Skip to main content
2023 BMES Annual Meeting Main banner
Icon Legend
This session is not in your schedule.
This session is in your schedule. Click again to remove it.
Back

    Biomedical Imaging and Instrumentation

    (E-194) Hyperspectral Imaging Technique to Identify the Composition of Bacteria in Marine Biofilms

    Friday, October 13, 2023
    9:30 AM – 10:30 AM PDT
    Location: Exhibit Hall - Row E - Poster # 194

      Presenting Author(s)

    • Shantanu M. Kore

      Graduate Student
      Clemson University
      Clemson, South Carolina, United States

      • Co-Author(s)

    • RF

      Reece Fratus

      PhD Student
      Clemson University
      Clemson, South Carolina, United States

    • BV

      Bruce Vogelsberg

      Graduate Student
      Clemson University, United States

    • LS

      Lucas Schmidt

      Lecturer
      Clemson University, United States

    • DK

      David Karig

      Associate Professor
      Clemson University, United States

      • Primary Investigator(s)

    • BG

      Bruce Z. Gao

      Professor
      Clemson University, United States

    Introduction:: A combination of different bacterial strains can be used to form marine biofilms. The thickness, roughness, and composition can vary depending on the environmental conditions. Therefore, the distribution of the different bacteria strains needs to be tracked. Hyperspectral imaging is a non-destructive technique that can collect spectral data from a sample at a high spatial resolution. This technique can be used to detect the location and identify the strain of bacteria within the biofilm.


    Materials and Methods:: A free space hyperspectral setup was used to image the lab-built biofilms, and the hyperspectral data was collected. The imaging system developed consisted of a supercontinuum light source that emits a broad range of wavelengths, a diffraction grating of 1800 lines/mm that disperses the light reflected by the sample into a spectrum of wavelengths, and a 1024x1024 pixel CMOS camera that captures the spectrum. An algorithm was developed to process the data to generate datasets for identifying the different bacterial strains in the sample.
    The biofilms were formed on a glass substrate within a flow chamber filled with low-nutrient media. The hyperspectral data were captured from a wavelength range of 600-900nm to identify the different bacterial strains in the biofilm under various flow and nutrient conditions.


    Results, Conclusions, and Discussions:: For different bacterial strains, a variation in intensity at around 830nm was observed for each bacterium sample. It was also confirmed that certain wavelengths had similar intensities for different samples of the same strains of bacteria. The spectral intensity curve associated with each wavelength for the bacteria strains had small differences that were used to label the bacterial strain. The hyperspectral imaging system was able to show bacterial development in the biofilms as well. For example, it is shown that the composition of biofilms that are formed on glass surfaces can be affected by changes in the flow and nutrient content in the water.

    The hyperspectral datacube can identify the bacteria and their position using the spectral fingerprint in combination with the spatial information. A database was built for storing the spectral information of different bacterial strains. In the future, this database can train machine-learning algorithms to automate the identification of bacterial strains in marine biofilms. The results demonstrate that hyperspectral imaging is a promising technique for identifying the composition of bacteria in marine biofilms.




    Acknowledgements (Optional): :

    References (Optional): :