Machine learning to improve the interpretation of intercalating dye-based quantitative PCR results.

This study aimed to evaluate the contribution of Machine Learning (ML) approach in the interpretation of intercalating dye-based quantitative PCR (IDqPCR) signals applied to the diagnosis of mucormycosis. The ML-based classification approach was applied to 734 results of IDqPCR categorized as positive (n = 74) or negative (n = 660) for mucormycosis after combining "visual reading" of the amplification and denaturation curves with clinical, radiological and microbiological criteria. Fourteen features were calculated to characterize the curves and injected in several pipelines including four ML-algorithms. An initial subset (n = 345) was used for the conception of classifiers. The classifier predictions were combined with majority voting to estimate performances of 48 meta-classifiers on an external dataset (n = 389). The visual reading returned 57 (7.7%), 568 (77.4%) and 109 (14.8%) positive, negative and doubtful results respectively. The Kappa coefficients of all the meta-classifiers were greater than 0.83 for the classification of IDqPCR results on the external dataset. Among these meta-classifiers, 6 exhibited Kappa coefficients at 1. The proposed ML-based approach allows a rigorous interpretation of IDqPCR curves, making the diagnosis of mucormycosis available for non-specialists in molecular diagnosis. A free online application was developed to classify IDqPCR from the raw data of the thermal cycler output ( http://gepamy-sat.asso.st/ ).

ß LACTA test performance for detection of extended-spectrum ß-lactamase-producing Gram-negative bacilli directly on bronchial aspirates samples: a validation study.

OBJECTIVES: Incidence of extended-spectrum ß-lactamase-producing Gram-negative bacilli (ESBL-PE-GNB)-related infections is worryingly increasing worldwide. ESBL-PE-GNB detection directly on bronchial aspirate samples (BAS) performed for suspected pneumonia may help save empirical carbapenems. Our objectives were to optimize ß-LACTA™ test (BLT) realization and evaluate BLT performance for ESBL-PE-GNB detection directly on BAS. METHODS: We studied BLT technical optimization using BAS of different matrix types spiked with increasing concentrations of CTX-M-15-producing Klebsiella pneumoniae; in vitro validation of BLT diagnostic performance on 17 ESBL enzymes, belonging to CTX-M, SHV, TEM, OXA, GES, VEB and PER groups; and clinical validation of BLT performance on 126 BAS prospectively collected from seven intensive care units. RESULTS: After optimization, BLT detected with 100% sensitivity the presence of CTX-M-15-producing K. pneumoniae spiked in sterile BAS for inoculums upon two or more GNB per field upon microscopic Gram staining examination (MGSE). The BLT accurately detected the 17 ESBLs tested at 106 CFU/mL and all ESBLs except Pseudomonas aeruginosa-related OXA-14 at 104 CFU/mL. Among the 126 BAS of the validation cohort, 21 (17%) gave positive BLT (ten in BAS positive and 11 in BAS negative on MGSE). All BLT-positive BAS grew with ESBL-PE-GNB, including five hyper-L2-producing Stenotrophomonas maltophilia strains. BLT detected ESBL-PE-GNB directly on clinical BAS positive for GNB on MGSE and/or growing with ≥104 CFU/mL with 100% sensitivity, specificity, and positive and negative predictive values. CONCLUSIONS: BLT is an accurate tool for ESBL-PE-GNB detection directly on BAS. Further studies are needed to evaluate the impact of BLT-guided early antimicrobial de-escalation strategies.

Selection during cefepime treatment of a new cephalosporinase variant with extended-spectrum resistance to cefepime in an Enterobacter aerogenes clinical isolate.

Enterobacter aerogenes resistant to cefepime (MIC, 32 microg/ml) was isolated from a patient treated with cefepime for an infection caused by a strain of E. aerogenes overproducing its AmpC beta-lactamase (MIC of cefepime, 0.5 microg/ml). The AmpC beta-lactamase of the resistant strain had an L-293-P amino acid substitution and a high k(cat)/K(m) ratio for cefepime. Both of these modifications were necessary for resistance to cefepime.