AI in Radiology and Interventions: Workflow Automation, Accuracy, and Efficiency Gains Today-and What's Coming

This review analyzes the current landscape and future outlook of artificial intelligence (AI) and automation in selected diagnostic imaging and image-guided therapy workflows. Evaluating four procedures-MRI cancer screening, CT lung screening, coronary stenting, and ultrasound-guided liver cryoablation-it assesses AI impact on workflow optimization, accuracy, procedure times, and new clinical insights. As of 2024, 903 AI-enabled medical devices have received FDA authorization (76.6% in radiology). The integration of deep learning, generative AI, and automation technologies is transforming diagnostic accuracy, reporting efficiency, and interventional guidance. By 2030, near-universal adoption across both diagnostic and interventional workflows is projected, with AI increasingly serving as a collaborative tool for clinicians. Key implementation challenges include data quality, transparency, workforce adaptation, and regulatory barriers. Overall, AI augments, rather than replaces, human expertise, driving substantial improvements in healthcare delivery.