Synthetic Scientific Image Generation with VAE, GAN, and Diffusion Model Architectures

Sordo, Zineb; Chagnon, Eric; Hu, Zixi; Donatelli, Jeffrey J.; Andeer, Peter; Nico, Peter S.; Northen, Trent; Ushizima, Daniela

doi:10.3390/jimaging11080252

This is an early access version, the complete PDF, HTML, and XML versions will be available soon.

Open AccessArticle

Synthetic Scientific Image Generation with VAE, GAN, and Diffusion Model Architectures

by

Zineb Sordo

¹

,

Eric Chagnon

¹,

Zixi Hu

¹

,

Jeffrey J. Donatelli

¹

,

Peter Andeer

²

,

Peter S. Nico

³

,

Trent Northen

²

and

Daniela Ushizima

^1,4,5,*

¹

Lawrence Berkeley National Laboratory, Applied Math and Computational Research Division, 1 Cyclotron Rd., Berkeley, CA 94720, USA

²

Lawrence Berkeley National Laboratory, Environmental Genomics and Systems Biology Division, 1 Cyclotron Rd., Berkeley, CA 94720, USA

³

Lawrence Berkeley National Laboratory, Earth Sciences Division, 1 Cyclotron Rd., Berkeley, CA 94720, USA

⁴

Bakar Computational Health Sciences Institute, University of California, 480 16th St., San Francisco, CA 94158, USA

⁵

Berkeley Institute for Data Science, 621 Sutardja Dai Hall, University of California, Berkeley, CA 94720, USA

^*

Author to whom correspondence should be addressed.

J. Imaging 2025, 11(8), 252; https://doi.org/10.3390/jimaging11080252 (registering DOI)

Submission received: 16 June 2025 / Revised: 16 July 2025 / Accepted: 24 July 2025 / Published: 26 July 2025

(This article belongs to the Special Issue Celebrating the 10th Anniversary of the Journal of Imaging)

Download

Browse Figure

Versions Notes

Abstract

Generative AI (genAI) has emerged as a powerful tool for synthesizing diverse and complex image data, offering new possibilities for scientific imaging applications. This review presents a comprehensive comparative analysis of leading generative architectures, ranging from Variational Autoencoders (VAEs) to Generative Adversarial Networks (GANs) on through to Diffusion Models, in the context of scientific image synthesis. We examine each model’s foundational principles, recent architectural advancements, and practical trade-offs. Our evaluation, conducted on domain-specific datasets including microCT scans of rocks and composite fibers, as well as high-resolution images of plant roots, integrates both quantitative metrics (SSIM, LPIPS, FID, CLIPScore) and expert-driven qualitative assessments. Results show that GANs, particularly StyleGAN, produce images with high perceptual quality and structural coherence. Diffusion-based models for inpainting and image variation, such as DALL-E 2, delivered high realism and semantic alignment but generally struggled in balancing visual fidelity with scientific accuracy. Importantly, our findings reveal limitations of standard quantitative metrics in capturing scientific relevance, underscoring the need for domain-expert validation. We conclude by discussing key challenges such as model interpretability, computational cost, and verification protocols, and discuss future directions where generative AI can drive innovation in data augmentation, simulation, and hypothesis generation in scientific research.

Keywords: image generation; generative AI; Generative Adversarial Networks; diffusion; synthetic data

Graphical Abstract

Share and Cite

MDPI and ACS Style

Sordo, Z.; Chagnon, E.; Hu, Z.; Donatelli, J.J.; Andeer, P.; Nico, P.S.; Northen, T.; Ushizima, D. Synthetic Scientific Image Generation with VAE, GAN, and Diffusion Model Architectures. J. Imaging 2025, 11, 252. https://doi.org/10.3390/jimaging11080252

AMA Style

Sordo Z, Chagnon E, Hu Z, Donatelli JJ, Andeer P, Nico PS, Northen T, Ushizima D. Synthetic Scientific Image Generation with VAE, GAN, and Diffusion Model Architectures. Journal of Imaging. 2025; 11(8):252. https://doi.org/10.3390/jimaging11080252

Chicago/Turabian Style

Sordo, Zineb, Eric Chagnon, Zixi Hu, Jeffrey J. Donatelli, Peter Andeer, Peter S. Nico, Trent Northen, and Daniela Ushizima. 2025. "Synthetic Scientific Image Generation with VAE, GAN, and Diffusion Model Architectures" Journal of Imaging 11, no. 8: 252. https://doi.org/10.3390/jimaging11080252

APA Style

Sordo, Z., Chagnon, E., Hu, Z., Donatelli, J. J., Andeer, P., Nico, P. S., Northen, T., & Ushizima, D. (2025). Synthetic Scientific Image Generation with VAE, GAN, and Diffusion Model Architectures. Journal of Imaging, 11(8), 252. https://doi.org/10.3390/jimaging11080252

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Menu

Synthetic Scientific Image Generation with VAE, GAN, and Diffusion Model Architectures

Abstract

Share and Cite

Article Metrics

Article Access Statistics

Further Information

Guidelines

MDPI Initiatives

Follow MDPI