Try Records
Fix one value. See what changes.
Start with a prepared synthetic Patient. Understand a terminology finding, correct the value and check the result.
No account or installation required · Synthetic and public test data
Your first correction
Find the issue.
Choose Validate resource, then follow the finding at
name[0].use.Correct one value.
In JSON, replace
officiallwithofficial.Check what changed.
Choose Revalidate, then review the resolved findings under Changes.
The sandbox uses a temporary test copy. Your correction does not change a source server.
Choose your next question.
Explore a public server, compare profiles or investigate a dataset.
Browse
Follow a finding to its field.
Browse public FHIR resources and inspect the requirement behind a validation finding.
Browse public serversProfiles
Compare profile requirements.
Check the same synthetic Patient against FHIR R4 and ISiK, then inspect what each profile requires.
Compare the profilesData quality
Explain a quality indicator.
Explore a prepared dataset and trace a quality indicator to the affected resource.
Try the dataset exampleSYNDERAI · Release 5
Start with a complete synthetic document.
Open a laboratory report, patient summary or discharge report. Follow its embedded resources and inspect their profile requirements and findings. Both sources can be browsed without an account.
Public demo
SYNDERAI on Medplum
Explore the Release 5 laboratory, patient-summary and discharge-report Bundles. Open a Bundle to find its embedded resources.
Public demo
SYNDERAI on Health Samurai
Explore laboratory and patient-summary Bundles on Aidbox. Discharge reports are excluded from this source after an R4 status-binding rejection.
The two sources have different document scopes. A resource inside a Bundle is not necessarily a separately searchable resource. These are synthetic examples; findings and availability can change.
Watch a correction and regression checkSee the correction
One value changes. Two findings resolve.
Replace officiall with official in the synthetic Patient, then run the checks again.
After correction, two terminology findings resolve and four information findings remain. The test copy is not written to the source server or saved as workspace evidence.
Use synthetic test data.
Public servers are shared and their contents can change. Review which resources were checked and read the basis and limits with each result.
Bring your own workflow.
For private servers, recurring checks, quality policies or shared baselines, plan a workspace evaluation around one delivery and a finding your team needs to resolve.
Plan your evaluation