Implemented pilot · Life Sciences & Biostatistics
Clinical QC Workspace: implemented pilot
Clinical QC Workspace is an implemented pilot for CROs and mid-sized pharmaceutical companies. It brings clinical-data intake, study-specific quality checks, findings review and reporting into one browser workspace. Independent R jobs continue after the browser closes; a completed job and an accepted quality outcome are reported separately.
Leap Dynamics GmbHUpdated:
Data intake
Upload raw, SDTM or ADaM datasets, or select an authorised server delivery.
Reviewer workspace
Configure study checks and inspect progress, logs and findings in R Shiny.
Independent execution
A background worker runs isolated R jobs outside the browser session.
Results and evidence
Review searchable findings and export reports linked to the delivered inputs.
A review workflow around each data delivery
A dataset delivery creates several different questions: can the files be processed, are subjects and dates consistent, and do the findings meet the study’s criteria? The pilot keeps these questions connected without treating successful processing as proof of data quality.
Checks cover dataset structure, subject consistency, dates, treatment exposure, adverse events, laboratory results and selected analysis derivations. Study-specific configuration determines the checks and acceptance criteria.
Execution status and quality gates
Reviewers can search and filter findings. Optional Define-XML comparisons and Pinnacle 21 result imports bring additional metadata and validation evidence into the review workflow.
| Outcome | What it tells the reviewer |
|---|---|
| Execution status | Whether the processing run completed or needs technical attention. |
| Quality-gate outcome | Whether the evaluated results meet configured acceptance criteria. A completed run can still contain findings. |
Outputs that remain connected to their inputs
The described functions are implemented pilot capabilities. They are not a measured claim about faster review or fewer errors in a production study.
- Detailed Excel workbook, standalone HTML summary and structured CSV/JSON exports.
- Input snapshots, file fingerprints, configurations and execution logs linking each delivery to its results.
- PostgreSQL for study, access and job history; artifact storage for snapshots and reports.
Deployment and the first discussion
Primary deployment is inside the customer’s environment on Linux, with browser access from Windows and Linux. Optional dedicated managed hosting is specific to this platform and scoped separately.
Start with the dataset formats, an approved representative delivery, the intended checks, review owners and operating responsibilities. The pilot complements formal CDISC validation and independent statistical programming QC. Agree those boundaries and the acceptance evidence before planning a study-specific evaluation.