Syntegon Technology recently installed the first fully validated visual inspection system that utilises AI in an automated inspection machine in a pharmaceutical customer’s production line. The technology was four years in the making.
The move constitutes a major step in the company’s effort to introduce AI to pharmaceutical visual inspection, which is a largely uncharted territory for the industry.
As one of the most challenging stages in the pharmaceutical manufacturing process, inspection requires ever more sophisticated visual systems to process increasingly complex products.
“Especially for high-cost pharmaceuticals, every single false reject is one too many,” said Dr Jose Zanardi, who is responsible for vision inspection development and applications at Syntegon.
“This important move is the joint result of long-standing visual inspection expertise, solid software and pharmaceutical validation competence, courage to cross boundaries, and an excellent partnership with our customer.”
AI applications have the potential of further increasing detection rates and decreasing the number of false rejects in difficult products like highly viscous parenteral solutions with air bubbles, which are sometimes hard to differentiate from harmful particles.
Zanardi explains that this is due to AI utilising Deep Learning algorithms, which are capable of accurately identifying recurring patterns and deviations.
“A growing number of Deep Learning vision applications are already on the market. Our task was to adapt those applications for pharmaceutical purposes, which essentially also includes validation,” adds Zanardi.
“In fact, thanks to thorough in-house software and visual inspection expertise, Syntegon was able to develop a solution that only requires moderate modifications to the already existing vision systems.”
Global biotechnology company, Amgen, uses the system to reliably distinguish air bubbles at the syringe’s rubber stopper from foreign particles, where conventional vision technology often mistakenly identifies safe products containing bubbles as defective.
In this customer project, Syntegon’s AI-based vision system was able to increase the particle detection rate by 70 per cent, while reducing the false detection rate by 60 per cent.
“This challenging project required a lot of dedication and expertise and we have implemented the world-first syringe inspection machine with AI and underline our market position, both in biotechnology production and technology,” said Manuel Soto, principal process development engineer at Amgen.
Based on this success, Syntegon is set to implement AI in further inspection machines for different products and container types, thus contributing to the production of safe and reliable pharmaceutical products worldwide.
