How Pariveda built real-time clinical voice notes for Henry Schein One using Amazon Nova and Amazon Bedrock AgentCore

This article describes how Pariveda and AWS built a real-time voice documentation system for Henry Schein One's dental practice management platforms using Amazon Nova and Amazon Bedrock AgentCore.

• Reduces clinical documentation time by 65-70% with real-time note generation during patient visits
• Integrates directly into existing Ascend and Dentrix practice management systems via micro frontend architecture
• Uses Amazon Transcribe for streaming speech-to-text and Amazon Nova 2 Lite for structured clinical note extraction
• Applies Amazon Bedrock Guardrails for content filtering and protected health information redaction
• Processes 1.35 million exams monthly with 300-400 model invocations per minute at production scale
• Implements chunk-level processing to maintain predictable latency regardless of appointment length
• Enforces security through AWS KMS encryption, IAM access control, and full audit trails via CloudWatch and CloudTrail

The solution demonstrates that production-scale real-time clinical AI can be embedded within existing healthcare systems through careful architectural design prioritizing latency, structured output, and operational governance.