Sepsis accounts for 215,000 deaths in the US each year. The average cost of treating a case of sepsis is $22,100, with annual total costs of $16.7 billion nationally. Early detection of sepsis in the critical care setting is essential to improve outcomes, reduce length of stay and contain economic costs. The project focuses on analysis of continuous patient telemetry, e.g., bedside monitor data, using advanced anomaly detection algorithms to provide early sepsis detection in the critical care setting. The algorithms will be used to generate a model of normal patient behavior from selected cases of complication-free patient cases. The model provides a dynamic baseline for comparison with real-time monitored data, with the aim of revealing subtle indications of incipient sepsis when it occurs. The goal of the project is to provide software monitoring of hospital data for earliest possible sepsis detection. It is envisioned this continuous telemetry analysis platform can be expanded to detect a variety of important health conditions in monitored patients at critical early stages.
Sepsis kills 215,000 people in the US each year and costs $16.7 billion annually. Early detection of sepsis in the critical care setting is essential for successful treatment. The project will develop software to automatically detect sepsis early in hospitalized patients.
