LoD estimator

FAQ

What is LoD versus LoB?

Limit of blank (LoB) is the highest measurement likely to be observed for a blank sample. Limit of detection (LoD) is the lowest analyte amount that can be detected with a stated confidence. Classic blank-based sketches often use LoB ≈ mean_blank + 1.645×SD_blank; full LoD methods add a low-level precision term. CLSI EP17 is a commonly cited evaluation protocol for LoB, LoD and LoQ.

Why does this tool default z to 1.645?

1.645 is the approximate one-sided 95% normal quantile often used in blank-based LoB sketches. You can change z, but the official claim still depends on the study design, replicates and error model in your protocol — not on this form alone.

Does this replace a CLSI EP17 study?

No. EP17-style studies specify blank and low-level sample plans, replicates across days/instruments, and how LoB, LoD and LoQ are estimated and verified. This page only computes mean + z×SD from values you enter.

How should LoD relate to clinical cutoffs?

Analytical LoD is not automatically a clinical decision threshold. Intended-use cutoffs, measuring interval and precision near the cutoff belong in the clinical validation and labeling discussion.

What units should I use?

Use the same concentration or signal units for blank mean, blank SD and the reported LoD. Mixing units silently invalidates the arithmetic.

When is a simple blank model a bad fit?

Heteroscedastic assays, censored blanks, non-normal errors and multi-reagent lots often need richer models. If blanks are not normally distributed, do not treat mean + 1.645×SD as definitive.