Evaluation of the use of artificial intelligence for the detection of VRE using two different agar types
This study was conducted to evaluate artificial intelligence (AI)-based VRE-detection algorithms in combination with different chromogenic culture media. Fast and reliable microbiological diagnostic is crucial for targeted antimicrobial therapy. Culture media and reading protocol are essential for minimizing time to result.
Over a 3-month period, 478 patient samples inoculated on Brilliance VRE® agar (Thermo Fisher) and ChromID® VRE agar (bioMérieux) were evaluated for VRE detection. APAS Independence (CCS) AI-Algorithms was used to classify growth after 24 and 48 hours. All plates were also read by experienced medical laboratory technicians and microbiologists after 48 hours. Results of AI-based classifications were compared to conventional plate reading.
Among 478 samples, a total of 27 (5.6%) were positive for VRE. At 48 hours, there was 100% agreement between AI-based and conventional reading of all culture media. The results from this study indicate, that using APAS Independence (CCS) AI-Algorithms for detection of VRE is a good alternative to conventional plate reading and works with both Brilliance VRE® and ChromID® VRE plates (100 % sensitivity at 48 hours).
Poster Presentation: by Labor Dr Wisplinghoff
Conference: ECCMID 2021, Online
Date: July 2021
Authors: Giglio S, Jazmati N, Nowag A, Pohl B, Quante X, Wirth S, Wisplinghoff H
Citation: Nowag A, et al. ECCMID 2021. Evaluation of the use of artificial intelligence for the detection of VRE using two different agar types