Agentic Post-Market Surveillance (A-PMS)

This repository contains the implementation of our study titled "Prototyping Agentic Post-Market Surveillance via ISO 13485 Digital Twins".

Submitted to MEDICON 2026 - XVII Mediterranean Conference on Medical and Biological Engineering and Computing - September 14-17, 2026 Siena, Italy

Authors

• Ali Salman - Department of Medical Biotechnologies, University of Siena, 53100 Siena, Italy.
• Ernesto Iadanza - Department of Medical Biotechnologies, University of Siena, 53100 Siena, Italy.

Abstract

This study presents a functional prototype for autonomous post-market surveillance of medical devices, bridging the execution gap between theoretical AI frameworks and live hardware orchestration. Using the OpenClaw multi-agent system powered by Gemma 4 language model and Exa for real-time regulatory intelligence, we developed a proactive technical loop that interfaces a whitelisted WhatsApp control layer with digital twin telemetry. To ensure deterministic device identification and minimize AI hallucinations, we implemented a declarative metadata layer for autonomous agent discovery. The system's reliability was validated through forced-stress simulations where the orchestrator autonomously refactored device logic to trigger and report critical thermal anomalies in compliance with ISO 13485:2016. Results demonstrate 100% discovery accuracy and successful closed-loop alerting, proving that generative AI integrated with deterministic safety gates can effectively automate the link between hardware drift and mandatory vigilance reporting. This architecture provides a scalable, human-in-the-loop framework that maintains strict adherence to the safety mandates of the EU AI Act.

Architecture

The framework is organized into four distinct layers:

1. User Interface: A secure WhatsApp gateway for human-in-the-loop oversight.
2. Agentic Orchestration: A multi-agent system powered by Gemma 4 and OpenClaw.
3. Regulatory Intelligence: Real-time compliance grounding via the Exa API.
4. Physical Simulation: A logic bridge connected to a Digital Twin (MD-01).

Key Features

• Deterministic Grounding: Uses code-level metadata to prevent LLM hallucinations regarding hardware states.
• Autonomous Alerting: Transitions from passive logging to active intervention upon detecting thermal breaches (40.0°C threshold).
• Regulatory Synthesis: Automatically fetches ISO 13485:2016 standards during anomaly detection.

Citation

If you use this work, please cite it as:

@misc{salmanMEDICON2026,
  author       = {Ali Salman and Ernesto Iadanza},
  title        = {Prototyping Agentic Post-Market Surveillance via ISO 13485 Digital Twins},
  year         = {2026},
  institution  = {Department of Medical Biotechnologies - University of Siena, Siena, Italy},
  url          = {https://github.com/alexsalman/MEDICON2026/medicon2026_unisi_it.pdf}
  urldate      = {2026-04-10}
}

⸻
© 2026 Ali Salman · MIT License