Introducing Bactovision, our straightforward Jupyter widget that helps researchers to accelerate the analysis of bacterial growth data — photos of petri dishes with multiple bacterial colonies growing under different conditions and with controlled interactions. In studies such as examining the interactions between Pseudomonas aeruginosa and Staphylococcus aureus, having the right tools for the analysis of hundreds of images becomes a crucial part of the research. Bactovision offers a practical, user-friendly interface that streamlines the integration of computer vision into microbial research, facilitating quicker insights and supporting scientific discovery.
A screenshot of the widget — an interface for a set of computer vision tools including image segmentation, contrast correction, and tools for manual correction of the segmentation procedure. The results are saved in a format that reflects the grid structure of the colony positions, simplifying the further analysis.
This work is a collaboration between the ML Colab and Dr. Laura Camus from the University of Tübingen