Simple and Rapid Discrimination of Methicillin-Resistant Staphylococcus aureus Based on Gram Staining and Machine...

Area of Science:

• Clinical microbiology
• Bacteriology
• Diagnostic technology

Background:

• Methicillin-resistant Staphylococcus aureus (MRSA) poses significant clinical challenges due to high morbidity and mortality.
• Current methods for detecting antibiotic resistance often require lengthy incubation periods.

Purpose of the Study:

• To develop a simple, rapid identification method for MRSA.
• To significantly reduce the time required for antibiotic resistance detection.

Main Methods:

• Utilized oxacillin sodium salt, a cell wall synthesis inhibitor, in conjunction with Gram staining.
• Employed machine vision (MV) analysis to detect color changes indicative of bacterial response.
• Applied machine learning models, including linear discriminant analysis (LDA) and artificial neural network (ANN), for data interpretation.

Main Results:

• Methicillin-susceptible S. aureus (MSSA) cell walls were destroyed by oxacillin, appearing Gram-negative, while MRSA remained Gram-positive.
• MV analysis successfully detected these color changes.
• LDA and ANN models achieved high accuracies of 96.7% and 97.3% for MRSA identification, respectively.

Conclusions:

• The combined approach of oxacillin, Gram staining, and MV analysis offers a rapid and simple method for MRSA identification.
• This strategy significantly shortens detection time to under one hour, avoiding overnight incubation.
• The method shows potential for broader application in detecting antibiotic resistance in other clinical bacteria.