Sign in Subscribe
Artificial neural networks

AI Accelerates Materials Design and Development in Nanotechnology

Strategic Insights

— 1 min read

Nanotechnology researchers at Johns Hopkins University have discovered that artificial intelligence (AI) models can significantly accelerate materials design and development. By employing convolutional neural networks (CNN) models, the researchers were able to characterize DNA origami nanostructures, which have numerous applications in biomedicine. The study, published in the Journal of Chemical Information and Modeling, demonstrated that fine-tuned VGG16 models can quickly and accurately determine the number of ligation sites of nanostructures in large transmission electron microscopy (TEM) images.

Key Takeaways:

  • The researchers pre-trained 9 CNN models using a large image dataset of 720 images from coarse-grained molecular dynamics (MD) simulations, and then fine-tuned them using a small experimental TEM data set with 146 TEM images.
  • All CNN models had similar computational time requirements, despite differences in model sizes and performances.
  • Among the pre-trained models, ResNet50 and VGG16 had the highest and second-highest accuracies, while among the fine-tuned models, VGG16 was found to have the highest agreement with test TEM images.
  • The study demonstrated the potential of AI in accelerating materials design and development in nanotechnology.
  • The research has been peer-reviewed and published in the Journal of Chemical Information and Modeling.

Statistics:

  • 720: the number of images used to pre-train the CNN models
  • 146: the number of TEM images used to fine-tune the CNN models
  • 9: the number of CNN models benchmarked in the study
  • 20: the number of test MD images used to evaluate the models
  • 79.5%: the accuracy of ResNet50 in characterizing DNA origami nanostructures
  • 78.2%: the accuracy of VGG16 in characterizing DNA origami nanostructures

Sources:

  • "Artificial intelligence (AI) models remain an emerging strategy to accelerate materials design and development." - Johns Hopkins University
  • "Characterizing DNA Origami Nanostructures in TEM Images Using Convolutional Neural Networks." Journal of Chemical Information and Modeling, 2025.
  • NewsRx. Johns Hopkins University Reports Findings in Nanostructures (Characterizing DNA Origami Nanostructures in TEM Images Using Convolutional Neural Networks). Journal of Engineering. July 7, 2025; p 1594.

Read more