Mobiolink

Multi-level validation

A configurable, multi-level validation framework safeguards the integrity of diagnostic data before export. Two standard levels (Val1 and Val2) enforce the four-eyes principle seamlessly, while borderline or erroneous results are automatically flagged for manual review. Mobiolink prepares, sorts and traces; the final decision always rests with the validator.

Architecture and safety levels

Validation relies on an access-control matrix that ensures compliance and data fidelity. Each target result passes through up to two standard levels. Mobiolink never makes a diagnostic decision on the validator's behalf and stays outside the scope of the IVDR: the software prepares the review and records it in a fully traceable way.

  • Level 1 validator (Val1): performs the first technical or clinical review.
  • Level 2 validator (Val2): performs the final verification; when assigned to different people, this enforces a strict four-eyes principle.
  • Expert validation: activated manually in cases of high uncertainty; the result stays locked until an authorised person makes the final decision.

Interactive validation workflows

Depending on assay design and lab preference, validation runs assay-wise (all results of a run at once, optimised for high throughput) or target-wise (one target across several samples, for focused assessment). For clear, compliant runs a single-click global validation moves every standard result forward; for anomalies, a detail view opens the raw data, allows measured values to be adjusted (with the right permissions), or flags a sample for a repeat run.

Automatic gates and compliance rules

Automatic criteria are built directly into the logic, sparing the validator avoidable sources of error:

  • Flagged results: any result classified as borderline or erroneous is excluded from one-click validation and must be reviewed individually.
  • Mandatory justification: accepting an erroneous or borderline result requires a formal justification; for quality-control (QC) samples outside the reference values it is strictly required.
  • Multiplex dependencies: for multiplex assays, every target result of a sample must be validated before a single result of that sample can be exported.
  • Run interlocks: export of every measured value stays blocked until all run KPIs and QC samples are validated or fall within the reference values.
Next Repeat management