sGAN

Design pharmaceutical microstructures in silico with digiM’s GAN technology. Simulate formulations, predict performance, and accelerate development.

Tablet
sGAN

sGAN

sGAN

sGAN

Digital Formulation Development

Our structure synthesis platform is a patent-pending approach that uses generative adversarial networks (GAN) to create 3D microscopic volumes replicated from imaging scans (WO2024138171A1). Through training on images like X-ray micro-CT and FIB-SEM, it generates new images with controlled modifications, while adhering to some component of the original sample's structure (eg a different drug load, but with the same particle size). The result is a new digital formulation in the form of a microscopic dataset, enabling scientists to digitally explore parameters like drug load, dispersion patterns, and particle size. Combined with released predictions and dissoLab, GAN structure synthesis enables true digital formulation design. 

Read the Nature Communications paper, published with Genentech and Merck, to see structure synthesis applied to implant and oral development.

A cute icon of a lightbulb

3D Generation from 2D Images

Generate realistic 3D microstructures directly from 2D images, saving resources in imaging collection while adhering to real world structural attributes.

A cute icon of a CSV file

Process Design Space

Incorporating information on manufacturing and process parameters, GAN structure synthesis can help in identifying optimal process conditions.

A cute icon of an image landscape

Microstructure Modifications

Easily adjust product attributes such as porosity, particle size, or drug loading to simulate a wide range of formulation design parameters.

A cute icon of a cube.

Test Formulations In Silico

Overcome lengthy trial by testing approaches. Digital formulations combined with in silico dissolution and in vitro release predictions can rapidly assess the impact of drug load and particle size.

For pharmaceutical scientists, this means a powerful new tool to speed up formulation design and reduce experimental costs, especially when drug substance samples for testing may be limited. Structure synthesis experiments can be combined with in silico dissolution to simulate drug release profiles and assess how changes in formulation or processing may impact performance. By digitizing and simulating key quality attributes, structure synthesis helps teams make informed decisions earlier, with fewer physical trials, and greater confidence in final product performance.

sGAN

Explore Microstructural Impacts on Performance

Explore how subtle shifts in microstructure like API distribution, particle size, or dispersion affect performance, without making new batches. Our structure synthesis platform can simulate these complex design scenarios in silico, testing ideas that may be too expensive or time-consuming to manufacture and test. Whether you are fine-tuning a controlled-release profile, managing risk from manufacturing variability, or comparing formulation strategies, this digital sandbox opens the door to a new kind of formulation intelligence built on exploration, not trial and error. Watch our webinar to earn more about structure synthesis.

sGAN

Ready to See it in Action?

Whether you're designing new formulations, optimizing release, or assessing process parameters, structure synthesis brings speed and clarity to your drug product development. If you're ready to revolutionize your research, contact us to learn more about how GAN can support your.

GAN structure synthesis is available as a module through digiM I2S or via contract research projects.

Start Your Project

Revolutionize your Product Development

Learn how to advance your pharmaceutical development with GAN's advanced AI capabilities today.