A review of mature machine learning- and artificial intelligence-enabled applications in aortic surgery

Abstract

IntroductionAfter years of both increasing enthusiasm and skepticism by surgeons and patients, the first artificial intelligence (AI) and machine learning (ML) algorithms for treating patients with aortic disease are now maturing. Multiple AI- and ML-enabled tools in this space are now U.S. Food and Drug Administration approved, being commercialized and, in turn, having an increasing impact across the care continuum. The goal of this article is to review the major themes in mature AI and ML algorithms in aortic surgery with a focus on image analysis, diagnosis, and open and endovascular surgical planning and intraoperative guidance. MethodsWe used a natural language AI-enabled tool called ChatGPT 4.0 to perform this narrative review of the most clinically mature AI- and ML-enabled tools in aortic surgery. We describe the major themes of the first devices reaching clinical care and the risk and barriers to further adoption. ResultsAI- and ML-enabled tools are being used to evaluate radiologic imaging to increase the speed and accuracy of diagnoses, including aortic dissection and aneurysm. Other tools are assisting with surgical planning, intraoperative guidance, decreasing radiation exposure, and facilitating or performing postoperative surveillance. These technologies are continuing to mature, and their integration with the health care system is expected to accelerate in the coming years as algorithm performance improves, regulatory framework solidifies, and hospital systems, payors, and physicians recognize their utility. ConclusionsAI- and ML-enabled tools are clinically maturing and impacting patient care, including those with aortic disease. The first of these tools are primarily being used for radiologic image evolution, diagnosis, and risk stratification. Although barriers still exist, as these algorithms and their enabled tools become integrated with other emerging technologies, the rapidity at which new products mature will increase. We authored and revised this article about AI and ML in the aortic surgery space with the help of a natural language AI-enabled tool. This highlights just one example of how technologies from completely unrelated spaces may accelerate progress in vascular surgery in unpredictable ways.