Legion Diagnostics

Running a verification or validation experiment

Create an experiment for an analyte, work through its statistical modules, and finalize and sign it when every module is complete.

An experiment is one analyte on one analyzer, and it holds the statistical modules you run for it. You find experiments under validation and verification.

Creating an experiment

Click "New Experiment", then "Single-analyte experiment". Give it a unique title, and pick the site group, lab area, analyzer and analyte. If a plan exists for that analyte, choose it to fill in the design; the experiment then takes the plan's study type. Without a plan, choose the "Study type" yourself: verification or validation. You can also start from a plan by clicking "Create Experiment", or add a plan to an experiment later, which also sets its study type to match the plan.

The experiment page shows its setup, links to the analyte's manufacturer claims and the plan, and an audit trail.

The statistical modules

The module grid lists every module that applies to the analyte, each marked "Not Started", "Pending" or "Completed":

  • Precision
  • Method Comparison and Multi-Instrument Method Comparison
  • Qualitative Method Comparison (qualitative assays)
  • Calibration Verification & Linearity
  • Reference Interval
  • Reagent Lot Comparison
  • Limit of Blank and Limit of Quantitation
  • Carryover and Interference
  • Factor Sensitivity, Manual INR Check, INR Geometric Mean & VRI, and PT/INR Method Comparison for coagulation

Qualitative analytes show only precision and qualitative method comparison. Quantitative analytes show the rest.

Each module works the same way. Enter or upload your data, review it in the "Data Entry" form, and submit. The analysis runs on its own, the results and charts appear, and you sign the module's report. The tutorials for each module explain its design and results in detail.

Finishing the experiment

When the modules you need are complete, generate a composite report and click "Finalize & Sign". See Composite reports and signing an experiment.