Real-time artificial intelligence system for bacteremia prediction in adult febrile emergency department patients

• Published in: International journal of medical informatics (Shannon, Ireland) Ireland), 2023-10, Vol.178, p.105176-105176, Article 105176
• Main Authors: Tsai, Wei-Chun, Liu, Chung-Feng, Ma, Yu-Shan, Chen, Chia-Jung, Lin, Hung-Jung, Hsu, Chien-Chin, Chow, Julie Chi, Chien, Yu-Wen, Huang, Chien-Cheng
• Format: Article
• Language: English
• Subjects: Adult; Artificial intelligence; Bacteremia; Emergency department; Febrile; Hospital information system; Implementation; Random forest
• ISSN: 1386-5056; 1872-8243
• DOI: 10.1016/j.ijmedinf.2023.105176

•Bacteremia can be predicted among adult febrile patients in the emergency department (ED) via an implemented real-time AI system.•To predict the risk of bacteremia, random forest AI model is more accurate than other four AI models and qSOFA score.•This real-time AI prediction system could have the collateral benefit of reduce up to 25% of unnecessary empiric antibiotics among adult febrile patients in the ED. Artificial intelligence (AI) holds significant potential to be a valuable tool in healthcare. However, its application for predicting bacteremia among adult febrile patients in the emergency department (ED) remains unclear. Therefore, we conducted a study to provide clarity on this issue. Adult febrile ED patients with blood cultures at Chi Mei Medical Center were divided into derivation (January 2017 to June 2019) and validation groups (July 2019 to December 2020). The derivation group was utilized to develop AI models using twenty-one feature variables and five algorithms to predict bacteremia. The performance of these models was compared with qSOFA score. The AI model with the highest area under the receiver operating characteristics curve (AUC) was chosen to implement the AI prediction system and tested on the validation group. The study included 5,647 febrile patients. In the derivation group, there were 3,369 patients with a mean age of 61.4 years, and 50.7% were female, including 508 (13.8%) with bacteremia. The model with the best AUC was built using the random forest algorithm (0.761), followed by logistic regression (0.755). All five models demonstrated better AUC than the qSOFA score (0.560). The random forest model was adopted to build a real-time AI prediction system integrated into the hospital information system, and the AUC achieved 0.709 in the validation group. The AI model shows promise to predict bacteremia in adult febrile ED patients; however, further external validation in different hospitals and populations is necessary to verify its effectiveness.