Amazon Bedrock and HealthLake automate health insurance claims processing

AWS has published detailed guidance on building an automated AI pipeline for processing medical insurance claims using Amazon Bedrock and AWS HealthLake.

Two Tools — One Process

Bedrock Data Automation is responsible for intelligent data extraction from insurance claim forms: the system analyzes unstructured documents — scans, PDFs, handwritten forms — and transforms them into machine-readable structured fields.

Amazon Bedrock AgentCore hosts and runs an AI agent that receives extracted data, validates it against specified rules, and transforms it into FHIR resources (Fast Healthcare Interoperable Resources) — an international standard for medical data exchange. Ready resources are written directly to AWS HealthLake, a fully managed cloud repository for medical data.

Two services work sequentially as a single pipeline: Data Automation handles document recognition, AgentCore — validation logic and transformation.

How the Process Works

The pipeline goes through several sequential stages that together replace manual processing:

• Form upload — an insurance claim arrives in the system (CMS-1500, UB-04 or other standard formats)
• Data extraction — Bedrock Data Automation parses the document and extracts fields: patient data, ICD-10 diagnosis codes, CPT procedure codes, payment amounts, dates of service
• Validation — AgentCore launches an AI agent with a set of business rules: checks code correctness, date correspondence, presence of mandatory fields
• Transformation — the agent converts validated data into FHIR-compatible resources: Patient, Claim, Coverage, Organization
• Writing to HealthLake — ready resources are saved to storage, where they are available for analytics, audit and further routing

At each stage, built-in automatic checks intercept errors before writing. Manual intervention remains only for non-standard cases.

Why FHIR in Insurance

FHIR is an open HL7 standard adopted by hospitals, insurers and regulators worldwide. In the USA it has been mandatory in the CMS (Centers for Medicare & Medicaid Services) system since 2021. Converting data to FHIR at the pipeline entrance eliminates repeated conversions when transmitting information to government registries, hospital systems or BI platforms.

"You'll learn how to combine these services into an end-to-end workflow that reduces manual processing while maintaining accuracy through automatic checks," — AWS ML

Blog.

AWS HealthLake supports FHIR R4, built-in resource search, analytics via Amazon Comprehend Medical and export to S3. For insurers with large claim volumes, this is a standardized output format without developing their own converters.

What This Means

AWS is consistently building a vertically specialized stack of AI tools for healthcare. The Bedrock Data Automation + AgentCore + HealthLake bundle transforms manual processing of insurance forms — one of the most labor-intensive processes in medical insurance — into a managed automatic pipeline on fully managed services. Teams without MLOps experience don't need to train their own document recognition models or build FHIR integrations from scratch.