Extract data from superbills

A superbill is the itemized encounter receipt a provider hands a patient so the patient, not the provider, can seek reimbursement from an insurer. It is common in out-of-network and cash-pay care: a psychotherapist, a physical therapist, or an out-of-network specialist gives the patient a superbill after the visit, and the patient submits it to their plan. Unlike a professional claim on a CMS-1500 or an insurer's Explanation of Benefits (EOB), a superbill has not been adjudicated. It states what was done and what the patient paid, and it carries the exact coded data a payer needs: the rendering provider with a National Provider Identifier (NPI) and tax ID, the diagnoses in ICD-10-CM, and each service in CPT or HCPCS. Itemized service lines are the working part. Each line pairs a CPT code with a description, the units, the unit price, the date of service, and the charge, and the charges have to foot to the total. A single visit can carry an office evaluation, an immunization administration, and a blood count, each on its own line and each pointing to a diagnosis. Its header ties the encounter to the patient and the payer: patient name and member ID, payer name and plan, provider NPI and tax ID, place of service, and any prior authorization number. Because the patient submits it, the patient responsibility usually equals the amount paid out of pocket, while the allowed and paid columns stay blank until the plan processes the request. Talonic reads a superbill into a claim header plus a service-line array. One superbill dated 2026-06-15 from provider Dr. Helen Cho, NPI 1487654321, for patient Rosa Delgado returns three CPT lines billed at 220.00, 45.00, and 35.00 USD that foot to a total charge of 300.00 USD, each line with its diagnosis pointer and date of service, so the patient or a reimbursement service files from structured data. Being a patient's own record, the document is protected health information (PHI), and Talonic returns the codes and amounts as written, making no coverage determination and no claim decision.

What gets extracted from superbills

How extraction works for superbills

Superbills are produced by EHR and practice-management systems and by templated encounter forms, so the service-line grid and the placement of the codes vary by practice. Talonic classifies the document and maps it to the healthcare claims schema in the Field Registry, which separates the claim header from the per-service-line detail and keeps the ICD-10-CM diagnosis codes distinct from the CPT and HCPCS procedure codes. Each service line keeps its units, unit price, date of service, charge, and diagnosis pointer, and the line charges are checked to foot to the total claim amount, so a 220.00 plus 45.00 plus 35.00 USD encounter is verified to sum to 300.00 rather than trusted. Allowed and paid columns, present in the schema for adjudicated documents, stay empty on a patient-submitted superbill. Every value returns with a confidence score and pixel-region provenance under DIN SPEC 91491 conformity, and because the document is protected health information, the fields are released only to the submitting account. Read as written, the codes and amounts carry no coverage or reimbursement decision.

Sample extraction

A three-line out-of-network superbill given to the patient

{
  "document_number": "SB-2026-0615",
  "document_date": "2026-06-15",
  "claim_status": "active",
  "claim_type": "professional",
  "provider.name": "Dr. Helen Cho",
  "provider.npi": "1487654321",
  "provider.tax_id": "84-2210987",
  "patient.name": "Rosa Delgado",
  "patient.date_of_birth": "1990-07-24",
  "patient.member_id": "M40028117",
  "payer.name": "Cigna",
  "payer.plan_name": "Cigna PPO",
  "service_dates": "2026-06-15",
  "place_of_service": "office",
  "diagnosis_codes": "F41.1, Z00.00",
  "procedure_codes": "99214, 90471, 85025",
  "billed_amount": 300,
  "total_claim_amount": 300,
  "patient_responsibility": 300,
  "adjudication_outcome": "not yet adjudicated",
  "service_lines": [
    {
      "line_number": "1",
      "procedure_code": "99214",
      "procedure_description": "Office/outpatient visit, established patient",
      "units": 1,
      "unit_price": 220,
      "serviced_date": "2026-06-15",
      "charge_amount": 220,
      "patient_responsibility": 220,
      "diagnosis_pointer": "F41.1"
    },
    {
      "line_number": "2",
      "procedure_code": "90471",
      "procedure_description": "Immunization administration",
      "units": 1,
      "unit_price": 45,
      "serviced_date": "2026-06-15",
      "charge_amount": 45,
      "patient_responsibility": 45,
      "diagnosis_pointer": "Z00.00"
    },
    {
      "line_number": "3",
      "procedure_code": "85025",
      "procedure_description": "Complete blood count with differential",
      "units": 1,
      "unit_price": 35,
      "serviced_date": "2026-06-15",
      "charge_amount": 35,
      "patient_responsibility": 35,
      "diagnosis_pointer": "Z00.00"
    }
  ],
  "diagnoses": [
    {
      "diagnosis_sequence": "1",
      "diagnosis_code": "F41.1",
      "diagnosis_description": "Generalized anxiety disorder",
      "is_principal": true
    },
    {
      "diagnosis_sequence": "2",
      "diagnosis_code": "Z00.00",
      "diagnosis_description": "Encounter for general adult medical exam without abnormal findings",
      "is_principal": false
    }
  ]
}

Frequently asked