Artificial Intelligence (AI) in Medical Engineering
Understand Compliance, Risks, and Best Practices
Artificial Intelligence (AI) is transforming medical technology by enabling advanced diagnostic, monitoring, and decision-support capabilities. As regulatory expectations and industry adoption continue to evolve, medical device manufacturers must understand how AI and machine learning technologies can be safely, effectively, and compliantly integrated into medical products.
This instructor-led course provides a practical introduction to AI and machine learning in medical engineering while examining the regulatory, quality, cybersecurity, data management, verification, validation, and lifecycle requirements governing AI-enabled medical devices. Participants will gain valuable insights into current expectations under MDR, IVDR, the EU AI Act, FDA guidance, and Good Machine Learning Practice (GMLP).
After completing this course, you will be able to:
- Explain key AI and machine learning concepts used in medical engineering.
- Distinguish between various machine learning approaches and their applications.
- Determine when AI software qualifies as a medical device.
- Understand classification requirements for AI-enabled medical devices.
- Interpret key requirements from MDR, IVDR, EU AI Act, FDA guidance, and related standards.
- Apply Good Machine Learning Practice principles across the Total Product Lifecycle (TPLC).
- Identify AI-specific hazards, risks, and CAPA triggers.
- Assess data quality, traceability, transparency, explainability, and trustworthiness requirements.
- Recognize cybersecurity and privacy risks affecting AI-enabled healthcare systems.
- Understand verification and validation approaches for AI systems.
- Implement effective market surveillance and performance monitoring practices.
- Understand Predetermined Change Control Plan (PCCP) concepts for adaptive AI systems.
Important: TÜV SÜD tests and certifies through its Notified Bodies medical devices and their manufacturers. TÜV SÜD Akademie GmbH offers trainings in the field of medical devices.
The neutrality and independence of the conformity assessment procedures carried out by TÜV SÜD must be maintained. In the interest of our customers, hence the seminars of TÜV SÜD Akademie do not contain any product-specific, process-related or company-specific content or solutions that could fulfill the function of individual consulting.
In case of any questions or uncertainties, please do not hesitate to contact us at [email protected].
This course is intended for:
- Medical device manufacturers
- AI and machine learning engineers
- Software development professionals
- Regulatory Affairs professionals
- Quality Assurance and Quality Management personnel
- Risk Management specialists
- Product Managers
- Project Managers
- Systems Engineers
- Requirements Engineers
- Clinical Affairs professionals
- Cybersecurity professionals
- Data scientists involved in healthcare applications
- Auditors and compliance specialists
- Medical software consultants and service providers
- Suppliers supporting AI-enabled medical devices
Module 1: Introduction and AI Fundamentals
- AI terminology and definitions
- Types and classifications of AI systems
- AI applications in healthcare and medical devices
- AI performance measures and evaluation criteria
Module 2: Machine Learning Fundamentals
- Supervised, unsupervised, and reinforcement learning
- Decision trees and regression models
- Neural networks and deep learning
- Additional machine learning methods
- Common AI development pitfalls and model limitations
Module 3: Data Quality, Privacy and Cybersecurity
- Data quality requirements
- Representative datasets and bias considerations
- Data leakage and overfitting
- Privacy by Design principles
- Differential Privacy and Federated Learning
- Cybersecurity threats to AI systems
- Model poisoning, model stealing, membership attacks, and prompt injection
Module 4: Regulatory Requirements for AI Medical Devices
- MDR requirements
- IVDR requirements
- EU AI Act requirements
- FDA expectations and guidance
- AI software classification as medical devices
- International guidance and standards overview
Module 5: Trustworthy and Explainable AI
- Transparency and explainability
- Human-in-the-loop concepts
- Ethical considerations
- Building trustworthy AI systems
Module 6: Risk Management and CAPA
- AI-specific risks
- ISO 14971 considerations
- BS/AAMI 34971:2023 overview
- Risk controls and mitigation strategies
- CAPA processes and AI-specific triggers
Module 7: Good Machine Learning Practice (GMLP)
- GMLP principles
- Total Product Lifecycle approach
- Quality management integration
- Data governance and monitoring
Module 8: Verification, Validation and Lifecycle Management
- AI-specific verification approaches
- Validation methodologies
- Clinical evaluation considerations
- Performance monitoring and post-market surveillance
- Predetermined Change Control Plans (PCCP)
Module 9: Practical Exercises and Case Studies
- Interactive discussions
- Risk assessment exercises
- Regulatory examples
- Real-world implementation scenarios
Course Summary and Q&A
Artificial Intelligence and Machine Learning are rapidly becoming integral components of modern medical devices, supporting clinical decision-making, diagnostics, monitoring, and personalized healthcare. As AI-enabled products continue to expand across the healthcare sector, manufacturers, developers, and quality professionals must understand both the technical foundations of AI and the regulatory frameworks that govern these systems.
This course provides a comprehensive overview of AI in medical engineering, beginning with key concepts such as machine learning, neural networks, supervised and unsupervised learning, performance measurement, and common challenges associated with AI model development. Participants will gain a practical understanding of how AI systems function and how they are applied within the medical device industry.
The training explores the evolving global regulatory landscape for AI-enabled medical devices, including MDR, IVDR, the EU AI Act, FDA guidance, and international industry recommendations. Participants will learn when AI software qualifies as a medical device, how such systems are classified, and what documentation, quality management, and conformity assessment requirements apply throughout the product lifecycle.
Special emphasis is placed on AI-specific risk management, cybersecurity, data quality, privacy protection, transparency, explainability, and trustworthiness. Participants will examine common AI-related risks such as model bias, data leakage, cybersecurity threats, model poisoning, and privacy challenges, while learning practical approaches to mitigation and compliance.
The course also introduces Good Machine Learning Practice (GMLP), AI lifecycle management, verification and validation strategies, post-market monitoring, CAPA processes, and Predetermined Change Control Plans (PCCP). Through practical examples, case studies, and interactive exercises, participants will gain the knowledge necessary to support the development, validation, deployment, and maintenance of safe, effective, and compliant AI-enabled medical devices.
Upon completion of this course, you will be able to:
- Understand fundamental AI and machine learning concepts relevant to medical devices.
- Interpret current regulatory requirements for AI-enabled medical products.
- Navigate MDR, IVDR, EU AI Act, FDA, and international guidance expectations.
- Apply Good Machine Learning Practice (GMLP) principles throughout the AI lifecycle.
- Identify AI-specific risks and implement appropriate mitigation strategies.
- Improve data quality, traceability, transparency, and documentation practices.
- Understand cybersecurity and privacy considerations unique to AI systems.
- Support verification and validation activities for AI-enabled devices.
- Develop effective lifecycle management approaches, including monitoring and change control.
- Increase organizational readiness for audits, inspections, and regulatory submissions involving AI technologies.
This instructor-led virtual classroom course combines expert lectures, practical examples, interactive discussions, and guided exercises to promote effective learning and knowledge transfer.
The training includes:
- Live instructor-led presentations
- Interactive discussions and Q&A
- Real-world medical device examples
- Group exercises and workshop activities
- Case-study analysis
- Risk assessment activities
- Knowledge checks and practical exercises
- Peer-to-peer learning and experience sharing
Participants interact with the instructor and fellow attendees in real time through webinar technology.
Participants who attend at least 90% of the total training duration will receive an official Certificate of Attendance from TÜV SÜD Academy.
There are no mandatory prerequisites.
However, a basic understanding of the following is beneficial:
- Medical device development processes
- Medical device regulations (MDR, IVDR, FDA)
- Quality management systems (ISO 13485)
- Software development concepts
- Risk management principles
No prior expertise in artificial intelligence or machine learning is required.
What is the main focus of this course?
This course focuses on the application of artificial intelligence and machine learning within medical devices, covering technical fundamentals, regulatory requirements, risk management, cybersecurity, and lifecycle management.
Is this course technical or regulatory?
The course provides a balanced combination of both. Participants gain foundational AI knowledge while also learning about regulatory and quality requirements applicable to AI-enabled medical devices.
Do I need prior AI experience?
No. The course introduces core AI and machine learning concepts before moving into more advanced regulatory and quality topics.
Which regulations are covered?
The course covers MDR, IVDR, the EU AI Act, FDA expectations and guidance, along with relevant international standards and industry recommendations.
Does the course cover Good Machine Learning Practice (GMLP)?
Yes. Participants will learn the key GMLP principles and how they can be applied throughout the AI product lifecycle.
Are cybersecurity topics included?
Yes. The course addresses AI-specific cybersecurity concerns such as model poisoning, model stealing, prompt injection, privacy risks, and secure AI development practices.
Will data quality and bias be discussed?
Yes. Data quality, representativeness, data governance, bias mitigation, and traceability are important topics throughout the course.
Does the course explain how AI software is classified as a medical device?
Yes. Participants will learn when AI software qualifies as a medical device and how classification requirements apply under relevant regulations.
Is risk management included?
Yes. The course covers AI-specific risk management concepts, ISO 14971 considerations, BS/AAMI 34971:2023 principles, and AI-related CAPA processes.
What is a Predetermined Change Control Plan (PCCP)?
A PCCP is a documented approach for managing planned modifications to AI-enabled software after market release. The course explains PCCP concepts and regulatory expectations.
Does the course include practical exercises?
Yes. Interactive exercises, case studies, discussions, and practical examples are integrated throughout the course.
How is the course delivered?
The training is delivered live online in a virtual classroom led by an expert instructor.
Will participants receive a certificate?
Participants who meet the attendance requirements will receive an official Certificate of Attendance.
Who should attend this course?
The course is suitable for regulatory, quality, software, engineering, cybersecurity, product management, clinical, and compliance professionals working with AI-enabled medical devices.
What will I be able to do after completing this course?
You will be able to understand the technical and regulatory foundations of AI-enabled medical devices, identify risks and compliance requirements, and support the development and maintenance of AI systems throughout the product lifecycle.
The seminar will be conducted by an expert with many years of practical experience in the medical technology industry. It conveys complex regulatory requirements in a clear and practical way.