Deep generative modeling of annotated bacterial biofilm images

Deep generative modeling of annotated bacterial biofilm images Full article

Journal

npj Biofilms and Microbiomes
ISSN: 2055-5008

Output data

Year: 2025, Volume: 11, Number: 1, Article number : 16, Pages count : DOI: 10.1038/s41522-025-00647-4

Authors

Holicheva Angelina A. ¹ , Kozlov Konstantin S. ² , Boiko Daniil A. ² , Kamanin Maxim S. ¹ , Provotorova Daria V. ^1,2 , Kolomoets Nikita I. ² , Ananikov Valentine P. ^2,3

Affiliations

1	Tula State University, Lenin pr. 92, Tula, 300012, Russia
2	Zelinsky Institute of Organic Chemistry, Russian Academy of Sciences, Leninsky Prospekt 47, Moscow, 119991, Russia
3	Organic Chemistry Department, RUDN University, 6 Miklukho-Maklaya St, Moscow, 117198, Russia

Biofilms are critical for understanding environmental processes, developing biotechnology applications, and progressing in medical treatments of various infections. Nowadays, a key limiting factor for biofilm analysis is the difficulty in obtaining large datasets with fully annotated images. This study introduces a versatile approach for creating synthetic datasets of annotated biofilm images with employing deep generative modeling techniques, including VAEs, GANs, diffusion models, and CycleGAN. Synthetic datasets can significantly improve the training of computer vision models for automated biofilm analysis, as demonstrated with the application of Mask R-CNN detection model. The approach represents a key advance in the field of biofilm research, offering a scalable solution for generating high-quality training data and working with different strains of microorganisms at different stages of formation. Terabyte-scale datasets can be easily generated on personal computers. A web application is provided for the on-demand generation of biofilm images.

Holicheva A.A. , Kozlov K.S. , Boiko D.A. , Kamanin M.S. , Provotorova D.V. , Kolomoets N.I. , Ananikov V.P.
Deep generative modeling of annotated bacterial biofilm images
npj Biofilms and Microbiomes. 2025. V.11. N1. 16 . DOI: 10.1038/s41522-025-00647-4 WOS OpenAlex

Web of science:	WOS:001396252500001
OpenAlex:	W4406380883

Пока нет цитирований