贝克曼库尔特转发了
This research shows that when designed properly, AI can mitigate human bias rather than perpetuate it!
Excited to share our latest research, "Enhancing Emergency Department Triage Equity With Artificial Intelligence: Outcomes From a Multisite Implementation," now published in Annals of Emergency Medicine! In this study, our team examined the implementation of AI-informed clinical decision support (TriageGO) across three emergency departments, analyzing its impact on the accuracy, performance, and equity of triage decision-making. This study is one of the first to demonstrate the positive impact of AI on equity in healthcare. Here are the major takeaways: Improved Sensitivity and Specificity - AI-CDS enhanced the identification of patients with critical illness (ICU admission or in-hospital death), increasing sensitivity from 78.9% to 83.1% and specificity from 74.3% to 76.8%. Significant Reduction in Inequities - Historically marginalized groups—Black, Hispanic, non-English-speaking, and female patients—saw the greatest improvements in identifying critical illness: Sensitivity for Black patients increased by 7.7%, Sensitivity for non-English speakers improved by 10.7%, Hispanic patients experienced an 8.3% increase, Women saw sensitivity rise by 4.7%, All of these performance improvements significantly improved equity gaps. A huge thank you to my co-authors, including Jeremiah Hinson, MD PhD, Rohit Sangal, Arjun Venkatesh, Benjamin Steinhart, and Chris Chmura, as well as the incredible clinical teams at Yale New Haven Health Check out the full article here: https://lnkd.in/eVj5Vk_R #EmergencyMedicine #ArtificialIntelligence #HealthcareEquity #TriageInnovation #AIInHealthcare Yale Biomedical Informatics & Data Science Yale School of Medicine Yale Emergency Medicine