Home icon

Rede Mater Dei de Saúde: Monitoring AI agents in the revenue cycle with Amazon Bedrock AgentCore

Machine Learning Blog



This article describes how Rede Mater Dei de Saúde, a major Brazilian healthcare network, deployed 12 AI agents using Amazon Bedrock AgentCore to address claim denials and revenue cycle inefficiencies.

  • Brazilian hospitals face 15.89% claim denial rates, costing sector ~R$10 billion annually
  • Rede Mater Dei implemented suite of 12 AI agents covering entire revenue cycle
  • Key agents: Contracts, Parameterization, and Authorization agents automate critical processes
  • Three-layer architecture: Data Execution, Agent Execution, and Trust/Compliance layers
  • AgentCore Evaluations added monitoring and continuous improvement capabilities
  • Achieved 517% ROI in first four months, 66% authorization time reduction
  • 33% reduction in surgery start times through improved operational efficiency
  • Built immutable audit trails for regulatory compliance and traceability
  • Real-time KPI visibility enables faster executive decision-making

The project demonstrates how AI agents with proper governance and observability can transform healthcare revenue cycles, delivering measurable financial and operational gains while maintaining compliance and auditability.



Go to article

The AWS News Feed is currently looking for gold sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.

Related articles

Mar 31
2026
Build reliable AI agents with Amazon Bedrock AgentCore Evaluations
May 21
2026
Build AI agents for business intelligence with Amazon Bedrock AgentCore
May 27
2026
Building AI agents for business support using Amazon Bedrock AgentCore
May 27
2026
Powering agentic AI sales strategy with Amazon Bedrock AgentCore

The AWS News Feed is currently looking for silver sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.