Deep learning approach for automatic microplastics counting and classification

Lorenzo-Navarro, Javier; Castrillón-Santana, Modesto; Sánchez-Nielsen, Elena; Zarco, Borja; Herrera, Alicia; Martínez, Ico; Gómez, May

doi:10.1016/j.scitotenv.2020.142728

Cited by 61 publications

(22 citation statements)

References 30 publications

(42 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Deep learning is promising in plastic classification work, but not routinely used (Garcia-Garin et al, 2021). For MP, automated sampling remains challenging, but automation is advanced for extraction, analysis and identification (Primpke et al, 2017;da Silva et al, 2020;Lorenzo-Navarro et al, 2021). Regarding the monitoring of litter and MP in the Arctic, the future may bring some opportunities for satellite imagery, autonomous tools such as Autonomous Underwater Vehicles, wave gliders and drones.…”

Section: Opportunities Automationmentioning

confidence: 99%

Future monitoring of litter and microplastics in the Arctic—challenges, opportunities, and strategies

et al. 2023

View full text Add to dashboard Cite

The Arctic Monitoring and Assessment Programme (AMAP) has published a plan and guidelines for the monitoring of litter and microplastics (MP) in the Arctic. Here we look beyond suggestions for immediate monitoring and discuss challenges, opportunities and future strategies in the long-term monitoring of litter and MP in the Arctic. Challenges are related to environmental conditions, lack of harmonization and standardization of measurements, and long-term coordinated and harmonized data storage. Furthermore, major knowledge gaps exist with regard to benchmark levels, transport, sources and effects, which should be considered in future monitoring strategies. Their development could build on the existing infrastructure and networks established in other monitoring initiatives in the Arctic, while taking into account specific requirements for litter and MP monitoring. Knowledge existing in northern and Indigenous communities, as well as their research priorities, should be integrated into collaborative approaches. The monitoring plan for litter and MP in the Arctic allows for an ecosystem-based approach, which will improve the understanding of linkages between environmental media of the Arctic, as well as links to the global problem of litter and MP pollution.

show abstract

Section: Opportunities Automationmentioning

confidence: 99%

Future monitoring of litter and microplastics in the Arctic—challenges, opportunities, and strategies

et al. 2023

View full text Add to dashboard Cite

show abstract

“…A N U S C R I P T 14 approaches have been developed in identifying and counting plastic particles (Chaczko et al, 2018;Lorenzo-Navarro et al, 2020;Bertoldi et al, 2021;Lorenzo-Navarro et al, 2021). Lorenzo-Navarro et al ( 2020) developed a software that could characterize the shape and count plastic particles at a rate twice as fast as a human expert.…”

Section: A C C E P T E D Mmentioning

confidence: 99%

The Current Status of Atmospheric Micro/Nanoplastics Research: Characterization, Analytical Methods, Fate, and Human Health Risk

Kung

Cheruiyot

et al. 2023

Aerosol Air Qual. Res.

View full text Add to dashboard Cite

Atmospheric plastic debris (microplastic and nanoplastic) research is comparatively little than in aquatic and terrestrial environments. Nonetheless, the research is important in understanding the risk and fate of these contaminants in the total environment. Generally, the research is limited by a lack of consensus on the characterization of plastic debris and the standardization of sampling and analysis A C C E P T E D MA N U S C R I P T 2 protocols. These limitations make it difficult to compare results from studies. In response, criteria for defining plastic debris beyond size characterization have been proposed to include polymeric composition, solubility, physical state, shape, color, and origin. There are also emerging techniques, such as Py-GC/MS, which can measure smaller particles in the nanoscale range, and TGA-FTIR-GC/MS, which can accurately identify more polymers. The identification of microplastics and nanoplastics sources and formation processes is challenging. Since most polymers are inert, the adverse health risks include endocytosis and accumulation in the liver and spleen. However, most of the toxic effects of these contaminants are related to surface-bound compounds, including heavy metals and persistent organic pollutants. Some polymers, such as expanded polystyrene, decompose to form carcinogens. This paper offers an overview of the current knowledge on plastic debris in the atmosphere and will be useful to researchers interested in this field.

show abstract

“…Following sample digestion, sample filtering and imaging have been successfully combined with classification methods to understand MP characteristics (such as polymer type, MP fragment or fibre, and damage). These consist of effectively filtering MPs from environmental samples and applying deep learning programs to count and analyze the plastics (Lorenzo-Navarro et al, 2020; Lorenzo-Navarro et al, 2021). These methods work well when other materials in the sample (other than the plastics) can be removed.…”

Section: Introductionmentioning

confidence: 99%

Characterizing microplastic ingestion, transformation, and excretion in insects using fluorescent plastics

Ritchie

Cheslock

Bourdages

et al. 2022

Preprint

View full text Add to dashboard Cite

Plastic pollution is a growing threat to our natural environment. Plastic waste/pollution results from high emissions of both macro (> 5 mm) and microplastics (MPs; < 5 mm) as well as environmental fractioning of macroplastics into microplastics. Microplastics have been shown to have a range of negative impacts on biota. Harmonized methods to accurately measure and count MPs from animal samples are limited, but what methods exist are not ideal for a controlled laboratory environment where plastic ingestion, transformation, and elimination can be quantified and related to molecular, physiological, and organismal traits. Here we propose a complete method for isolating and characterizing fluorescent MPs by combining several previously reported approaches into one comprehensive workflow. We combine tissue dissection, organic material digestion, sample filtering, and automated imaging techniques to show how fluorescently-labelled MPs provided to animals (e.g. in their diet) in a laboratory setting can be isolated, identified, and quantified. As a proof of concept, we fed crickets (Gryllodes sigillatus) a diet of 2.5% (w/w) fluorescently-labelled plastics and isolated and characterized plastic particles within the gut and frass.

show abstract

Deep learning approach for automatic microplastics counting and classification

Cited by 61 publications

References 30 publications

Future monitoring of litter and microplastics in the Arctic—challenges, opportunities, and strategies

Future monitoring of litter and microplastics in the Arctic—challenges, opportunities, and strategies

The Current Status of Atmospheric Micro/Nanoplastics Research: Characterization, Analytical Methods, Fate, and Human Health Risk

Characterizing microplastic ingestion, transformation, and excretion in insects using fluorescent plastics

Contact Info

Product

Resources

About