PlanktonScope: Affordable modular imaging platform for citizen oceanography

Background

   The planktonic communities within the oceans represent one of the most diverse and understudied ecosystems on the planet. Despite their importance, little is known about the diversity in these communities, or the scope that human activities are having on them, including urban, agricultural, industrial and nuclear waste, and acoustic perturbations. Despite efforts to understand and catalogue them, many species remain unknown, or are very poorly characterized. This is largely because monitoring such a vast number of species under such varied conditions, and over spatial scales that encompass entire oceans, remains difficult. Consequently, most of the ever-evolving oceanic habitats remain largely undersampled. In their work, Pollina et al built a modular imaging platform aiming at enabling citizens-based oceanographic research.

Key findings and developments

    Pollina et al present a modular imaging platform which can be adapted for various field applications, including, but not limited to, oceanography. The modular nature of the device allows adaptability for diverse applications; easy upgrades or replacements as required; and integration of novel applications in the future, as needed (Figure 1).

 

The authors present a flow-based imaging platform in two configurations: a) a modular, compartmentalized configuration enabling multi-functionality and adaptability and b) a compact version allowing robust performance and portability. The platform is built around an open-hardware principle, with off-the-shelf components, aimed at improving accessibility. The compartmentalized version is based on 6 units: a  single board computer coupled to a camera sensor; two reversed M12 lenses separated into 2 modules; a motorized stage and focus delta platform for sample manipulations; a module embedding independent programmable rings of LEDs for the illumination module; and a laser cut peristaltic pump. The other prototype shows the integration of those modules for specialized applications for oceanographic studies.

  Both designs are parametric, enabling the use of different thicknesses of the chosen material (i.e. acrylic, recycled plastic,  wood, metal, fiberboard, etc.). This allows rapid design iteration and enables a precise, low-cost method for aligning and spacing optical components. For coupling the units, three neodymium magnets are incorporated into the corners of the interface between modules. Moreover, each of the modules can be readily 3D-printed. The microscope can be used in vertical or horizontal configurations, and placed upright or inverted as required.

  The flow-through version simplifies the assembly of the modular prototype by using a minimal structure to position and align the components. A level of modularity is maintained by allowing the lenses and flow-cell to be quickly swapped. Both prototypes are based on a Raspberry Pi single board computer. This enables a centralized method to control the electronics, acquire and process the images, and serve as the user/machine interface. GPS for geolocalization in oceanographic studies can be incorporated too. Moreover, the use of Raspberry Pi and its community reach, allows for multiple other possibilities for new modules and more functions built on top of this platform. Both instruments can be powered either through standard wall AC power or from battery cells for field use. The fluidic strategy used includes a continuous flow mode and a stop flow mode. The stop-flow method synchronizes the rotation of the pump to the image capture in order to stop the pump when each image is taken. Therefore, with stop flow, quantitative analysis and high resolution are possible. The software architecture is based on existing programs and python libraries, including tools for classification and annotation. The workflow presented in this work allows image processing to be done automatically. This workflow includes background subtraction, thresholding, and labeling. To reduce storage requirements, large datasets can be compressed on-board by storing only relevant regions of interest. MorphoCut extracts various features of each region of interest, and generated a file ready to be uploaded to EcoTaxa.

As proof of principle, the authors used a continual flow mode to image monocultures of marine planktons Pyrocystis noctiluca, and Coscinodiscus wailesii. Identical samples were analysed in the Planktonscope and a FlowCam to create a quantitative comparison of the instruments. Both achieved enough resolution to allow a classification down to a genus or evenspecies level. Also as proof of principle, the authors aimed to answer an ecological question, namely how salinity in the Comau Fjord in southern Chile, affects the local planktonic community.

   Altogether, the presented versions of PlanktonScope are capable of autonomously imaging 1.7 ml per minute with a 1.5 μm resolution, and are built with under $400 in parts. The authors ultimately conclude that deploying a high-throughput microscope platform on a global scale will bring light to the habitats under-surveyed by research cruises. They hope that the possibility to connect platforms such as Planktonscope, with a network of climate-researchers, ecologists, citizen scientists, and others, worldwide, will allow for integrative research.

What I like about this preprint

I like work that is curiosity-driven and aims at making research and science, accessible to all. I think this work does this perfectly. Moreover, it successfully identified an important gap in current research possibilities in oceanographic enterprises, and addressed this with an imaging platform with enormous potential. The work also includes clear information on the prototypes, allowing reproducibility, and encourages also contributions from others – citizen science- to adapt the PlanktonScope to the research questions and needs as necessary. I also chose this preprint because biologically it indeed addresses an important gap in our knowledge on planktonic communities, and places them in a relevant picture of current concerns in terms of climate health.

Open questions

1. This is an exciting tool with enormous potential! What would you say are important limitations of the setups, or possible complications which you think you might encounter during oceanographic surveys, and how can they best be addressed?

2. In addition to characterization of the planktonic communities, how do you integrate measurements of parameters such as salinity or contaminants in water affecting such communities? Can a module for diverse measurements be incorporated too? Is this done independently?

3. As for Foldoscope, since PlanktonScope is aimed at citizen-science, do you intend to conduct similar training workshops for people who are interested in using this platform, as well as those who might help integrating information obtained from it, with other parameters (such as pollutants, etc)? This would potentially extend its use, and allow for more applications and research questions to be addressed, as has been the case for Foldoscope.

References

• Pollina T., Larson A.G., Lombard F., Li H., Colin S., de Vargas C., Prakash M., PlanktonScope: Affordable modular imaging platform for citizen oceanography, bioRxiv, 2020.

 

 

doi: https://doi.org/10.1242/prelights.21065

Read preprint (No Ratings Yet)