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

High-throughput identification of bacterial β-glucuronidase inhibitors using machine learning.

Gut microbes·2026
Same author

Biorelevant assessment of commercial probiotics using porcine gastrointestinal fluids.

Journal of applied microbiology·2026
Same author

Active learning in pharmaceutical 3D printing: a multi-dataset comparison.

Drug delivery and translational research·2026
Same author

A water-based probiotic supplement shows antipathogenic activity against Clostridium perfringens, Klebsiella pneumoniae, and Listeria monocytogenes.

International journal of pharmaceutics·2026
Same author

A multi-strain probiotic modulates gut microbiome composition, intestinal barrier integrity and inflammation in a multi-compartmental in vitro gut model of decompensated advanced chronic liver disease.

International journal of pharmaceutics·2026
Same author

Effects of a Four-Strain Probiotic on Gut Microbiota, Inflammation, and Symptoms in Parkinson's Disease: A Randomized Clinical Trial.

Movement disorders : official journal of the Movement Disorder Society·2025
Same journal

Approaches for enhancing bioavailability of macromolecular drugs.

International journal of pharmaceutics·2026
Same journal

Characteristics of asymmetric microcrystalline solidification pellets and a better prediction for bioequiavailability based on solubility-permeability theory.

International journal of pharmaceutics·2026
Same journal

A CFPD-FSI analysis of the impact of nasal hairs on airflow patterns, nasal resistance, and particle filtration in a realistic human nasal airway.

International journal of pharmaceutics·2026
Same journal

Transient hydrate formation during disintegration of film-coated pharmaceutical tablets.

International journal of pharmaceutics·2026
Same journal

Use of a hydrophilic excipient to modulate the release of dapivirine, copper ions and zinc ions from a matrix-type vaginal ring.

International journal of pharmaceutics·2026
Same journal

Reprogramming chronic wounds: the emerging role of microbiome-targeted therapies in diabetic foot ulcers.

International journal of pharmaceutics·2026
See all related articles

Related Experiment Video

Updated: Jan 11, 2026

Voxel Printing Anatomy: Design and Fabrication of Realistic, Presurgical Planning Models through Bitmap Printing
11:36

Voxel Printing Anatomy: Design and Fabrication of Realistic, Presurgical Planning Models through Bitmap Printing

Published on: February 9, 2022

3.2K

Machine learning recovers corrupted pharmaceutical 3D printing formulation data.

Olima Uddin1, Yusuf Ali Mohammed1, Simon Gaisford2

  • 1School of Biological and Behavioural Sciences, Queen Mary University of London, Mile End Road, London E1 4DQ, UK.

International Journal of Pharmaceutics
|November 18, 2025
PubMed
Summary
This summary is machine-generated.

Machine learning, specifically denoising autoencoders (DAEs), can effectively reconstruct corrupted pharmaceutical formulation data, safeguarding personalized medicine production against cyberattacks and ensuring patient safety.

Keywords:
Artificial IntelligenceDigital ResilienceDrug DevelopmentFused Deposition ModellingQuality Control

More Related Videos

Multimodal 3D Printing of Phantoms to Simulate Biological Tissue
05:11

Multimodal 3D Printing of Phantoms to Simulate Biological Tissue

Published on: January 11, 2020

8.0K
3D Printing of Preclinical X-ray Computed Tomographic Data Sets
11:06

3D Printing of Preclinical X-ray Computed Tomographic Data Sets

Published on: March 22, 2013

41.0K

Related Experiment Videos

Last Updated: Jan 11, 2026

Voxel Printing Anatomy: Design and Fabrication of Realistic, Presurgical Planning Models through Bitmap Printing
11:36

Voxel Printing Anatomy: Design and Fabrication of Realistic, Presurgical Planning Models through Bitmap Printing

Published on: February 9, 2022

3.2K
Multimodal 3D Printing of Phantoms to Simulate Biological Tissue
05:11

Multimodal 3D Printing of Phantoms to Simulate Biological Tissue

Published on: January 11, 2020

8.0K
3D Printing of Preclinical X-ray Computed Tomographic Data Sets
11:06

3D Printing of Preclinical X-ray Computed Tomographic Data Sets

Published on: March 22, 2013

41.0K

Area of Science:

  • Pharmaceutical manufacturing
  • Digital health
  • Machine learning applications

Background:

  • 3D printing of pharmaceuticals enables personalized medicine but introduces cybersecurity risks.
  • Cyberattacks can corrupt critical drug formulation data, jeopardizing patient safety.
  • Developing robust data protection methods is crucial for digital pharmaceutical manufacturing.

Purpose of the Study:

  • To investigate the efficacy of denoising autoencoders (DAEs) in reconstructing corrupted pharmaceutical formulation data.
  • To simulate and address potential cyberattack scenarios affecting digital drug manufacturing data.
  • To enhance the digital resilience of pharmaceutical data integrity.

Main Methods:

  • Utilized a dataset of 1,623 pharmaceutical formulations (336 ingredients, >545,000 data points).
  • Simulated cyberattacks by introducing data deletion (1%-50%) and noise injection.
  • Evaluated multiple DAE configurations for data recovery performance.

Main Results:

  • DAEs achieved high R² scores (0.989 at 1% corruption, 0.924 at 50% corruption).
  • DAEs accurately reconstructed active pharmaceutical ingredient and excipient values, indicating meaningful pattern recognition.
  • Traditional machine learning methods failed to recover corrupted data, unlike DAEs.

Conclusions:

  • Denoising autoencoders effectively safeguard pharmaceutical formulation data against corruption.
  • DAEs offer a promising solution for enhancing digital resilience in pharmaceutical manufacturing.
  • Machine learning plays a vital role in maintaining data quality and patient safety in the digital pharmaceutical sector.