Predicting Soil Microplastic Occurrence Using a Graph Convolutional Network with Kolmogorov–Arnold Activation

Deep Learning Approaches for Microplastic Identification in Microscopic Water Samples

Land use shapes the fate of soil microplastics in China: Insights from machine learning

AI summary · Read the abstract

A massive study analyzing nearly 1,500 soil samples across China found that farmland, especially orchards and fields covered with plastic sheeting (mulch), had far more tiny plastic particles than forests or natural areas, with mulched fields showing about 6 times more contamination. Since these are the soils growing our food, this matters because it suggests the crops we eat may be absorbing more microplastics from heavily farmed, plastic-mulched land, highlighting a need for smarter farming practices to reduce this exposure.

Machine Learning Approaches for Predicting Microplastic Removal

AI summary · Read the abstract

Researchers applied Bayesian optimization combined with machine learning models — including boosted regression trees, neural networks, and support vector regression — to predict microplastic removal during coagulation, finding the BOA-BRT hybrid outperformed conventional methods by up to 71%, with microplastic particle size identified as the most influential variable.

Advanced Microplastic Identification in Marine Environments via Hybrid Deep Learning

AI summary · Read the abstract

Researchers propose a hybrid deep learning architecture combining 3D convolutional neural networks and Vision Transformers applied to hyperspectral imagery to detect and classify microplastics in turbid marine environments, capturing both local spectral signatures and global contextual patterns that single-model approaches miss.

Defining Irregular Microplastics: A Machine Learning Approach for Morphometric Characterization

AI summary · Read the abstract

Researchers applied a machine learning decision-tree classifier to hyperspectral images of 129 nano- and microplastic particles and identified five morphometric descriptors—circularity, roundness, solidity, long diameter, and perimeter-area ratio—that together distinguish irregular particles from spherical and fibrous forms with 96% accuracy, providing a standardized definition for the most environmentally prevalent but least-defined shape class.