Image
Dorthe Wildenschild
Johanna Carson
Dorthe Wildenschild uses deep learning techniques to enhance the speed and accuracy of the imaging microbial biofilms.

Researchers are using deep learning and advanced imaging to visualize biofilm growth

Key Takeaways

Biofilms play a crucial role in the health of ecosystems by cleaning up pollution in soil and water.

Research seeks to answer questions about how biofilms respond to environmental stresses.

This research uses deep learning to enhance 3D imaging of biofilms, and minimize exposure to damaging X-rays.

Environmental engineering researchers at Oregon State University are using cutting-edge technology to better understand biofilms – communities of microorganisms that play a vital role in the environment. This research, led by Dorthe Wildenschild, professor and DeVaan Chair and Executive Director for Clean Water Technology, is funded by a $1.8 million grant from the U.S. Department of Energy’s Bioimaging Research Program.

Understanding biofilms

Biofilms are groups of microorganisms, such as bacteria, that stick together and create protective communities. While problematic when they grow on medical implants or in drains, biofilms are also crucial for cleaning up pollution in soil and water. Understanding how biofilms function is key to harnessing their beneficial properties and mitigating their negative impacts.

Using advanced imaging and deep learning

Studying biofilms in their natural environments, such as soil or rocks, is difficult because these environments are complex and opaque. To overcome these challenges, the research team will use advanced imaging techniques, including high-resolution X-ray imaging (micro-CT) and special contrast agents. These techniques will allow the researchers to create detailed 3D images of biofilms growing in materials like sand and beads, providing valuable insights into their structure and growth.

Image
Glass beads covered with microbial biofilm
Manipulating genes in S. oneidensis (shown here in purple growing on glass beads) reveals the role of biofilm movement and communication in survival.
Photo by: Ostvar S. et al. (2018) Advances in Water Resources

"The thrust of our research is looking for answers to some fundamental questions around biofilms and their ability to respond intelligently to stress,” Wildenschild said. “But, in order to do that, we have to invent a whole new imaging technology that will allow us to image 3D biofilms as they grow. So, we need to reduce the amount of radiation that the bacteria are exposed to. We’re going to do that by using the latest in deep learning models.”

The novel aspect of this project is the use of deep learning techniques, such as convolutional neural networks and generative adversarial networks, to enhance the speed and accuracy of the imaging process. These methods will help the researchers minimize radiation exposure to prevent damage to the biofilms, and also to more effectively and reliably differentiate between different materials (biofilm, air, water, and solid).

The impact of biofilm research

The goal of the research is to investigate how biofilms respond to stress, such as lack of water. By manipulating specific genetic traits in a type of bacteria called S. oneidensis, the researchers will observe the role of biofilm movement and communication in the survival of the biofilm community. The research is expected to provide crucial insights into the behavior of biofilms, which will have implications for various fields, including bioremediation, bioenergy, and water treatment.

Partner institutions are the University of Virginia and the University of New South Wales in Sydney, Australia. A former student of Wildenschild, Ryan Armstrong, B.S. bioengineering ’06, Ph.D. environmental engineering ’12, is a key collaborator. Armstrong, now a professor at UNSW, brings expertise in applying deep learning to advanced imaging techniques.

Dec. 5, 2024

Related Researchers

Related Stories