Comparison of an Automated Assessment System (APAS Independence) and Artificial Intelligence (AI) to Manual Plate Reading of Methicillin-resistant Staphylococcus aureus Chromagar Surveillance Cultures

Microbiology is a discipline that is heavily predicated upon image interpretation, whether it is by microscopy of stained smears or manual interpretation if colony morphology of organisms growing on a broad range of media. Therefore, platforms that use artificial intelligence image analysis to automate interpretation and separate negative cultures from those requiring technologist intervention could increase both laboratory efficiency and test accuracy. The Automated Plate Assessment System (APAS® Independence) is an automated imaging station linked with interpretive software that detects colonies on chromogenic media and triages samples into categories of negative or presumptive positive.

Over a 3-month period, 4,603 BBL CHROMagar MRSA pates were read in parallel between APAS® Independence and manual reading. The results showed no false negatives as the APAS Independence achieved 100% positive percent agreement. The APAS Independence identified a further 3 positive MRSA samples, previously missed by manual reading.

Poster Presentation: by Johns Hopkins Hospital Conference: ASM Microbe Online 2020

Date: June 2020

Authors: Gammel N, Ross T, Lewis S, Olson M, Henciak S, Harris R, Carroll K.C

Citation: Gammel N, Ross T, Lewis S, Olson M, Henciak S, Harris R, Carroll K.C., ASM Microbe 2020. Comparison of an Automated Plate Assessment System (APAS Independence) and Artificial Intelligence (AI) to Manual Plate Reading of Methicillin-resistant Staphylococcus aureus Chromagar Surveillance Cultures.