Live demo
Charting for bedside & remote care-management nurses, the EMR Bridge for IT, and a care-plan delta workflow — all on synthetic data. Nothing leaves the redactor.
Names, MRNs, dates, addresses, and phones are tokenized before any AI call.
Lovable AI Gateway with google/gemini-3-flash-preview. Same path production will use.
Every patient, ID, and date below is fabricated. Outputs are not medical records.
Charting pipeline: PHI redaction → AI draft → nurse sign.
Pick a case
Bedside, telephonic CCM, TCM, RPM, triage — all synthetic.
Encounter type
Drives the AI output template. Bedside uses your chosen SBAR/SOAP/Narrative/Handoff format below.
1 · Raw input
Synthetic note (editable)2 · Redacted
What the AI actually sees3 · AI-generated chart
Choose a format and runPhoto attach (constrained)
Local-only preview. EXIF/GPS stripped via canvas re-encode. No AI analysis.
Real patient photos are PHI and not supported here. Your selected photo never leaves this browser — no upload, no storage.
Drag & drop synthetic images here, or
JPEG/PNG/HEIC · re-encoded to JPEG · metadata removed
- AI wound staging (NPIAP stage, measurements, periwound signs)
- Medication label transcription (name, strength, sig)
- All inference behind a signed BAA; on-prem container option
Preview only — no analyze-image endpoint exists in this build.
What this demo proves — and what it does not
- The redaction happens in a server function before any model call. The raw input never reaches the AI gateway.
- Synthesis runs through Lovable AI Gateway with google/gemini-3-flash-preview.
- No requests are persisted; no audit rows are written from this demo page.
- This is a regex/heuristic redactor, not a HIPAA Safe Harbor de-identification service.
- Billing meters and time capture are documentation support — not billing advice. Thresholds are illustrative.
- No BAA, no SOC2 attestation, no production guarantees yet. Demo grade only.