Pharmaceutical manufacturing is evolving fast—and so are the tools that make innovation possible. A recent peer-reviewed publication in Chimica Oggi – Chemistry Today demonstrates how scientists are using SuntheticsML to unlock new insights in spray freeze drying (SFD) and dynamic lyophilization — two critical processes in developing stable, high-quality pharmaceutical products.

This study, led by Howard J. Stamato and colleagues from Meridion Technologies, showcases how researchers are applying our machine learning platform to achieve remarkable process understanding—all without coding or Machine Learning expertise.

Machine Learning Meets Spray Freeze Drying

Spray freeze drying is a cutting-edge process that transforms liquid formulations into frozen pellets, later dried under vacuum conditions. It’s used to enhance product stability, shelf life, and dosage flexibility — but it’s also notoriously complex. With so many interacting parameters (temperature, pressure, infrared energy, drum rotation speed, etc.), experimentation can become expensive and time-consuming.

That’s where machine learning—and specifically SuntheticsML—changes the game.

Using Bayesian Optimization, researchers were able to model over 80 experimental runs, visualize relationships between inputs and outputs, and even predict the most efficient drying conditions. The platform allowed them to analyze and clean their data, build predictive models in just hours, and generate new experimental designs automatically.

The result? Faster insight. Fewer experiments. Smarter science.

Key Takeaways from the Study

The publication highlights how SuntheticsML empowered the research team to:

• Explore large datasets from multiple instruments and scales without writing a single line of code.
  
  
• Understand variable importance — discovering which parameters most influenced drying time, yield, and moisture.
  
  
• Visualize complex relationships through dynamic 3D plots, correlation maps, and error analysis.
  
  
• Optimize for multiple objectives at once (like minimizing drying time while maximizing yield).
  
  
• Generate new experimental plans in minutes — reducing the need for trial and error.

This is a clear demonstration of how AI and ML can augment scientific intuition, helping experts validate hypotheses, uncover hidden trends, and accelerate R&D across the product lifecycle.

A Glimpse of the Future

At Sunthetics, it’s inspiring to see our platform enabling breakthroughs like this. These results reflect the true power of AI-driven modeling — not as a replacement for scientific expertise, but as a catalyst for it.

Machine learning is helping scientists spend less time cleaning data and running repetitive tests, and more time doing what they do best: innovating.

As more users across pharmaceuticals, materials, and chemical engineering continue to adopt SuntheticsML, we’re seeing a shift toward data-driven experimentation — where every result feeds back into a smarter, more sustainable future for R&D.

Curious to Learn More?

Explore how scientists are using SuntheticsML to accelerate discovery, improve process understanding, and reduce development costs.

Read the full publication in Chimica Oggi – Chemistry Today (Vol. 43, 2025): “Artificial Intelligence (AI) Driven Machine Learning Modeling for Process Characterization of Dynamic Freeze Drying (Lyophilization) After Spray Freezing."

Thank you, Howard J. Stamato from Stamato Solutions and Bernhard Luy, Matthias Plitzko, colleagues from Meridion Technologies.