Exploring AI/ML in Tablet Compression: Insights from LifeFactFuture

This autumn, Professor Anne Juppo and Dr. Yazid Bounab from the Faculty of Pharmacy at the University of Helsinki have been sharing the promising results from the LifeFactFuture project, particularly regarding the application of AI and machine learning (ML) in tablet compression. The response from audiences across Europe has been overwhelmingly positive, leading to numerous invitations for further presentations and collaborative opportunities with both academic and industry partners.

8th International Symposium on Solid Oral Dosage Forms

At the 8th International Symposium on Solid Oral Dosage Forms held on September 22–23 at AstraZeneca R&D Gothenburg in Mölndal, Sweden, Professor Juppo delivered her oral presentation as an invited speaker. This event provided a valuable platform to present our findings to an audience of around 200 international academics and industry professionals. A common theme emerged during discussions: the pharmaceutical industry is grappling with scattered and fragmented data. Participants highlighted how variations in analytical methods further complicate this issue. Notably, our findings aligned with other presentations at the conference, reinforcing the idea that deep neural networks (DNNs) hold significant promise for modeling pharmaceutical processes.

In addition, Dr. Bounab engaged in fruitful discussions at our poster session, which offered a deeper understanding of our platform’s feasibility. These conversations have sparked potential collaborations with University of Eastern Finland and various suppliers of excipients and production machinery.

PSSRC Annual Meeting

Following the symposium, we attended the annual meeting of the Pharmaceutical Solid State Cluster from September 24–26 at TU Graz in Austria. Our poster garnered considerable attention among over 100 attendees from academia and industry. We initiated two academic collaborations and discussed plans for Dr. Bounab to undertake a LifeFactFuture secondment with Professor Johannes Khinast’s team at Technical University Graz in April next year.

Nordic POP and Nordic Pharma Train Workshop

On November 3–4, Professor Juppo was invited to speak at the Nordic POP and Nordic Pharma Train Workshop in Reykjavik, Iceland. Her talk focused on AI and ML from both manufacturing and formulation development perspectives. A striking observation was that, despite the growing interest in these technologies, only 33 scientific papers published in the last decade have genuinely employed novel AI/ML tools in formulation development. Most publications in this area have been either reviews or speculative in nature. The discussions highlighted the importance of high-quality data for training effective models, and among the 90 attendees, only about five had practical experience with AI/ML tools in their research.

Bioseminar Series at Uppsala University

Lastly, on November 6, Professor Göran Frenning invited Professor Juppo to present at the Bioseminar series at Uppsala University. This presentation revisited key insights from the Mölndal conference, sparking discussions on the philosophical dimensions of “intelligence” in AI and the critical role of data quality and quantity in model development. The audience expressed great interest in the LifeFactFuture project and eagerly anticipates future updates on our findings.

As we continue to explore the intersection of AI/ML and pharmaceutical manufacturing, we are grateful for the engagement and support from our peers and look forward to advancing our collaborative efforts in this exciting multidisciplinary field.

Anne Juppo, Professor 
anne.juppo@helsinki.fi

Yazid Bounab, Postdoctoral Researcher
yazid.bounab@helsinki.fi

Division of Pharmaceutical Chemistry and Technology, Faculty of Pharmacy
University of Helsinki

– – –

Article photo: Professor Anne Juppo at the 8th International Symposium on Solid Oral Dosage Forms, Gothenburg in Mölndal, Sweden.