MicroICS: A New Computational Tool for Biofilm Image Analysis

• Post by: Jure Pohleven
• 3.November, 2026
• Comments off

Biofilms – highly organised microbial communities – are often the hidden cause of recurrent infections in humans and animals, as well as contamination in food and pharmaceutical production. As these communities grow into 3D structures, microscopy is becoming increasingly important for their investigation, while imaging data processing requires multidisciplinary expertise. At the IJS, through collaboration between three departments (Biotechnology – B3, Knowledge Technologies – E8 and Condensed Matter Physics – F5), we have developed a new tool for analysing biofilm images called MicroICS, currently the only tool that combines analysis of confocal fluorescence images of biofilms with machine learning. It consists of three modules: the first converts images into quantitative structural features; in the second, we train models to understand how these features relate to the biological context; and in the third, we use the trained models to predict and classify new images. The tool is designed to investigate links between the 3D structure of biofilms and their function, a field that is developing rapidly, driven by technological advances in analytical methods. It is open source and available to researchers for further development and expansion. We have published an article on its development and functionalities in the journal npj Biofilms and Microbiomes (https://www.nature.com/articles/s41522-026-01083-8).