Automated Scoring of Chromogenic Media for the Detection of MRSA Using the WASPLab Image Analysis Software
Categories: Automation, MRSA Swab Sampling
Publication Date: 12/30/2015
Conference or Journal: Journal of Clinical Microbiology
Author(s): Matthew L. Faron, Blake W. Buchan, Chiara Vismara, Carla Lacchini, Alessandra Bielli, Giovanni Gesu, Theo Liebregts, Anita van Bree, Arjan Jansz, Genevieve Soucy, John Korver, and Nathan A. Ledeboer
Recently systems have been developed to create total laboratory automation for clinical microbiology. These systems allow for automation of specimen processing, specimen incubation and imaging of bacterial growth. In this study we used the WASPLab to validate software that discriminates and segregates positive and negative chromogenic MRSA plates by recognition of pigmented colonies. A total of 57,690 swabs submitted for MRSA screening were enrolled in the study. Four sites enrolled specimens following their standard of care. Chromogenic agar used at these sites included: MRSASelect (Bio-Rad Laboratories, Redmond, WA), ChromID MRSA (BioMeriéux, Marcy-I’Etoile, France) and CHROMagar MRSA (BD Diagnostics, Sparks, MD). Specimens were plated and incubated using the WASPLab. The digital camera took images at 0 and 16-24h and the WASPLab software determined the presence of positive colonies based on a HSV (Hue, Saturation, Value) score. If the HSV score fell within a defined threshold, the plate was called positive. Performance of the digital analysis was compared to manual reading. Overall the digital software had a sensitivity of 100% and a specificity of 90.7% with the specificity ranging between 90.0 and 96.0 across all sites. Results were similar when using the three different agars with sensitivity observed to be 100% and specificity ranging from 90.7 and 92.4%. These data demonstrate that automated digital analysis can be used to accurately sort positive from negative chromogenic agar cultures regardless of pigmentation produced.