Skip to main content
PDF4me APIIntegrationsKnowledge BaseExtensionsAPI TesterBlog

AI-Process Health Card in Make

What this module does

PDF4me – AI-Process Health Card reads a health insurance card image or PDF and returns structured insurance data as individually mappable fields — member ID, group number, plan type, insurance provider, member name, date of birth, coverage dates, prescription BIN, PCN, group, and copay information per service type. There is no template to create and no field positions to configure. The AI recognises card layouts from any insurer automatically, making it practical for patient intake portals, insurance verification workflows, and benefits pre-authorization scenarios in Make where cards arrive from many different providers.

Authenticating Your API Request​

Every PDF4me module in Make requires a valid Connection. Create or select one that holds your PDF4me API key so the scenario can call the AI health card service securely.

Get your API Key

Important Facts You Should Not Miss​

No template setup required — AI handles any card layout
Unlike OCR solutions that require field coordinates or per-insurer templates, this module uses a trained AI model that recognises card layouts automatically. You supply the card image or PDF and the AI locates the member ID, group number, prescription fields, and copay details without any configuration — supporting hundreds of insurer card formats out of the box.
Supports PDF, PNG, JPG, and JPEG inputs
Pass binary content from any source — a patient portal file upload, an email attachment, a Dropbox download, or a Google Drive file — directly to the Document field. There is no need to convert images to PDF or perform any pre-processing step. Map File Name with the correct extension (.pdf, .png, .jpg, or .jpeg) so the module detects the format correctly.
Each output field is a separate mappable token
The module returns all extracted fields as individual tokens in the output bundle — memberId, groupNumber, planType, insuranceProvider, memberName, dateOfBirthStr, effectiveDateStr, rxBin, rxPcn, rxGroup, and copayInfo. No JSON parsing, no custom functions. Click any token in Make's mapping panel to wire it directly into an Airtable field, a Router condition, an email body, or an EHR API call.
Make PDF4me AI-Process Health Card module showing Connection set to My PDF4me connection, File set to Map with File Name mapped from 1. File Name and Document mapped from 1. Data

Set File to Map, then wire File Name and Document from the step that downloaded or received the health card file.

Parameters​

Required: Connection, File Name, and Document. Set File to Map first — this reveals the File Name and Document fields in the module panel.

ParameterRequiredWhat it doesExample mapping
ConnectionYesPDF4me API connection that authenticates the AI health card request. Click Add to create one by pasting your PDF4me API key — reused automatically across all PDF4me modules in your scenarios.My PDF4me connection
FileNoHow the health card file is supplied to the module. Choose Map when passing binary content from a prior download step. Choose Dropbox – Download a File to pull directly from a Dropbox trigger without a separate download step.Map
File NameYesFilename of the health card including extension (.pdf, .png, .jpg, or .jpeg). Used by the AI to detect the input format. Map from the filename output of your source module — e.g. the File Name field of a Dropbox or Google Drive download step.1. File Name
DocumentYesBinary content of the health card file. Must be raw file bytes from a download or upload module — the Data field in Dropbox, Google Drive, or OneDrive. Do not pass a URL string or filename here; the module requires actual binary to run AI processing.1. Data

Output Fields​

FieldTypeWhat it contains
memberIdStringUnique member identification number printed on the card. Map into patient records or use in a Router to look up existing members.
groupNumberStringGroup or policy number. Used in insurance verification API calls and claims submissions.
planTypeStringType of health plan — PPO, HMO, EPO, Medicare, Medicaid, etc. Use in a Router to branch scenario logic per plan type.
insuranceProviderStringName of the insurance company (e.g. Blue Cross Blue Shield, Aetna, UnitedHealth). Map into EHR or billing records.
memberNameStringFull name of the insured member as printed on the card.
dateOfBirthStrStringMember date of birth in ISO 8601 format. Use to cross-verify against registration data.
effectiveDateStrStringDate when coverage became or becomes effective. Use in a Filter to reject expired coverage before scheduling.
rxBinStringPrescription BIN number for drug coverage routing through pharmacy networks.
rxPcnStringPrescription PCN (Processor Control Number) for pharmacy claims.
rxGroupStringPrescription group identifier for pharmacy benefit management.
copayInfoArrayArray of copay entries — each with a service type (e.g. Primary Care, Specialist, Emergency) and copay amount. Map into a Slack notification or patient-facing email.

Quick Setup​

  1. Add PDF4me → AI-Process Health Card to your Make scenario.
  2. Select Connection (or click Add to create one with your PDF4me API key).
  3. Under File, choose Map to reveal the File Name and Document fields.
  4. Map File Name to the filename output of your source step — include the extension (.pdf, .png, .jpg, or .jpeg).
  5. Map Document to the binary data field of the same step — typically named Data in Dropbox, Google Drive, or OneDrive modules.
  6. Save and click Run once. Expand the output bundle — each extracted field is a separate mappable token ready to wire into EHR APIs, databases, or notification modules.

Workflow Examples​

Workflow ExamplesCommon Make scenario patterns using AI-Process Health Card.
Patient portal upload → AI extract → create EHR record
  1. A new patient uploads their health card image via a Typeform or JotForm intake portal.
  2. AI-Process Health Card extracts memberId, planType, insuranceProvider, and copayInfo from the uploaded image.
  3. A Router branches by planType — PPO patients route to one EHR workflow, HMO to another.
  4. An HTTP module creates a new patient record in the EHR system with all extracted insurance fields pre-populated.
  5. Gmail sends a confirmation to the patient with their copay amounts for common services.
Email attachment → AI parse → benefits verification → appointment prep
  1. Gmail triggers when a patient emails their health card image as an attachment before an appointment.
  2. AI-Process Health Card extracts memberId, groupNumber, effectiveDateStr, and rxBin.
  3. A Filter blocks the scenario if effectiveDateStr is in the past — an automated email asks the patient to provide updated coverage.
  4. An HTTP module calls the insurer's eligibility API with memberId and groupNumber to confirm active benefits.
  5. A Slack message posts the copay amounts and plan type to the front-desk channel before the patient arrives.
Annual re-verification batch — Dropbox folder → re-extract → update records
  1. A scheduled scenario lists all health card images in a Dropbox "annual-reverification" folder.
  2. An iterator loops over each file and passes it to AI-Process Health Card.
  3. The module re-extracts insuranceProvider, planType, effectiveDateStr, and groupNumber for each card.
  4. An Airtable or Google Sheets module updates the patient record with the refreshed insurance data.
  5. A second Filter flags records where the insurer or group number changed since last year for manual review.

Frequently Asked Questions​

What types of health cards can this module process?+
The module processes standard health insurance cards from US and international insurers including PPO, HMO, EPO, and Medicare/Medicaid plans. It reads both physical card scans and digital card images sent via patient portal upload, email attachment, or mobile photo. The AI is trained on diverse card layouts across hundreds of insurers, so it does not require a fixed template or field-coordinate configuration for each provider — you supply the card and the AI extracts the data.
Does the module support image files or only PDFs?+
Both formats are fully supported. You can pass .pdf, .png, .jpg, and .jpeg files directly to the Document input. For smartphone photos of physical cards, .jpeg or .png work without any conversion step. For scanned card documents delivered as multi-page PDFs, the module reads the card from the PDF pages. Set File Name with the correct extension so the AI detects the input format accurately — passing a .png with a .pdf extension may cause a format mismatch error.
What data fields does the AI extract from a health card?+
The module returns memberId, groupNumber, planType, insuranceProvider, memberName, dateOfBirthStr, effectiveDateStr, rxBin, rxPcn, rxGroup, and copayInfo (an array listing copay amounts for service types such as Primary Care, Specialist, and Emergency). Each field is an individual mappable token in Make — no JSON parsing or custom functions are needed. Fields that are absent from a card or not readable will return empty strings rather than causing an error.
How accurate is the AI extraction, and how do I handle low-confidence results?+
For clear, well-lit card photos or high-resolution scans at 200 DPI or above, accuracy on core fields like memberId and insuranceProvider typically exceeds 90%. Blurry photos, low-contrast cards, or non-standard layouts produce lower confidence. Add a Make Filter after the module to check the success field — route failed extractions or empty memberId results to a human review queue in Airtable or a Slack alert rather than passing them directly to billing or EHR systems.
How do I use the extracted data in the next Make module?+
After the module runs, expand its output bundle in the Make mapping panel. Each extracted field appears as an individual selectable token — click memberId to insert it into an EHR record field, planType to set a Router branch condition, effectiveDateStr to wire into a Filter that rejects expired coverage, or copayInfo to populate a patient-facing email template. The full output is also available as a single JSON object if you need to store the entire record in a database in one step.

Get Help​

Make IntegrationGet your API KeyMake Documentation