Referral intake
Turn inbound referral faxes into structured referrals matched to the right patient.
Documents · Intake
Medical Document Processing Automation
An agent that ingests inbound faxes, PDFs, and scanned forms, reads them with OCR, extracts the fields that matter, and posts structured, FHIR-ready data to the chart, with a human check on anything uncertain.
Human-in-the-loopHIPAA-ReadyAudit LoggedFHIR R4
Trusted by global innovators
The fax machine still runs healthcare
AI document processing works the fax and scanned-PDF queue that still carries referrals, prior-auth packets, lab reports and intake forms. Agnotic Technologies builds the agent to OCR each page, classify the document type, extract structured fields and match the right patient before posting to the chart as FHIR resources.
Without it, staff retype each document into the EHR, so documents sit for days and details get miskeyed. Anything low-confidence is flagged for a quick human review instead of silently guessed.
Architecture
Reference architecture for how this agent connects to your EHR and data systems.
Key Capabilities
Concrete document-handling steps: from raw scan to structured, matched, chart-ready data.
OCR for messy inputs
Reads low-quality faxes, multi-page PDFs, and handwritten forms, deskewing and cleaning pages for reliable extraction.
Document classification
Identifies whether a document is a referral, lab report, prior-auth packet, or intake form and routes it to the right template and queue.
Structured field extraction
Pulls demographics, ordering provider, diagnoses, medications, and results, normalizing codes to ICD-10, LOINC, and SNOMED.
Patient + encounter matching
Matches each document to the correct patient and encounter, flagging ambiguous matches rather than attaching to the wrong chart.
Who runs it
Where the document agent runs
Lab + imaging results
Extract results from faxed reports into DiagnosticReport resources.
Prior-auth packets
Parse payer correspondence and attach it to the right authorization workflow.
New-patient intake
Digitize scanned intake and history forms into structured demographics and problems.
What document automation changes
Directional impact, validated per deployment rather than guaranteed.
Auto-triage
Every inbound document classified and routed
Field-level
Confidence scoring gates what needs review
FHIR-native
Structured output written to the chart
Compliance-First Healthcare App Development Services Backed by Global Standards
15-Minute Scoping Call01
HIPAA
Health Insurance Portability and Accountability Act
Protect PHI with privacy-first architecture, encrypted storage and transmission, strict access controls, and traceable audit logs.
02
GDPR
General Data Protection Regulation
Implement lawful consent flows, data minimization, retention controls, and secure processing for sensitive health data.
03
FHIR
Fast Healthcare Interoperability Resources
Enable standardized health data exchange across apps, care teams, and systems through standards-based FHIR APIs.
04
HL7
Health Level Seven International
Support enterprise-grade interoperability with HL7-based integrations for records, events, and clinical messaging workflows.
05
HITRUST
Health Information Trust Alliance
Align security programs to healthcare-specific control and risk management practices trusted by providers and ecosystem partners.
06
HITECH
Health Information Technology for Economic and Clinical Health Act
Design with breach notification readiness, digital record safeguards, and operational controls that support regulated care programs.
07
SaMD
FDA Software as a Medical Device
Plan software quality, traceability, and documentation pathways for products that may require SaMD review and submission.
08
MDR (EU)
Medical Device Regulation (European Union)
Prepare EU market-ready processes for risk classification, evidence tracking, and lifecycle governance under MDR expectations.
09
SAMHSA
Substance Abuse and Mental Health Services Administration
Apply confidentiality controls and consent-aware sharing models for behavioral and mental health data experiences.
Standards
What it is built against
HIPAASOC 2FHIRHL7LOINC
How It Works
From fax to structured record
Four stages: capture the document, extract under a confidence guardrail, review the uncertain, and write back to the chart.
1.
Capture the document
Inbound faxes, PDFs, and scans land in the pipeline, where OCR converts every page to clean, searchable text.
OCR capture
2.
Classify + extract
The document type is identified, a template extracts structured fields, and codes are normalized, each value scored for confidence.
Extraction + guardrails
3.
Match + review
The document is matched to a patient and encounter; low-confidence fields and ambiguous matches go to a reviewer to confirm or correct.
Human review
4.
Write back to the chart
Approved data posts to the EHR as FHIR R4 resources with the source image attached, and the whole path is logged for audit.
EHR write-back
Related proof
Read Case StudyLera Health: compliant women's health platform
Related delivery proof from our compliant, AI-assisted healthcare work, not this exact agent.
Compliance & Guardrails
15-Minute Scoping CallGuardrails on every document
In healthcare, the guardrails matter as much as the capability.
PHI handling
Documents are processed under a BAA on HIPAA-ready infrastructure, encrypted at rest and in transit, with access scoped to the intake team.
Human-in-the-loop
Confidence thresholds send uncertain extractions and unmatched patients to a reviewer, so nothing questionable posts unchecked.
Explainability
Every extracted field links back to its exact location on the source page, so a reviewer can verify each value.
Audit logging
Ingestion, extraction, review decisions, and write-backs are logged against the document and patient.
Frequently Asked Questions
Yes. Documents are processed under a BAA on HIPAA-ready infrastructure, encrypted in transit and at rest, with access scoped to the intake team. Every ingestion, extraction, and review decision is logged against the document and patient record.
No. It does the mechanical reading, classifying, and keying, and routes anything low-confidence to staff for a quick check. Your team stops retyping every fax and reviews only the exceptions it flags.
It isn't conversational. It's a document pipeline (OCR, classification, extraction, patient matching, and FHIR write-back) that runs unattended on your fax and PDF queue with a human review gate for uncertain fields.
A single high-volume document type, often referrals or results, can be live in a few weeks once fax and EHR access are in place. We validate accuracy on a labeled sample before enabling write-back, then add types incrementally.
Clear the fax queue
Tell us your workflow and systems, and we'll return an architecture review and a phased rollout plan.
contact@agnotic.com
Partnerships
contact@agnotic.com