Name of session: Machine Learning Prediction of Central Line Associated Bloodstream Infections (CLABSI)
Presented by: Evan Sylvester, MPH, AL-CIP, CIC, LTC-CIP, WFR, MT(ASCP) and Laura Ebinger, CIC
Key learnings:
a. While there are environmental considerations for use, using tools like AI can increase efficiency and maximize increasingly limited IP resources; it's not taking jobs, it's amplifying our work! Hello, new power couple!
b. AI includes Machine Learning, predictive models, Natural Language Processing (NLP), and Computer Vision. Machine Learning (ML) is a branch of AI that enables algorithms to uncover hidden patterns within datasets – it can uncover new, similar relationships without explicit intervention of programming for each new task!
c. Predictive modeling (there are multiple types) can analyze data and look for patients that are higher risk for developing a CLABSI, prompting early intervention prior to infection development. High risk patients can be identified for closer attention vs low risk patients that may not need current intervention or as much IP review, if any. Some data elements include demographic data, device-related info (CL type and location, CL days, etc), vitals, medical conditions, test results, and medications. The presenter spoke of their model being used to identify patients at risk for a CLABSI in the next 24-72 hours!
How I will apply learnings: I love to find safe and FUN opportunities to use AI! I will continue evaluating AI tools for ways to use them efficiently to amplify IP work and maximize resources. Knowing more ways AI is used in IP helps drive educated, informed decisions when asked to weigh in on new products or tools!
-------------------------------------------