Investigating the Accuracy and Efficiency of AI-Based EKG Report Summary Writing in Cardiovascular Care in Thailand: A Proof of Concept Version

Abstract

Integrating artificial intelligence (AI) into healthcare has the potential to revolutionize diagnostic processes by enhancing accuracy and efficiency. In cardiovascular care, AI systems like WellBeat can automate tasks such as EKG summary writing, reducing physician workload while maintaining high standards of care. This study evaluates WellBeat’s accuracy and time-saving potential in a clinical setting in Thailand, involving three doctors who assessed its performance. Accuracy was tested on both trained and previously unseen (unknown) data, with results showing 81% accuracy on trained data and 98% on unseen data, demonstrating strong generalization. Time efficiency analysis revealed an 85% reduction in time spent on EKG summary writing, translating to approximately 9.5 hours saved per clinic weekly. The findings highlight WellBeat’s ability to streamline clinical workflows and improve patient care. Despite its strong performance, variability in doctor acceptance for complex cases emphasizes the necessity of human oversight. With further refinement, WellBeat shows promise for large-scale adoption in high-volume healthcare settings.

Ähnliche Arbeiten