GIFSQ Tools · Module 4 — Microbiology Methods & Laboratory Operations · Tool 73 of 500
Calculate the limit of detection (LOD) from blank measurements and the calibration slope, per ISO 11843 and the IUPAC definition.
Try sample values — fills the form with the worked example below.
–
LOD signal (blank + 3.3 s)
–
Detection limit (CFU/g)
–
LOQ for reference, 10 s ÷ slope (CFU/g)
Equation: LOD signal = blank mean + 3.3 × s, where s is the standard deviation of the blank. LOD (concentration) = LOD signal ÷ slope. For reference, LOQ = (blank mean + 10 × s) ÷ slope.
Method: ISO 11843-2 (capability of detection) and the IUPAC definition of the limit of detection: the lowest amount that can be detected with reasonable certainty, distinct from the limit of quantification.
Assumptions: the blank measurements are normally distributed; the calibration is linear through the low range; the slope is constant near the detection limit.
Limitations: this is a statistical detection limit for the measurement system, not a method detection limit for a food matrix — matrix effects, recovery and enrichment steps can raise the practical limit. A result below the LOD is reported as not detected, never as zero.
Rounding: LOD and LOQ to 3 decimal places; the signal to 4 decimal places. The 3.3 factor is applied to unrounded values.
This is a calculation aid, not a laboratory report — confirm against your accredited method before citing a value.
Method: ISO 11843-2 (capability of detection) and the IUPAC definition of the limit of detection: the lowest amount that can be detected with reasonable certainty, distinct from the limit of quantification.
Assumptions: the blank measurements are normally distributed; the calibration is linear through the low range; the slope is constant near the detection limit.
Limitations: this is a statistical detection limit for the measurement system, not a method detection limit for a food matrix — matrix effects, recovery and enrichment steps can raise the practical limit. A result below the LOD is reported as not detected, never as zero.
Rounding: LOD and LOQ to 3 decimal places; the signal to 4 decimal places. The 3.3 factor is applied to unrounded values.
This is a calculation aid, not a laboratory report — confirm against your accredited method before citing a value.
Worked example: Blank mean 0.05, blank standard deviation 0.02, slope 0.5 signal per CFU/g. LOD signal = 0.05 + 3.3 × 0.02 = 0.1160; LOD = 0.1160 ÷ 0.5 = 0.232 CFU/g; the reference LOQ is 10 × 0.02 ÷ 0.5 = 0.400 CFU/g.
Enter values above to see results.
A detection limit is a property of the method and the matrix, not a fixed number for an organism. Re-establish it when the matrix, the enrichment or the instrument changes. Tool 73 of 500 — GIFSQ
