Same-day antimicrobial susceptibility test using acoustic-enhanced flow cytometry visualized with supervised machine learning

• Published in: Journal of medical microbiology Vol. 69; no. 5; pp. 657 - 669
• Main Authors: Inglis, Timothy J J, Paton, Teagan F, Kopczyk, Malgorzata K, Mulroney, Kieran T, Carson, Christine F
• Format: Journal Article
• Language: English
• Published: England Microbiology Society 01.05.2020
• Subjects: Acoustics; Anti-Infective Agents - pharmacology; Bacteria - drug effects; Editor's Choice; Flow Cytometry - methods; Flow Cytometry - standards; Humans; Microbial Sensitivity Tests - methods; Microbial Sensitivity Tests - standards; Software; Supervised Machine Learning; Time Factors; Workflow; flow cytometer; machine learning; antimicrobial susceptibility test; Escherichia coli; Klebsiella pneumoniae; Staphylococcus aureus
• ISSN: 0022-2615; 1473-5644
• DOI: 10.1099/jmm.0.001092

Antimicrobial susceptibility is slow to determine, taking several days to fully impact treatment. This proof-of-concept study assessed the feasibility of using machine-learning techniques for analysis of data produced by the flow cytometer-assisted antimicrobial susceptibility test (FAST) method we developed. We used machine learning to assess the effect of antimicrobial agents on bacteria, comparing FAST results with broth microdilution (BMD) antimicrobial susceptibility tests (ASTs). We used (1), (1) and (2) strains to develop the machine-learning algorithm, an expanded panel including these plus (2), (3), (1), (1), (2) and (1), tested against FAST and BMD (Sensititre, Oxoid), then two representative isolates directly from blood cultures. Our data machines defined an antibiotic-unexposed population (AUP) of bacteria, classified the FAST result by antimicrobial concentration range, and determined a concentration-dependent antimicrobial effect (CDE) to establish a predicted inhibitory concentration (PIC). Reference strains of and tested with different antimicrobial agents demonstrated concordance between BMD results and machine-learning analysis (CA, categoric agreement of 91 %; EA, essential agreement of 100 %). CA was achieved in 35 (83 %) and EA in 28 (67 %) by machine learning on first pass in a challenge panel of 27 Gram-negative and 15 Gram-positive ASTs. Same-day AST results were obtained from clinical (1) and (1) isolates. The combination of machine learning with the FAST method generated same-day AST results and has the potential to aid early antimicrobial treatment decisions, stewardship and detection of resistance.