15th Visual Inspection of Parenterals Conference - Live Online Training

Programm

• Compendial requirements
  • 100% visual inspection & AQL testing
  • PharmEur, USP, JP - similarities and differences
• News from the Annex I
• Risk management considerations

• Recent FDA Form 483 and Warning Letter observations involving visual inspection and visible particulate manufacturing controls
• The regulatory and compendial basis for FDA concerns about visual inspection
• Risk assessment techniques for the evaluation of visual defect criticality, particle source, and patient impact

• External sources (packaging material, filter, abrasion..)
• Internal sources (product and inherent particles)
• Potential risks for patients
• Route cause detection and particle identification
• Avoidance and depletion of particles

• Applicable regulations & guidance
• Current requirements for pharmaceutical Industry
• Expectations of a GMP Inspector
• Examples of observations

• Framework alignment: <790> (compliance), <1790> (science), TR 79 (lifecycle/DIPs), ECA v5.0 (practice)
• Human inspection capability: PoD and station standards
• Defect taxonomy & risk classification (DIP-aware)
• Method capability & qualification — per defect category
• Automated inspection (AVI): PoD‑based validation & requalification
• Process control, trending & CAPA (lifecycle)
• Inspector training & requalification — per defect class
• Test‑set design — separate, blinded kits per category with adequate blanks
• Defect relevance, CCI and special containers
• Continuous improvement & robustness

• The growing role of Automated Visual Inspection (AVI) in quality, efficiency, and compliance
• Key challenges of AVI implementation from a CDMO perspective
• Balancing standardisation with customer-specific flexibility
• Managing regulatory expectations, internal quality requirements, and client demands
• Practical aspects of AVI governance and validation
• Collaboration with customers and technology suppliers throughout the system lifecycle
• Lessons learned from implementation, operation, and continuous improvement of AVI systems

• Inspection of Difficult to Inspect Parenteral Products
• Single chamber and multi-chamber bags
• Inspection of Blow-Fill-Seal containers
• Inspection of Form-Fill-Seal containers
• Manual, semi-automated and fully automated approaches
• Two step inspection

• Use of artificial intelligence
• General approach machine learning
• Training, validation and testing
• False reject rates
• Limitations

• Development of robust, reliable and production-ready models in 4 phases
  • Phase 1: Problem identification & description
  • Phase 2a: Specification of inspection Concept
  • Phase 2b: Definition of the sample sets (artificial and production samples), creation of the datasets, clarification of the labelling strategy
  • Phase 3: Model design, training and verification - a risk-based Approach
  • Phase 4: Qualification & validation
• Processes & Technologies
  • Technologies for efficient image data acquisition, variable model technologies, transfer learning / pre-trained models, labelling application
  • Documentation of model development: traceability, risk minimisation and build-up of confidence

• Multispectral lighting and image processing enable the analysis of multiple information channels from a single image, improving the visibility of relevant details
• A single-camera approach can replace more complex multi-camera setups and provide the required inspection information in a compact Format
• Combined lighting modes such as transmitted, reflected, and dark-field illumination help to highlight different material characteristics and defect types
• Intelligent image evaluation supports the reliable detection of defects such as particles, scratches, and contamination while reducing interference from regular packaging features.