How integrating AI and clinical decision support systems can help in the ER

The deployment of artificial intelligence (AI) for point-of-care clinical decision support in emergency medicine is still in its early stages. Despite media attention and numerous AI studies, there is limited evidence of successful translation to clinical practice. Andrew Taylor, an associate professor of emergency medicine at Yale University School of Medicine, emphasizes the potential for revolutionizing care delivery by integrating AI and clinical decision support (CDS) in emergency medicine.

In an upcoming session at the HIMSS24 Global Conference & Exhibition, Taylor will discuss the complex environment of the emergency department (ED) and how AI-CDS tools can streamline processes, improve patient outcomes, and optimize resource utilization. The focus will be on various applications, including triage, patient disposition, diagnosis, and risk assessment. Taylor stresses the importance of developing AI systems that complement and support clinicians rather than replacing them, emphasizing the human-centric care at the core of medicine.

The discussion will delve into how AI-CDS facilitates rapid and precise triage, enhances risk assessment, and improves diagnostic accuracy in the high-stakes ED environment. Taylor emphasizes the critical role of stakeholders, including clinicians, healthcare staff, and patients, in the acceptance and integration of AI-CDS systems. The success of these tools depends not only on technological sophistication but also on their alignment with the core values of healthcare, such as compassion, privacy, and equity.

The session will also highlight the significance of establishing a robust infrastructure for AI-CDS deployment, focusing on user-friendly design, adaptability, and the implementation of machine learning operations (MLOps) for monitoring, maintenance, and continuous improvement. Taylor concludes that meticulous attention to operational infrastructure and cultivating a symbiotic relationship between AI-CDS tools and clinical workflows are essential for the success and sustainability of AI in emergency care settings. The session is scheduled for March 12 at HIMSS24 in Orlando.