Deep Learning for Predicting Phlebitis in Patients with Intravenous Catheters

Abstract

This study presents a deep learning model to predict phlebitis in patients with peripheral intravenous catheter (PIVC) insertions. Leveraging electronic health record data from 27,532 admissions and 70,293 PIVC events at a hospital in Seoul, South Korea, the study involved analyzing patient demographics, PIVC-specific features, and drug-related information. The developed deep learning model was benchmarked against various machine learning models, demonstrating superior performance with an accuracy of 0.93 and an AUC of 0.89. This highlights its potential as an effective tool for early detection of phlebitis, promising enhanced patient outcomes and healthcare efficiency.