We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Quantitative and Qualitative Evaluation of Microplastic Contamination of Shrimp Using Visible Near-Infrared Multispectral Imaging Technology Combined with Supervised Self-Organizing Map
AI summary Read the abstract
Scientists developed a non-destructive imaging method using visible and near-infrared multispectral technology combined with machine learning to detect and identify microplastics in shrimp. The system could distinguish between four common microplastic types (PET, PE, PP, and PS) on both minced shrimp and shrimp shell surfaces. This approach offers a faster alternative to traditional microplastic detection methods for screening seafood contamination.
Microplastic (MP) contamination is a growing environmental concern with significant impacts on ecosystems, the economy, and potentially human health. However, accurately detecting and characterizing MPs in biological samples remains a challenge due to the complexity of biological matrices and analytical limitations. This study presents a novel, non-destructive visible near-infrared multispectral imaging (Vis-NIR-MSI) method combined with a supervised self-organizing map (SOM) to enable rapid qualitative and quantitative analysis of MPs in seafood. We specifically aimed to identify and differentiate four types of microplastics, namely PET, PE, PP, and PS, in the range 1–4 mm, present on the surface of minced shrimp and shrimp shell. For quantification, MPs were incorporated into minced shrimp surface at concentrations ranging from 0.04% to 1% w/w. The modified model achieved a high coefficient of determination (R2 > 0.99) for PE and PP quantification. Unlike conventional techniques, this approach eliminates the need for pre-sorting or chemical processing, offering a cost-effective and efficient solution for large-scale monitoring of MPs in seafood, with potential applications in food safety and environmental protection.
More Papers Like This
Assessment of microplastic contamination in shrimp utilizing multispectral imaging, fluorescence, and infrared spectroscopy
AI summary Read the abstract
This study evaluated mid-infrared spectroscopy, fluorescence spectroscopy, and multispectral imaging for detecting and classifying microplastic contamination in minced shrimp, finding all three techniques capable of predicting contamination levels of polystyrene, polypropylene, polyethylene, and PET.
In‐Situ Detection of Microplastic Particles on Food Using Hyperspectral Imaging With One‐Dimensional Convolutional Neural Network and Artificial Neural Network
AI summary Read the abstract
Scientists trained AI models on hyperspectral camera images to detect microplastic particles directly on the surface of raw seafood (tilapia) without needing to isolate or remove the particles first, achieving a 96.3% detection score for 600-micron particles. Current food safety testing for microplastics requires laborious physical separation, making routine screening impractical; this approach could enable rapid, non-destructive screening in food processing facilities. The method represents a practical step toward monitoring microplastic contamination in seafood before it reaches consumers.
Identification and visualization of environmental microplastics by Raman imaging based on hyperspectral unmixing coupled machine learning
AI summary Read the abstract
Researchers developed a new method combining Raman imaging with machine learning to identify and visualize microplastics in environmental samples without destroying them. The technique can distinguish between different polymer types and map their distribution within a sample. The study offers a faster, more accurate approach to microplastic detection that could improve environmental monitoring efforts.
Development of a Near-Infrared Imaging System for Identifying Microplastics in Water
AI summary Read the abstract
Researchers developed a near-infrared imaging system capable of automatically identifying and characterizing microplastics suspended in water, successfully obtaining material identification images without the manual sorting typically required by conventional methods.
Hyperspectral imaging and self-organizing map approach for non-destructive monitoring of microplastic contamination in sandy substrates
AI summary Read the abstract
Scientists developed a camera-based scanning technique that can spot and measure tiny plastic fragments, including pieces smaller than 1mm, buried in sand, without destroying the samples. This matters because microplastics on beaches and in soil are notoriously hard to detect and measure accurately, and better detection tools are a key first step toward understanding how much of this pollution surrounds us and eventually ends up in our bodies and food chain.
Research digests by email
When a large batch of papers lands in the Atlas, we read through it and send a short write-up of what stood out.