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 Experiment Videos

Pharmaceutical granulation and tablet formulation using neural networks

J G Kesavan1, G E Peck

  • 1Department of Industrial and Physical Pharmacy, School of Pharmacy, Purdue University, West Lafayette, Indiana 47906, USA.

Pharmaceutical Development and Technology
|December 1, 1996
PubMed
Summary
This summary is machine-generated.

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

Drug physical state and drug-polymer interaction on drug release from chitosan matrix films.

Journal of controlled release : official journal of the Controlled Release Society·2001
Same author

Physical properties and molecular behavior of chitosan films.

Drug development and industrial pharmacy·2001
Same author

Accelerated fluid bed drying using NIR monitoring and phenomenological modeling.

Drug development and industrial pharmacy·2000
Same author

The effect of formulation on radioiodide thyroid uptake in the hyperthyroid cat.

Drug development and industrial pharmacy·1999
Same author

The effect of aluminum hydroxide dissolution on the bleeding of aluminum lake dyes.

Pharmaceutical research·1993
Same author

The effect of swelling characteristics of superdisintegrants on the aqueous coating solution penetration into the tablet matrix during the film coating process.

Pharmaceutical research·1993
Same journal

<i>Camellia sinensis</i> Silver Nanoparticles Inhibit Proliferation and Induce PARP-Dependent Apoptosis in A549 Cells: Formulation, Characterization, and Biological Evaluation.

Pharmaceutical development and technology·2026
Same journal

Enhanced solubility and bioavailability of zanubrutinib nanocrystals: Physicochemical and Pharmacokinetic evaluation.

Pharmaceutical development and technology·2026
Same journal

Intradermal delivery of lidocaine hydrochloride using dissolving microneedles: <i>In vitro</i> dermatokinetic test and <i>in vivo</i> anesthetic activity assessment.

Pharmaceutical development and technology·2026
Same journal

Quality by design-guided development of silica-enabled lipid hybrid nanoparticles for enhanced olaparib dissolution.

Pharmaceutical development and technology·2026
Same journal

Surface engineering of nanocarriers with polymer coatings: materials, strategies, functional outcomes, and clinical translation.

Pharmaceutical development and technology·2026
Same journal

Compression equations in pharmaceutical tableting: from classical compaction models to data-driven and mechanistic approaches.

Pharmaceutical development and technology·2026
See all related articles

Neural networks can predict pharmaceutical formulation characteristics, improving upon traditional trial-and-error methods. This study demonstrates their applicability in modeling granulation and tablet properties, offering better predictions than regression models.

Area of Science:

  • Pharmaceutical Sciences
  • Computational Chemistry
  • Chemical Engineering

Background:

  • Pharmaceutical formulation often relies on trial-and-error due to undefined relationships between input (material/process) and output (characteristics) variables.
  • Neural networks offer a data-driven approach to capture complex relationships and enable predictive modeling.

Purpose of the Study:

  • To predict granulation and tablet system characteristics using neural networks based on material and process variables.
  • To evaluate the performance of neural network models against traditional regression methods for formulation prediction.

Main Methods:

  • Developed neural network models using formulation data, including granulation equipment, diluent, binder addition method, and binder concentration.
  • Trained and tested models with material, process, granulation evaluation, and tablet evaluation data.

Related Experiment Videos

  • Compared neural network predictions with regression model outcomes.

    • Main Results:

    • Both granulation and tablet neural network models showed rapid convergence during training.
    • Satisfactory predictions were achieved for granulation variables (particle size, flow, densities) and tablet properties (disintegration, thickness).
    • Neural network predictions generally outperformed or were comparable to regression methods across all variables.

    Conclusions:

    • Neural networks are applicable to pharmaceutical formulation modeling, offering a more predictable approach than trial-and-error.
    • The study highlights the potential of artificial intelligence in optimizing drug development processes.
    • Further investigation into scenarios where neural networks may not perform optimally is warranted.