(AI) Artificial Intelligence in a GxP Environment - Live Online Training

Programm

Gesamtes Programm als PDF herunterladen

• History of AI
• Types of AI
• Real life examples

• Technological Basics
• Different learning / training Methods
• Example use cases

• Challenges for pharmaceutical Industry
• The GAMP-perspective on AI/ML
• GMLP: SDLC for AI/ML
• Use cases / Known scenarios

• Pre-requisites for AI: Data Governance, Data Science, Data Understanding, FAIRification.
• Specificities in the lab
  • Lack of standardization, Lack of connectivity.
  • Chances with standards: AniML, SiLA
• Specificities in clinical development

• US / FDA: Expectations and guidance
• EU / EMA: Expectations and guidance
• Special case: Generative AI

• Pharmaceutical laws (AMG and other)
• EU-GMP Guide Annex 11
• Concept Paper Revision of Annex 11
• Software as Medical Device

• Maturity: Increasing autonomy and transferring Control
• Governance: Developing and operating AI solutions in GxP-regulated areas

• Power with control: Explaining the outcomes of trained models
• Applying QRM to development and operation of AI applications

• Inspection strategy
• What do inspectors expect from the regulated user?

• Apply Advanced Analytics to enable predictive Titer/Yield and reduce variability while increasing mean toward high-end value
• In-Flight predictive and adaptive process oversight for shop floor to target Titer/Yield Golden Batches
• Prove the value of utilizing Advanced Analytics as a digital product leveraging different data sources and advanced predictive algorithms
• Build site future capabilities required for a sustainable way of working using Advanced Analytics

• Introduction on the regulations in Medical Device area
• AI in Medical Device:
  • Patient risk: more direct than in Pharma?
  • Reality not future: FDA list of devices released.
• Guidelines on AI:
  • (FDA GMLP > optional as already covered in one of the other sessions)
  • AI & Cybersecurity (ENISA guideline)
  • NMPA Guideline on AI
• Examples of a use case:
  • Electrocardiogram analysis with AI

• Pain points in visual inspection
• A machine-agnostic AI solution Framework
• Strategies for developing robust and reliable AI models
• Qualification and necessary documentation
• Results of AI powered visual inspection

• Introduction and Basics
• Application and Challenges
• Approach and ML Training
• Testing and Qualification
• Conclusion

• Introduction and Overview of the Use Case
• Consolidating the Tablet Production Process into a Comprehensive Dataset
• Theoretical Basics of Machine Learning and Gradient Boosting Decision Trees
• Application of Machine Learning to Identify Critical Factors Impacting Production Target Variables
• Key Takeaways and Lessons Learned