Powered by AWS

Intelligent PHI Redaction at Scale

Automatically detect and redact personally identifiable information from documents using advanced AI. Built for healthcare and enterprise compliance workflows.

See Redaction in Action

Watch how the AI identifies and redacts sensitive information while preserving document context.

Original Document

Patient Information Name: John Michael Smith DOB: March 15, 1985 SSN: 482-55-7891 Phone: (555) 234-8901 Email: jsmith@email.com MRN: MR-2024-881456 Visit Notes Patient presents with recurring headaches for the past 2 weeks. Reports stress from work at Acme Corporation as contributing factor. Assessment: Tension-type headache. Follow-up scheduled for April 2, 2024. Physician: Dr. Sarah Johnson License: MD-45892

AI Processing
Redacted Output

Patient Information Name: [PATIENT_NAME] DOB: [DATE_OF_BIRTH] SSN: [SSN] Phone: [PHONE_NUMBER] Email: [EMAIL_ADDRESS] MRN: [MEDICAL_RECORD_NUMBER] Visit Notes Patient presents with recurring headaches for the past 2 weeks. Reports stress from work at [ORGANIZATION] as contributing factor. Assessment: Tension-type headache. Follow-up scheduled for [DATE]. Physician: [PHYSICIAN_NAME] License: [LICENSE_NUMBER]

HIPAA 18 Protected Health Information Identifiers:
1. Names 2. Geographic Data 3. Dates 4. Phone Numbers 5. Fax Numbers 6. Email Addresses 7. SSN 8. Medical Record Numbers 9. Health Plan Beneficiary Numbers 10. Account Numbers 11. Certificate/License Numbers 12. Vehicle Identifiers 13. Device Identifiers 14. Web URLs 15. IP Addresses 16. Biometric Identifiers 17. Full-face Photos 18. Other Unique Identifiers

Built for Enterprise

Every feature designed with security, scalability, and compliance in mind.

Claude-Powered Detection

Leverages Claude Sonnet via Amazon Bedrock for context-aware PHI detection that understands document semantics.

Async Processing

Process multiple documents concurrently with automatic scaling. SQS-based queue ensures reliable delivery.

Human-in-the-Loop

Built-in review dashboard with diff viewer. Edit redactions inline and approve individual notes or entire batches.

Enterprise Security

Cognito authentication with secure access controls. All data stays within your AWS account.

Infrastructure as Code

Complete AWS CDK stack for one-command deployment. Version controlled, reproducible, and customizable.

HIPAA 18 Detection

Detects all 18 HIPAA identifiers: names, SSNs, dates, geographic data, contact info, medical records, device IDs, biometrics, and more.

Full Demo Video

See the complete workflow from upload to approval.

Architecture

Infrastructure defined as code with AWS CDK. One-command deployment to your AWS account.

Architecture diagram for the PHI de-identification platform
Serverless Event-Driven Cloud-Native Secure Authentication

Technology Stack

Serverless architecture on AWS with automatic scaling, fault tolerance, and pay-per-use pricing.

AWS Lambda Serverless Compute
Amazon S3 Document Storage
Amazon SQS Message Queue
Amazon Bedrock Claude AI
Amazon Cognito Authentication
API Gateway REST API
AWS CDK Infrastructure as Code
React Frontend UI
TypeScript Type Safety
Python Backend Logic
DynamoDB Batch State
AWS Amplify Frontend Hosting
CloudWatch Metrics & Logs

Ready to protect sensitive data?

Deploy the full stack to your AWS account with a single command.
Open source and fully customizable.