Paper Type
Complete
Abstract
Venous thromboembolism (VTE) is a leading cause of preventable hospital death in the United States, significantly impacting patient outcomes and healthcare resources. This study explores the development of an AI-enabled VTE surveillance system designed to assist in clinical decision-making and resource management. Using machine learning models trained on electronic medical record (EMR) data from 57,490 inpatient encounters, we developed a predictive tool to help prioritize VTE potential while minimizing the need for complex laboratory and observational inputs. The system employs a surveillance orientation integrated with AI predictions, allowing an ability for clinicians to allocate staff appropriately. A pilot empirical evaluation demonstrated high agreement between clinician assessments and AI-generated priority rankings. Our findings suggest that AI-driven surveillance can enhance early VTE identification, optimize clinician workload, and improve patient care. Future work will focus on integrating additional clinical features and conducting a translational study for broader implementation.
Paper Number
1196
Recommended Citation
Russell, Stephen; Montes Suros, Fabio; Goodnow, Venessa; and Zeng, Jia, "VTE Surveillance: AI Enabled Human Machine Teaming for Veinous Thromboembolism" (2025). AMCIS 2025 Proceedings. 6.
https://aisel.aisnet.org/amcis2025/health_it/sig_health/6
VTE Surveillance: AI Enabled Human Machine Teaming for Veinous Thromboembolism
Venous thromboembolism (VTE) is a leading cause of preventable hospital death in the United States, significantly impacting patient outcomes and healthcare resources. This study explores the development of an AI-enabled VTE surveillance system designed to assist in clinical decision-making and resource management. Using machine learning models trained on electronic medical record (EMR) data from 57,490 inpatient encounters, we developed a predictive tool to help prioritize VTE potential while minimizing the need for complex laboratory and observational inputs. The system employs a surveillance orientation integrated with AI predictions, allowing an ability for clinicians to allocate staff appropriately. A pilot empirical evaluation demonstrated high agreement between clinician assessments and AI-generated priority rankings. Our findings suggest that AI-driven surveillance can enhance early VTE identification, optimize clinician workload, and improve patient care. Future work will focus on integrating additional clinical features and conducting a translational study for broader implementation.
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.
Comments
SIGHEALTH