Normalized claim
Time: 40 seconds
Average ADR letter processing time is approximately 40 seconds.
MDaudit SmartScan.ai automates payer external audit document processing with Amazon Textract and Amazon Comprehend
Use case typeIntelligent document processingUpdated Jun 13, 2026
MDaudit provides a cloud-based billing compliance and revenue integrity SaaS platform for more than 70,000 healthcare providers and 1,500 healthcare facilities. To address growing post-payment payer external audit requests, MDaudit built SmartScan.ai to automate the extraction and formatting of data from unstructured ADR PDFs that customers upload through a web portal. The solution uses Amazon S3 for document ingestion, Amazon Textract for OCR, Amazon Comprehend for entity extraction, and Amazon Augmented AI for human-in-the-loop review of unrecognized formats.
- Organization
- MDaudit
- Industry
- Healthcare
- Location
- United States
- Published
- September 2023
Reported outcomes
Time: Approximately 40 seconds
Time & speed
Why do we believe this?Outcome claims, sources, and evidence checks
Normalized claim
Accuracy: 95%
Accuracy reaches about 95%+.
Normalized claim
Revenue: 95%
Customers retained approximately 95%+ in revenue.
Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks
- Customer
- MDaudit
- Provider
- AWS
- Maturity
- Unknown
- Linked source
- AWS Machine Learning Blog
No explicit deployment-stage evidence found.
Customer identity supportedSource describes one deploymentMaturity evidence evaluated
Primary read
Use case focus
Showing 3 of 3
- 1Intelligent document processing
- 2Workflow automation
- 3Human-in-the-loop AI
- SmartScan.ai ingests ADR documents into Amazon S3.
- Amazon Textract converts scanned PDFs into machine-readable text.
- Amazon Comprehend identifies key entities such as name, date of birth, and date of service.
- Outputs are compared against preexisting data-object configurations, and unrecognized formats are routed to Amazon Augmented AI for human review.
- Approved configurations are stored for future scans, with client-level document isolation.
- Average ADR letter processing time is approximately 40 seconds.
- In the 12 months after launch, customers responded to about $100M of payer audit requests.
- The company estimated tenfold productivity gains for letter management.
Architecture
Customers upload ADR documents through a web portal into Amazon S3. A function initiates processing. Amazon Textract performs OCR on scanned PDFs, Amazon Comprehend extracts key entities, and both outputs are compared with stored data-object configurations. Unknown formats are sent to Amazon Augmented AI for human review. Approved configurations are stored for reuse. AWS services are used for monitoring metrics, events, and logs.
Sources & evidence1
Evidence: Low35/100Evidence strength
- Customer explicitly identified
- Quantified outcome available
- Technical implementation details available
Type: Blog PostPublished: Sep 27, 2023Publisher: AWSEvidence: VendorConfidence: Medium
AI-generated summary. Verify important details with the linked sources before relying on this case.
Explore related AI use cases
Customer evidence
Provider evidence
Was this useful?
Community
Comments
No published comments yet.