Microplastic and nanoplastic analysis methods, tests and reference materials

The quantification of microplastic debris is critical to understanding how plastic litter accumulates across the world's oceans and identifying hotspots that should be targeted for early cleanup efforts. Currently, the most common monitoring method to quantify microplastic requires physical sampling followed by manual counting and laboratory analysis. The need for manual counting is time-consuming, operator-dependent, and incurs high costs, thereby preventing scalable deployment of consistent marine plastic monitoring worldwide. Here, we describe a workflow combining a simple experimental setup with advanced image processing techniques to conduct quantitative and qualitative assessments of microplastic (0.05 cm ¡ particle size ¡ 0.5 cm). The image processing relies on deep learning models designed for image segmentation and classification trained on 4795 particles based on four categories (hard fragments, lines, pellets, and foams). For each particle, multiple morphological and geometrical characteristics were inferred by the workflow after conducting the image analysis such as the color, the length of the minimum and maximum axis, the surface, the perimeter, and the orientation. We evaluated our workflow by randomly partitioning our data into training and test samples. We employed four validation datasets that had been manually annotated using two techniques and repositioned multiple times. The results demonstrated comparable or superior performance in comparison to manual identification. Thus, the use of the model offers an efficient, more robust, and less labor-intensive alternative to particle counting in addition to the relative simplicity of the network architecture used that made it easy to train. Furthermore, besides the efficient particle quantification, ongoing efforts are directed towards improving and developing the model's capabilities to classify the annotated particles – presenting encouraging and promising prospects for better-standardized reporting of plastic particles in the environment. We also made the models and datasets open-source and created a user-friendly web interface for directly annotating new images. Also see: https://micro2024.sciencesconf.org/557208/document