Technical Documentation for AI-Enabled Medical Devices: Integrating MDR and EU AI Act Requirements

Artificial intelligence is increasingly used in medical devices and medical device software, from diagnostic support and image analysis to patient monitoring, prediction, triage, therapy support and clinical decision support. These technologies offer significant opportunities, but they also introduce specific challenges related to safety, performance, clinical validity, data quality, bias, transparency, robustness, human oversight, lifecycle control and post-market monitoring.

For AI-enabled medical devices, technical documentation must demonstrate not only conformity with Regulation (EU) 2017/745, but also a structured and risk-based approach to AI-specific requirements. Where the EU AI Act applies, manufacturers need to understand how AI Act expectations can be integrated into existing MDR documentation, quality management, risk management, clinical evaluation, software lifecycle and post-market surveillance processes.

This training provides medical device professionals with a practical understanding of how to build and assess technical documentation for AI-enabled medical devices. It focuses on the integration of MDR expectations with relevant EU AI Act requirements and shows how AI-specific evidence can be structured in a way that supports conformity assessment, regulatory submissions, audits and Notified Body reviews.

Key Topics Covered:

• Regulatory landscape for AI-enabled medical devices under the MDR and, where applicable, the EU AI Act

• Classification considerations under MDR Rule 11 and EU AI Act high-risk classification logic

• Integration of AI Act requirements into MDR technical documentation structures

• Intended purpose, clinical claims, AI functionality and boundaries of use

• AI-specific risk management, including bias, data representativeness, performance degradation, automation bias and human oversight

• Data governance expectations, including training, validation, test data, representativeness, data quality and data traceability

• Model development and validation evidence, including performance metrics, clinical relevance, robustness and generalizability

• Software lifecycle documentation for AI-enabled medical device software, including links to IEC 62304

• Transparency, user information, instructions for use, limitations and human-machine interaction

• Cybersecurity, robustness, logging, monitoring, and post-market performance follow-up

• Change control for AI systems, including model updates and lifecycle management

• Common gaps and pitfalls in AI-related technical documentation and conformity assessment

• Case studies, practical examples and Q&A sessions

Benefit:

• Understand how MDR and EU AI Act expectations interact for AI-enabled medical devices
• Learn how to structure technical documentation for AI systems embedded in medical devices and medical device software
• Identify AI-specific evidence needs for risk management, clinical evaluation, software lifecycle, data governance and post-market surveillance
• Understand how to document model development, validation, performance monitoring and change control in a regulatory context
• Recognize common gaps identified during technical documentation reviews and Notified Body assessments
• Strengthen the ability to prepare, review, or assess AI-related technical documentation for conformity assessment
• Build a practical bridge between AI innovation, MDR compliance and emerging EU AI Act requirements.

Certificate: Participants receive a Certificate of Attendance of DQS MED GmbH

Target Audience: Usability, human factors, design, engineering and risk management professionals in medical device development, Regulatory Affairs, Quality Management, clinical, software, audit and technical documentation teams, professionals involved in usability engineering documentation, validation, audits and submissions. Applicable to hardware devices, medical device software, combination products and various user environments. Suitable for beginners and experienced professionals applying IEC 62366-1 and integrating usability with risk management. Basic knowledge of medical device development, risk management or QMS is beneficial.