coronary plaque characterization, while DL supports imaging interpretation, robotic navigation, and catheter tracking. Clinical applications show AI's potential in predicting myocardial infarction, guiding personalized treatment, and improving resource management. Despite these benefits, challenges such as data privacy, algorithm transparency, and generalizability remain. Addressing these requires collaborative efforts and robust data sharing. Future priorities include integrating AI into routine clinical workflows, resolving regulatory barriers, and ensuring interpretability. Multidisciplinary collaboration is essential to address ethical considerations and uphold patient safety. The integration of AI in interventional cardiology offers significant potential to enhance patient care, procedural precision, and resource utilization. However, its adoption must be guided by careful attention to ethical, technical, and regulatory constraints. Overcoming these barriers through coordinated efforts may allow AI to redefine standards in cardiovascular care and enable a more precise, efficient, and patient-centered approach to interventional cardiology.

Keywords: ai & robotics healthcare; ai and machine learning; artificial intelligence (ai); cognitive computing; convolutional neural networks (cnn); deep learning (dl); interventional cardiology; natural language processing (nlp).

Figure 1. PRISMA flow diagram

PRISMA: Preferred reporting items for systematic reviews and meta-analyses