Central Line-Associated Bloodstream Infection (CLABSI) Prediction Model

­

Application

A predictive model of central line-association bloodstream infection (CLABSI).

Key Benefits

  • Specifically designed to improved predictive analysis of patients who are at risk for CLABSI.
  • Scored higher predictive values on average than similar predictive modeling without the hypotheses driven approach (Bidirectional LSTM with Focal Loss and Attention Mechanism).

Market Summary

Central venous catheters (CVC) are commonly used in critically ill patients and offer several advantages to peripheral intravenous access. However, indwelling CVCs have the potential to lead to bloodstream infections, with the risk increasing with an array of characteristics such as catheter choice, catheter location, insertion technique, and catheter maintenance. Central line-associated bloodstream infections (CLABSIs) are a significant cause of healthcare-associated infections among hospitalized children and contribute to increased morbidity, length of hospital stay, and cost. Early detection and treatment of CLABSIs can prevent adverse outcomes, reduce costs, and improve the quality of care.

Technical Summary

Researchers have developed a predictive model to predict the onset of central line-associate bloodstream infection (CLABSI) in children with a central during the next 48 hours of their hospitalization utilizing electric health record data. Researchers found that Temperature and Platelet count were the 2 largest factors indicating the onset of CLABSI in a juvenile patient. The researchers’ model combined the factors of Bidirectional LSTM with Focal Loss and Attention Mechanism outperformed the other model examples in every category except for Sensitivity and Negative Predicative Value indicating the model is capable of being implemented in a real-time setting and serve as a clinical support system.

Patent Information

Tech ID: 21042
Published: 7/6/2022