AI-powered clinical trial eligibility and safety using Amazon Bedrock AgentCore

Summary

Manual chart review across fragmented clinical sources slows clinical trial enrollment, contributing to timeline delays and high screen failure rates. To address this bottleneck, an architecture on AWS automates eligibility and safety assessments while preserving human clinician authority. AWS HealthLake ingests and normalizes records into FHIR R4 resources, while Amazon Bedrock AgentCore orchestrates pre-screening, detailed screening, and site enrollment agents. Amazon Bedrock AgentCore Evaluations scores screening determinations using an LLM-as-a-judge approach for clinical accuracy, operational effectiveness, and safety compliance. This human-in-the-loop workflow produces structured recommendations with source citations, reducing patient matching time from days to minutes while maintaining compliance audit trails.

Context

Eighty percent of clinical trials miss enrollment timelines, with delays costing an estimated $500,000 daily. Decisions currently depend on manual chart reviews across fragmented electronic health records, lab portals, imaging reports, and medication histories as trial protocols grow increasingly complex.

Approach / What changed

Architecting an AI-assisted screening pipeline using AWS HealthLake for FHIR-native data normalization, Amazon Bedrock AgentCore to orchestrate specialized screening agents behind guardrails, Amazon Bedrock Knowledge Bases for trial criteria, and Amazon Bedrock AgentCore Evaluations with human-in-the-loop review.

Takeaways

• Amazon Bedrock AgentCore orchestrates three specialized agents for pre-screening, detailed protocol and safety screening, and site enrollment logistics.
• Amazon Bedrock AgentCore Evaluations uses LLM-as-a-judge evaluators to score decisions on clinical accuracy, operational effectiveness, and safety compliance.
• The pipeline reduces patient matching time from days to minutes while capturing workflow histories designed to support FDA 21 CFR Part 11 requirements.

Related reading

• Dropbox · Jun 25, 2026

  How we used DSPy to turn AI evaluations into better responses in Dash chat

  Dropbox improved its Dash chat agent by establishing an automated optimization loop powered by DSPy and LLM-as-judge evaluations. Engineers first calibrated their LLM judges against human-annotated interaction traces, then used those judges to systematically optimize the agent's system prompts via offline counterfactual replay. This automated workflow reduced incomplete responses by 26% while decreasing overall token consumption.

  Ilya Yakovlev,Andrew Cheung,Binoy Dash,Simran Jumani,Dmitriy Meyerzon

• AWS · Aug 26, 2026

  Closing the AI agent trust gap with graduated autonomy

  Operating AI agents under binary full-access or read-only permissions creates an operational trust gap due to unpredictable model behavior across updates. The graduated autonomy architectural pattern closes this gap by dynamically granting or revoking tool permissions through sustained reliability scoring. A scoring engine computes a composite score across five weighted dimensions over a rolling 50-action window, mapping agents across four tiers from probation to full autonomy. In-process pre-execution filters check for prompt injection and sensitive targets, while Amazon Bedrock AgentCore Gateway and Cedar policies strictly enforce deny-by-default access at the infrastructure level. Post-execution state capture in Amazon DynamoDB supports reversibility, and AWS CodePipeline delivery gates block releases failing adversarial evaluation benchmarks.