Klaus Feuerhelm
Regierungspräsidium Tübingen
Referent:innen
Felix Krumbein
Head ECA Visual Inspection Group
Haluk Dönmez
B. Braun
All times mentioned are CET.
Zielsetzung
It is the aim of this event to inform about the possibilities and limitations of Artificial Intelligence in the automated visual inspection of parenterals. In addition, Good Machine Learning Practices throughout the entire process will be explained and solutions will be presented on how AI projects can be established and validated in a riskbased and traceable manner in the GMP Environment.
Hintergrund
The pharmaceutical industry is increasingly interested in AI for the visual inspection of parenterals to optimize and enhance process efficacy. However, the lack of specific regulatory requirements for AI validation poses challenges from a Good Manufacturing Practice (GMP) perspective, such as data representativeness, model design, and data integrity throughout the product lifecycle.
In visual inspection, AI aims to improve efficiency by reducing the false acceptance rate (FAR) of defect units and the false reject rate (FRR) of good units, which together determine the misclassification rate and the inspection process‘s effectiveness. A high FAR is associated with a possible quality risk, while the FRR is a measure of the economic damage of the selected Control process.
Despite its potential, the FDA‘s guidance on automated Systems mentions AI only briefly, highlighting the need for comprehensive regulation and addressing technical challenges like training, domain knowledge, and data quality. Implementing AI Systems requires specialized expertise, precise data labelling, and Cloud computing for model training.
At this online training, we will be focussing on GMP regulation and technical aspects. Questions such as
- What expectations can be placed on the achievable false reject rates of AI-supported inspection systems?
- Are there applications or technical limitations that even AI-supported systems cannot solve?
- How to set up a project for switching to AI-based visual inspection?
- What GMP authority requirements are there for such systems?
will be discussed and possible solutions presented.
Zielgruppe
The target group for this event are specialists and managers in the pharmaceutical industry from the fields of engineering, production and quality assurance who are involved in the organisation or operation of visual inspection. This conference is also aimed at suppliers involved in the development and automation of inspection systems.
Date & Technical Requirements
Date
Thursday, 11 December, 09.00 to approx. 15.30 h
Technical Requirements
We use Webex for our live online training courses and webinars. At https://www.gmp-compliance.org/training/online-training-technical-information you will find all the information you need to participate in our events and you can check if your system meets the necessary requirements to participate. If the installation of browser extensions is not possible due to your rights in the IT system, please contact your IT department. Webex is a standard nowadays and the necessary Installation is fast and easy.
Programm
Gesamtes Programm als PDF herunterladen
Artificial Intelligence (AI) in Visual Inspection from a GMP Inspector‘s Perspective
- Legal Basis
- GAMP® and AI (ML)
- Validation
- Operation and raw data
Application, Project Planning and Qualification of AI in fully automated Visual Inspection
- 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 / pretrained models, labelling application
- Documentation of model development: traceability, risk minimisation and build-up of confidence
Usage of AI in the Inspection of hard-to-inspect Container-Systems
- Manual, semi-automated and fully-automated approaches
- Use of Artificial Intelligence
- Single chamber and multi-chamber bags
- Inspection of Blow-Fill-Seal containers
- Inspection of Form-Fill-Seal containers
- General Approach
- Training and Machine Learning
- Testing and Validation
- Limitations
| ECA-Member*: | € 990,- |
| Regular Fee*: | € 1190,- |
| EU/GMP Inspectorates*: | € 595,- |
| APIC Member Discount*: | € 1090,- |
Alle Preise zzgl. MwSt. Wichtige Hinweise zur Umsatzsteuer.
* auch unkompliziert per Kreditkarte bezahlbar
Weitere Termine vor Ort
Weitere Termine vor Ort nicht verfügbar
Weitere Termine online
Weitere Termine online nicht verfügbar
Aufzeichnung nicht verfügbar
Haben Sie noch Fragen?
Wir stehen Ihnen für weitere Auskünfte gerne zur Verfügung.
E-Mail: info@concept-heidelberg.de
Teilnehmerstimmen
Seminarprogramm als PDF
Sonderkonditionen DB BahnTeilnehmerstimmen - das sagen andere über unsere Seminare:
