Verified clinical-context layer
Give your clinical software a patient memory it can trust.
MedLineage is the verified clinical-context layer beneath your CRM, EHR, and clinical AI — one source-cited patient history your tools can read, act on, and prove.
100% provenance coverage · 1.00 extraction F1 on our published synthetic benchmark — raw JSON public, drift fails CI. JSON →
Prepares records for clinician review. Does not diagnose, does not recommend treatment, does not replace clinicians. Best-effort de-identification, not certified anonymization.
- Multi-format ingest
- Every claim cited
- Agents + human approval
- 5 languages
02What your software gains
A patient memory your software can act on.
- 01
One memory across every record
Your software finally sees the whole patient — PDFs, HL7, FHIR, lab CSVs, in five languages, with no clean-data assumption.
- 02
Context it can cite, not guess
Every fact traces to the source line it came from, so AI stops hallucinating the chart.
- 03
Verification that compounds
Clinicians attest into a tamper-evident ledger; trust becomes permanent and measurable.
- 04
Drop-in, alongside your stack
One API, MCP, and FHIR — MedLineage sits beneath your product; it does not replace your EHR.
- 05On the roadmap
Portable with the patient
On the roadmap: the same verified history follows the person between providers.
03Agent-ready
Grounded answers, by construction.
No SDK, no rewrite. Add one MCP server entry to the agent you already run — Claude Desktop, Claude Code, or any MCP client — and it gains read-only tools that return source-cited, ledger-attested patient context.
{ "mcpServers": {+ "medlineage": {+ "command": "python",+ "args": ["-m", "mcp_server"],+ "env": {+ "MEDRECORD_MCP_ENABLED": "true",+ "MEDRECORD_PATIENT_GRAPH_ENABLED": "true",+ "MEDRECORD_MCP_DEFAULT_PATIENT_ID": "pat_demo_neph"+ }+ } } }
{ "claims": [ { "text": "CEA 8.2 ng/mL (rising)", "citation": "[obs_4a]", "source_document_id": "doc_2", "verification_status": "attested" }, ... ], "counts": { "total": 14, "verified": 9 } }
Add one MCP server
Drop the medlineage entry into your MCP config (Claude Desktop/Code — or run it hosted at /mcp). No SDK, no schema changes.
Your agent gains the tools
get_verified_context, ask_patient, check_claims, verify_claim appear automatically. The patient is bound by the X-Patient-Handle header — never passed in a prompt.
Cited & ledger-attested
Every claim returns with an [obs_*] citation, its source document, and a Provenance-Ledger status — compare it with raw-PDF parsing via compare_context.
- ask_patient
- get_patient_context
- assess_readiness
- list_evidence_gaps
- resolve_citation
- verify_claim
- get_fhir_bundle
- get_verified_context
Shipped behavior, demonstrated on synthetic cases — flag-gated per deployment.
04Verification that compounds
Every clinician sign-off makes the layer harder to copy.
When a clinician verifies a fact, the verdict is signed into a tamper-evident ledger and reused for every future read. Demand for a fact prioritizes its review; verified context accrues.
- 01
An agent relies on a fact
- 02
A clinician verifies it
- 03
Every future agent reads verified context
05Measured, not promised
Numbers that reproduce from a pinned benchmark.
Synthetic multilingual gold set — not a clinical study. The scoreboard is committed to the repo and pinned in CI: any drift fails the build.
253 synthetic gold facts · 5 languages
every recovered fact cites its source document
no synthetic case tripped the safety net
Clinical CRM
Give care and sales teams a verified patient history inside the tools they already use.
EHR / EMR
Surface source-cited context beside the chart — no migration, no schema change.
Patient portals
Let patients see a structured, source-cited history they can actually understand.
Clinical AI & agents
Serve context other agents cite over MCP instead of re-parsing PDFs.
Our first clinical pilot is live with Sanity Health. Sanity Health is a small private clinic in Milan with around 500 patients. The clinic pushes its inbound clinical records straight into MedLineage Connect — we turn them into a source-cited patient-context layer its clinicians can review and attest, with every finding tracing back to the document, page, and source language it came from. Pilot in progress. MedLineage prepares records for clinician review — it does not diagnose, recommend treatment, or replace clinicians.
06On the roadmap
One day, your history follows you — not the building.
We are building toward a patient-controlled, consent-gated view of the same verified graph: a portable, source-cited history a patient could carry to a new hospital or specialist. This is where MedLineage is heading; it is not available yet.
- Patient-held access (planned)
- Consent-gated sharing (planned)
- Source-cited, portable history (planned)
- Every claim cites its document, page and source language
- Tenant-isolated · best-effort de-identification · dry-run by default
- Multi-format ingest & FHIR-like exports — no EHR migration, no lock-in
Not a medical device. Not certified for FHIR, GDPR or EHDS conformance. No diagnostic capability or anonymization guarantee is claimed.
Start with one packet. Grow into an agentic record-prep layer.
Generate a clinician-ready PDF today. Re-upload tomorrow to thicken the same Patient Graph. Let safe agents open the next review task and prepare a signed Data Room when you're ready to integrate — every step gated for human approval.
Working with hospitals or second-opinion services? Bring 5 real cases — we run them de-identified and you sign off the packets. Run a de-identified pilot
Coordinates record-preparation work. Does not diagnose, does not recommend treatment, does not replace clinicians. No certified FHIR / GDPR / EHDS / MDR / CE conformance is claimed.