ProductionEvidence: Low40/100

Automated Claims Adjudication Workflow Using Amazon Textract and Amazon Comprehend Medical

A generic healthcare payer institution implemented an automated claims adjudication workflow using AWS technologies to process medical insurance claims with minimal manual intervention. The workflow processes claims in PNG format, validates their authenticity and correctness, extracts medical procedure details using AI, and provides analytics for clinical and claims data. The solution employs a serverless architecture using Amazon S3, Amazon Textract, AWS Lambda, Amazon QuickSight, Amazon Comprehend Medical, Amazon Athena, and Amazon Simple Notification Service to automate and streamline claims adjudication and analytics.

Industry
Healthcare
Published
November 2019

Reported outcomes

Strategic outcomes

Speed & agilityAutomated claims adjudication workflowNew product / capabilityValidated claims for authenticity and complianceBetter decisions & insightEnabled scalable claims analytics and insightsNew product / capabilityExtracted medical procedure details from claims
Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks
Customer
Generic Healthcare Payer Institutions
Provider
AWS
Maturity
Production

Deployed a serverless, event-driven architecture using Amazon Textract to extract entities and relationships from claim documents stored in Amazon S3

Customer identity supportedSource describes one deploymentMaturity supported

Primary read

Use case focus

Showing 3 of 3

  • 1Claims Adjudication
  • 2Medical Document Processing
  • 3Healthcare Analytics
  • Manual processing of unstructured medical claims data was time-consuming and prone to delays and errors.
  • The need to validate claims accurately for authenticity, correctness, and insurance coverage compliance in a scalable manner.
  • To automate the medical claims adjudication workflow while enabling analytics from the extracted data.
  • Deployed a serverless, event-driven architecture using Amazon Textract to extract entities and relationships from claim documents stored in Amazon S3.
  • Used AWS Lambda functions to perform syntax and domain-level validation of extracted data.
  • Employed Amazon Comprehend Medical to extract medically relevant information such as procedures from claims.
  • Configured Amazon Simple Notification Service to send email notifications for claims failing validation.
  • Integrated Amazon Athena for querying aggregated clinical and claims data, and used Amazon QuickSight for visualization and analytics dashboards.
  • The system automates the medical claims adjudication process, significantly reducing manual intervention.
  • Improves accuracy and speed of claims validation and processing, enhancing operational efficiency.
  • Enables scalable analytics from claims data to drive insights on clinical procedures and claims patterns, assisting better decision-making.
Architecture

The architecture is a serverless, event-based workflow using Amazon S3 for document storage; Amazon Textract for entity extraction; AWS Lambda for validation and workflow logic; Amazon Comprehend Medical for medical data extraction; Amazon SNS for notifications; Amazon Athena for querying; and Amazon QuickSight for visualization.

Sources & evidence1
Evidence: Low40/100Evidence strength
  • Customer explicitly identified
  • Deployment status explicitly supported
  • Technical implementation details available
Type: Blog PostPublished: Nov 19, 2019Publisher: AWS Machine Learning BlogEvidence: VendorConfidence: Medium

AI-generated summary. Verify important details with the linked sources before relying on this case.

Explore related AI use cases

Was this useful?

Community

Comments

Loading comments...

Similar cases