Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Editorial: Educating the educators in digital STEM-education - the impact of teacher training and their further education.

Frontiers in psychology·2026
Same author

Crucial Gram-positive type IV secretion system protein TraF is a structural homolog of type VII secretion system protein EssB/YukC.

microLife·2026
Same author

Effect of tablet press residence time distribution on material traceability in a continuous direct compression process via mini-batch blending - part 1: corroboration and application.

International journal of pharmaceutics·2026
Same author

Microarrayed Allergen Molecules Distinguish IgE Sensitisation to Blomia tropicalis and Dermatophagoides pteronyssinus.

Allergy·2026
Same author

Allergen-specific human IgE isolated through an allergen-agnostic pipeline-understanding immune response and allergen recognition.

Communications biology·2026
Same author

Magnetic Properties of Ferritin at Different Levels of Degradation: Implications for MRI-Based Iron Quantification in the Brain.

Magnetic resonance in medicine·2025

Related Experiment Video

Updated: Jul 26, 2025

Author Spotlight: Process Development for the Spray-Drying of Probiotic Bacteria and Evaluation of the Product Quality
05:45

Author Spotlight: Process Development for the Spray-Drying of Probiotic Bacteria and Evaluation of the Product Quality

Published on: April 7, 2023

3.5K

A multi-step machine learning approach for accelerating QbD-based process development of protein spray drying.

Daniela Fiedler1, Elisabeth Fink2, Isabella Aigner2

  • 1Graz University of Technology, Institute of Process and Particle Engineering, Inffeldgasse 13/III, 8010 Graz, Austria.

International Journal of Pharmaceutics
|June 14, 2023
PubMed
Summary

A novel machine learning (ML) approach using surrogate materials significantly reduces experiments needed for developing protein spray drying design spaces (DS). This method optimizes complex biologic processes efficiently, minimizing costly trial-and-error for spray drying optimization.

Keywords:
Artificial neural networksBiologicsDesign of experimentsMachine learningProteinSpray drying

More Related Videos

A Workflow for Lipid Nanoparticle LNP Formulation Optimization using Designed Mixture-Process Experiments and Self-Validated Ensemble Models SVEM
13:54

A Workflow for Lipid Nanoparticle LNP Formulation Optimization using Designed Mixture-Process Experiments and Self-Validated Ensemble Models SVEM

Published on: August 18, 2023

4.7K
Process Optimization using High Throughput Automated Micro-Bioreactors in Chinese Hamster Ovary Cell Cultivation
09:28

Process Optimization using High Throughput Automated Micro-Bioreactors in Chinese Hamster Ovary Cell Cultivation

Published on: May 18, 2020

8.5K

Related Experiment Videos

Last Updated: Jul 26, 2025

Author Spotlight: Process Development for the Spray-Drying of Probiotic Bacteria and Evaluation of the Product Quality
05:45

Author Spotlight: Process Development for the Spray-Drying of Probiotic Bacteria and Evaluation of the Product Quality

Published on: April 7, 2023

3.5K
A Workflow for Lipid Nanoparticle LNP Formulation Optimization using Designed Mixture-Process Experiments and Self-Validated Ensemble Models SVEM
13:54

A Workflow for Lipid Nanoparticle LNP Formulation Optimization using Designed Mixture-Process Experiments and Self-Validated Ensemble Models SVEM

Published on: August 18, 2023

4.7K
Process Optimization using High Throughput Automated Micro-Bioreactors in Chinese Hamster Ovary Cell Cultivation
09:28

Process Optimization using High Throughput Automated Micro-Bioreactors in Chinese Hamster Ovary Cell Cultivation

Published on: May 18, 2020

8.5K

Area of Science:

  • Chemical Engineering
  • Materials Science
  • Biotechnology

Background:

  • Developing design spaces (DS) for spray drying proteins typically requires extensive Design of Experiments (DoE), which is costly due to expensive biologics.
  • Optimizing spray drying processes for biologics necessitates minimizing experimental runs while maintaining accuracy.

Purpose of the Study:

  • To investigate the efficacy of a material-efficient, multi-step machine learning (ML) approach for developing a DS for protein spray drying.
  • To evaluate the suitability of using a surrogate material, specifically lactose, in conjunction with ML for DS development.
  • To compare the predictive performance of ML models against traditional DoE models.

Main Methods:

  • A Design of Experiments (DoE) was conducted using a surrogate material (lactose) to generate training data for a machine learning (ML) model.
  • A multi-step ML approach was developed and compared against a benchmark DoE approach using multivariate regression.
  • Model predictions from both ML and DoE approaches were validated against experimental runs using actual protein formulations.

Main Results:

  • The ML approach, utilizing a surrogate material, demonstrated suitability for developing a protein spray drying DS, reducing experimental burden.
  • Lactose was found to be a suitable surrogate material, and the proposed ML approach showed advantages over traditional DoE.
  • Limitations were observed for protein concentrations exceeding 35 mg/ml and particle sizes larger than 6 µm.
  • Within the studied DS, protein secondary structure was preserved, with yields typically above 75% and residual moisture below 10 wt%.

Conclusions:

  • The proposed material-efficient ML approach offers a viable and cost-effective alternative for developing spray drying design spaces for proteins.
  • This method is particularly beneficial for expensive biologics where minimizing experimental runs is crucial.
  • Further investigation is needed to address limitations at higher protein concentrations and particle sizes.