Artificial intelligence and innovation of clinical care: the need for evidence in the real world.

The article discusses the potential of artificial intelligence (AI) in clinical care, specifically in the field of cardiovascular imaging. The authors highlight the challenges of identifying AI applications with real-world impact and emphasize the importance of evidence-based research for clinical implementation. They present a study by Pieszko et al. that developed an AI model to predict the likelihood of left atrial appendage thrombus in patients undergoing cardioversion. The model showed promising results, outperforming traditional diagnostic methods. The authors suggest that further research, including randomized controlled trials, is needed to assess the true value and impact of AI models in clinical practice. [Extracted from the article]