Estimates the lowest concentrations an assay can reliably distinguish from a blank (LoB), detect (LoD), and quantify (LoQ), from blank and low-level replicate measurements.
Near zero, an assay's noise can masquerade as signal. Three limits pin down how low it can go. The Limit of Blank (LoB) is the highest reading you'd plausibly see from a true-zero sample — the 95th percentile of blank replicates. The Limit of Detection (LoD) is the lowest true concentration that reliably reads above the LoB, adding the low-level material's own scatter. The Limit of Quantitation (LoQ) is the lowest concentration you can report with acceptable precision (a target CV or total-error goal).
The parametric estimate uses LoB = mean_blank + c·SD_blank and LoD = LoB + c·SD_low, with c = 1.645 (5% error) corrected for the finite number of replicates; a non-parametric LoB (rank-based 95th percentile) is reported alongside for robustness. This follows the classical CLSI EP17 detection-capability approach.
A column of blank (zero-concentration) replicate measurements and a column of low-level replicate measurements, in concentration units. Optionally a target CV% for the LoQ.
LoB (parametric and 95th-percentile), LoD, and an LoQ estimate, plus the blank/low means, SDs, and the low-level CV.
Report a numeric result below the LoQ as "< LoQ", not as a number — its imprecision is too large to quantify. Results between LoB and LoD are "detected but not quantifiable". LoB/LoD/LoQ should be re-established whenever reagents, calibrators, or instruments change.