Use case 01 · clinical AMR
Carbapenem resistance: where the platform is already validated.
Carbapenems are a last-resort antibiotic. When an Enterobacterales isolate produces a carbapenemase — KPC, NDM, VIM, OXA-48, IMP — the clinician is near the end of the therapeutic line, and reported in-hospital mortality for carbapenem-resistant infections runs between 30 and 50% depending on the cohort. BELMAP tracks these resistances in Belgium under a category of their own. This is the use case where a few hours of diagnostic delay changes an outcome, which is why we validated here first.
Culture sets the clock. At least six hours of growth before any test can run — so everything after it had better be fast.
Immunochromatographic strips
Fast, but blind to the unexpected
Quick and simple, widely used as a first-line screen — but they recognise known enzyme families only. An uncovered variant or atypical expression can pass unseen, and they say nothing about actual enzymatic activity.
Rapid phenotypic tests
Functional, but hard to read
These measure activity rather than a predefined genetic panel, which is their great advantage. But colour reading is subjective, sensitive to inoculum and operating conditions, and ambiguous at low activity levels.
Reference methods
Complete, but too slow to act on
Full AST needs 24 to 48 hours. PCR runs €40–50 per test on dedicated infrastructure and detects genes whether or not they are expressed. BELMAP frames both as complements to phenotypic work, not replacements.
The gap, and what we add
A pre-antibiogram, not another test to reconcile.
No universal phenotypic test returns robust information in under fifteen minutes, and no single device couples a rapid screen to a digitalised functional confirmation. So laboratories run several tests in cascade — complex interpretation, late decision, no native digital trace of how the conclusion was reached.
Our contribution is a short sequential protocol. A lateral flow assay built for phenotype — targeting the functional hydrolysis of sentinel antibiotics rather than an enzyme's shape — triages within fifteen minutes. The BYG electrochemical read then confirms and characterises the mechanism, measuring hydrolysis kinetics by differential pulse voltammetry. Molecular recognition on one side, enzyme kinetics on the other: two signals of different natures on the same sample, combined into one digital verdict.
We are not proposing to replace the reference methods. We are shortening the interval in which a clinician has nothing to act on — what the literature calls diagnostic stewardship: changing how tests are ordered, run and reported so that treatment improves, resistance pressure falls, and laboratory resources go where they matter (Fabre et al., 2023).