| title | Automated Imaging and Artificial Intelligence (AI²) |
|---|
A variety of instruments designed for automated image analysis of phytoplankton and microzooplankton are deployed across Northern Europe for comprehensive monitoring purposes. Frequent sampling is possible, e.g. every 20 minutes. A typical sample volume is 5-10 mL. Below, we provide descriptions of some of these cutting-edge instruments.
Fig 1. Images of plankton captured using the Imaging FlowCytobot in the Skagerrak, Kattegat and the Baltic Sea. The IFCB was operated by SMHI on R/V Svea.
The Imaging FlowCytobot (IFCB) is an automated microscope integrated into a flow cytometer that generates plankton images. The IFCB can be operated in situ or in air pumping water to the instrument. Water samples are collected automatically about every 20 minutes. The plankton in the sample passes trough a flow chamber and are imaged by a camera triggered by fluorescence or scattering of the plankton organism. Thousands of images are collected for each sample. To identify the plankton in the images machine learning methods are used. Phytoplankton identifications specialists annotate images to create training sets. Nordic Microalgae provides links to annotated images from Northern Europe that can be used for training your image classifier, see Annotated images.
Learn more about the Imaging FlowCytobot: Woods Hole Oceanographic Institute web site about IFCB
Fig. 2. The Imaging Flow CytoBot (Photo: Bengt Karlson)
The Cytosense/Cytobuoy is a flow cytometer developed for analysing plankton samples. Scattering properties and fluorescence fingerprints are used to characterize organisms. A large flow cell in this instrument makes it possible to cover a larger size range of organisms compared to the other instruments. This makes it possible to quantify phototrophic picoplankton, nanoplankton and microplankton. The CytoSense/Cytobuoy produces images of only a subset of the cells passing through its flow cell.
Fig. 4. The CytoSense (Photo Bengt Karlson)
The FlowCam was developed for analysing plankton samples but can also be used to analyse various man made particles. The instrument is not a true flow cytometer since water is simply pushed through the flow cell with no sheath fluid. This sometimes results in out of focus images. A unique feature with the FlowCam 8400 is that it produces colour images of the organisms while the other two instruments mentioned above use grayscale cameras. In the FlowCam 8400 it is possible to change objectives with different magnifications. A high magnification results in images showing more detail but also in longer sample processing time
Fig. 5. The FlowCam Macro. (Photo Bengt Karlson)
The PlanktoScope, is a citizen science approach to automated imaging and identification of plankton. This is low-cost alternative to the commercial instruments. The PlanktoScope can be used with objectives with different magnifications. At present, the PlanktoScope (version 4) is somewhat limited in functionality compared to the commercial alternatives and it may only be applicable to multicellular zooplankton and to larger phytoplankton species at the moment. Future development may change the situation.
- Karlson, B., Berdalet, E., Kudela, R.M., 2022. The GlobalHAB mini-symposium on automated plankton observations. Harmful Algae News 71, 1-4. http://www.e-pages.dk/ku/1549/
- Karlson, B., 2021. A European Imaging FlowCytoBot (IFCB) network. Harmful Algae News 68, 1. http://www.e-pages.dk/ku/1516/
- Olson, R.J., Sosik, H.M., 2007. A submersible imaging-in-flow instrument to analyze nano-and microplankton: Imaging FlowCytobot. Limnology and Oceanography-Methods 5, 195-203. https://doi.org/10.4319/lom.2007.5.195
- Sieracki, C.K., Sieracki, M.E., Yentsch, C.S., 1998. An imaging-in-flow system for automated analysis of marine microplankton. Marine Ecology Progress Series, 285-296.
- Sosik, H.M., Olson, R.J., 2007. Automated taxonomic classification of phytoplankton sampled with imaging-in-flow cytometry. Limnology and Oceanography-Methods 5, 204-216. https://aslopubs.onlinelibrary.wiley.com/doi/pdf/10.4319/lom.2007.5.204
- Thyssen, M., Alvain, S., Lefèbvre, A., Dessailly, D., Rijkeboer, M., Guiselin, N., Creach, V., Artigas, L.-F., 2015. High-resolution analysis of a North Sea phytoplankton community structure based on in situ flow cytometry observations and potential implication for remote sensing. Biogeosciences 12(13), 4051-4066.
- Thyssen, M., Grégori, G., Créach, V., Lahbib, S., Dugenne, M., Aardema, H.M., Artigas, L.-F., Huang, B., Barani, A., Beaugeard, L., Bellaaj-Zouari, A., Beran, A., Casotti, R., Del Amo, Y., Denis, M., Dubelaar, G.B.J., Endres, S., Haraguchi, L., Karlson, B., Lambert, C., Louchart, A., Marie, D., Moncoiffé, G., Pecqueur, D., Ribalet, F., Rijkeboer, M., Silovic, T., Silva, R., Marro, S., Sosik, H.M., Sourisseau, M., Tarran, G., Van Oostende, N., Zhao, L., Zheng, S., 2022. Interoperable vocabulary for marine microbial flow cytometry. Frontiers in Marine Science 9. doi 10.3389/fmars.2022.975877