Team Mola Mola

Venous thromboembolism (VTE) is a preventable condition that includes both pulmonary embolism (PE) and deep venous thrombosis (DVT). VTE is responsible for more than 100,000 deaths a year in the U.S. Diagnosing VTE is particularly problematic in the intensive care setting. In spite of awareness and prophylaxis, studies show a VTE incidence of up to 10% in medical-surgical ICUs. Existing clinical scoring systems have poor sensitivity and rely on patient reported symptoms and routine vital signs measurements. Furthermore, treatment or prophylaxis with heparin carries significant risks, making accurate risk assessment crucial to balancing the risk-benefit ratio of intervention. Using high frequency physiologic data and information about underlying disease states, we are developing a novel risk assessment algorithm using machine learning methods to develop a more accurate, sensitive, and specific model for predicting VTE. Such a model could provide clinicians with more timely and reliable information to prompt definitive testing while decreasing unnecessary testing.