Prior authorization forms arrive by fax, as scanned PDFs, or as electronic health record (EHR) printouts. Staff manually extracts Current Procedural Terminology (CPT) codes, diagnosis codes, National Provider Identifiers (NPIs), and coverage details before reentering them into payer portals. Nutrient Data Extraction API structures those fields — with confidence scores that flag incomplete submissions before they reach the payer.
Trusted by enterprises, governments, and teams building document workflows at scale
WALKTHROUGH
USE CASES
Structure provider, payer, procedure, diagnosis codes, service dates, and supporting clinical evidence from authorization packets.
Extract supporting clinical evidence, denial details, appeal dates, and case references from reconsideration packets and medical necessity letters.
Structure referring provider; patient demographics; reason for referral; and specialty, urgency, and insurance information.
Handle handwritten forms, low-quality faxes, and image-based records with intelligent character recognition and confidence scoring.
Organize multidocument packets containing clinical notes, payer forms, test results, and supporting records into per-document structured output.
GOVERNED EXTRACTION
LLMs can reason over a medical packet, but healthcare workflows need deterministic, traceable output grounded in the source document — not generated answers that vary between runs.
Typed output remains tied to the source document — not generated text that changes from one run to the next.
Flag uncertain values before they move downstream into authorization or payer systems.
Every extracted value is anchored to its source location for traceability and staff review.
Tables, forms, checkboxes, and key-value regions preserved — not flattened into plain text.
Support validation steps before structured data enters authorization, records fulfillment, or administrative systems.
Source context and page detail support audit trails in regulated healthcare environments.
WHAT YOU CAN EXTRACT
Name, date of birth, member ID, and contact details from intake forms, ROI requests, and authorization packets.
Ordering provider, facility, NPI, and referring provider details from prior authorization and referral documents.
Payer name, plan ID, group number, and coverage details from authorization forms and administrative records.
CPT, Healthcare Common Procedure Coding System (HCPCS), and International Classification of Diseases (ICD) codes extracted with confidence scores for validation before downstream use.
Requestor, record categories, date range, authorization dates, delivery instructions, and expiration information.
Checkbox states, handwritten values, and signature presence from faxed or scanned healthcare forms.
HOW IT WORKS
Parse
Turn medical records, faxed forms, and scanned healthcare packets into document structure.
Extract
Identify patient details, codes, payer information, signatures, checkboxes, and form fields.
Map
Define the fields the workflow needs using the schema generator in Studio, or write a JSON Schema directly. Get back validated, structured output.
Structure
Return typed JSON for administrative systems and validation, or Markdown for search and AI workflows.
Process
Send structured data into prior authorization, utilization management, referral, or revenue-cycle platforms.
OUTPUT FORMATS
Confidence scores, coordinates, and page context included.
Choose output: "json" or Markdown per request.
Spatial JSON
For extraction · validation · review
{ "type": "keyValueRegion", "id": "kv_h1i2j3", "bounds": { "x": 82, "y": 128, "width": 340, "height": 96 }, "confidence": 0.97, "page": { "pageIndex": 0, "pageNumber": 1 }, "pairs": [ { "key": { "entityType": "QUESTION", "value": "Patient name", "bounds": { "x": 82, "y": 128, "width": 110, "height": 24 }, "confidence": 0.99 }, "value": { "entityType": "ANSWER", "value": "James R. Morrison", "bounds": { "x": 200, "y": 128, "width": 150, "height": 24 }, "confidence": 0.99 }, "relationshipConfidence": 0.97 }, { "key": { "entityType": "QUESTION", "value": "Requested service", "bounds": { "x": 82, "y": 164, "width": 130, "height": 24 }, "confidence": 0.97 }, "value": { "entityType": "ANSWER", "value": "Lumbar MRI", "bounds": { "x": 220, "y": 164, "width": 110, "height": 24 }, "confidence": 0.97 }, "relationshipConfidence": 0.95 }, { "key": { "entityType": "QUESTION", "value": "CPT code", "bounds": { "x": 82, "y": 200, "width": 90, "height": 24 }, "confidence": 0.98 }, "value": { "entityType": "ANSWER", "value": "72148", "bounds": { "x": 180, "y": 200, "width": 70, "height": 24 }, "confidence": 0.98 }, "relationshipConfidence": 0.96 } ]}Markdown
For RAG · search · knowledge bases
# Prior Authorization Request
**Patient** James R. Morrison**Date of birth** 1971-08-22**Plan ID** BCBS-9934021
## Requested Service
| Field | Value || --- | --- || Procedure | Lumbar MRI || CPT code | 72148 || ICD-10 | M54.5 |
**Ordering provider** Dr. Susan Okafor**Request date** 2024-10-14Regional processing options
Choose supported processing regions for enterprise deployments.
SOC 2 Type 2 audited
Backed by Nutrient’s SOC 2 Type 2 security practices, built for use in compliance-sensitive workflows.
Trust and compliance
TLS encryption by default
All API communication is TLS-encrypted. Documents in transit are protected end to end.
HIPAA-compatible architecture
Encrypted transport and access controls designed to support HIPAA-sensitive workflows.
Yes. For hand-completed fields, checkboxes, and handwritten values, the API applies intelligent character recognition to capture what standard OCR misses. Every extracted value includes a confidence score so uncertain fields can be flagged for staff review before moving downstream.
Data Extraction API processes PDFs; images (including scans and photos); and Word, Excel, and PowerPoint files. It handles scanned records, faxed forms, and mixed digital/image-based documents without requiring a separate OCR pipeline.
Nutrient Data Extraction API is an extraction layer, not a prior authorization system or utilization management platform. It extracts structured data from prior authorization forms and clinical packets so that data can be reviewed, validated, and routed into the systems organizations already use.
Every extracted element includes confidence scores, page references, and coordinates so teams can flag low-confidence fields for review; trace values back to the source document; and validate results before they enter authorization, records fulfillment, or administrative systems.
Use the schema generator in Studio — upload example prior authorization forms, ROI requests, or referral documents and describe the document type. It generates a JSON Schema for the /extract endpoint. Fields like patient identifiers, codes, payer details, and authorization information can also be defined manually. Refer to the documentation for supported field types and limits.
Use agentic mode for healthcare documents that require the deepest visual understanding — degraded scans, faxed records, handwritten clinical notes, and documents with embedded images where understand mode produces insufficient results. Agentic mode adds a vision language model (VLM) on top of the understand pipeline. It costs 18 credits per page.