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

Comparison of neural network and multiple linear regression as dissolution predictors.

Pradeep M Sathe1, Jurgen Venitz

  • 1U.S. Food and Drug Administration, Office of Generic Drugs, Division of Bioequivalence, Rockville, Maryland 20855, USA. sathe@cder.fda.gov

Drug Development and Industrial Pharmacy
|May 14, 2003
PubMed
Summary

Artificial neural networks (NN) outperformed multiple linear regression (MLR) in predicting diltiazem dissolution profiles. NN demonstrated superior internal and external predictability for formulation variables influencing drug release.

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

Quantitative assessment of the in-vitro binding kinetics of antisickling aromatic aldehydes with hemoglobin A: A universal HPLC-UV/Vis method to quantitate Schiff-base adduct formation.

Journal of pharmaceutical and biomedical analysis·2022
Same author

Identifying a patient-centered outcome measure for a comparative effectiveness treatment trial in myasthenia gravis.

Muscle & nerve·2021
Same author

VZHE-039, a novel antisickling agent that prevents erythrocyte sickling under both hypoxic and anoxic conditions.

Scientific reports·2020
Same author

Exploration of Structure-Activity Relationship of Aromatic Aldehydes Bearing Pyridinylmethoxy-Methyl Esters as Novel Antisickling Agents.

Journal of medicinal chemistry·2020
Same author

Pharmacokinetic profile analyses for inhaled drugs in humans using the lung delivery and disposition model.

Biopharmaceutics & drug disposition·2019
Same author

Discovery of Second-Generation NLRP3 Inflammasome Inhibitors: Design, Synthesis, and Biological Characterization.

Journal of medicinal chemistry·2019

Area of Science:

  • Pharmaceutical Sciences
  • Computational Chemistry
  • Drug Delivery Systems

Background:

  • Accurate prediction of drug dissolution is crucial for pharmaceutical development.
  • Traditional statistical models like multiple linear regression (MLR) have limitations in capturing complex relationships in dissolution data.
  • Artificial neural networks (NN) offer a potential alternative for modeling intricate dissolution profiles.

Purpose of the Study:

  • To compare the predictive performance of artificial neural networks (NN) against first-order multiple linear regression (MLR).
  • To evaluate the predictability of Weibull function parameters (alpha and beta) derived from diltiazem immediate release tablet dissolution data.
  • To determine the influence of formulation variables on dissolution markers using both MLR and NN models.

Main Methods:

Related Experiment Videos

  • Collected mean dissolution data from 28 diltiazem immediate release tablet formulations.
  • Fitted the Weibull function to dissolution profiles to obtain alpha and beta parameters.
  • Utilized a three-layered 8-5-2 feedforward artificial neural network (NN) and first-order multiple linear regression (MLR).
  • Assessed internal and external predictability using bias (mean prediction error; MPE) and precision (mean absolute error; MAE).

Main Results:

  • The artificial neural network (NN) proved to be an adequate descriptor of the dissolution data.
  • NN demonstrated superior internal and external predictive performance compared to MLR for the studied diltiazem formulations.
  • The study identified the order of formulation composition variables influencing dissolution parameters, with hydrogenated oil and microcrystalline cellulose having the most significant impact.

Conclusions:

  • Artificial neural networks (NN) are superior predictors to multiple linear regression (MLR) for diltiazem dissolution data.
  • NN models can effectively capture complex relationships between formulation variables and drug release characteristics.
  • The findings provide valuable insights for optimizing diltiazem tablet formulations for desired dissolution profiles.