We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Advancing Microplastic Monitoring: Automatic Correction of the Aggregation and Discontinuity Issues Based on Instrument Imaging
AI summary Read the abstract
Researchers developed an automated microplastic characterization method called the diffluent amodal instance segmentation former (DAISF) model, trained on a manually labeled dataset of 130,536 particles, which substantially improved detection accuracy for both fiber and non-fiber microplastics by correcting aggregation and discontinuity errors common in instrumental imaging.
Instrumental imaging accelerates the analysis of microplastics but suffers from reduced detection accuracy during the segmentation of fibers and nonfibers due to particle aggregation and discontinuities. Therefore, this study aimed to develop an automated analytical method to characterize environmental microplastics based on instrumental imaging. By leveraging a manually labeled data set (130,536 particles), our established diffluent amodal instance segmentation former (DAISF) model greatly improved the ability to correct the aggregation and discontinuity issues due to the use of the Gauss-Laplace operator, which has superior segmentation performance. Compared to the instrument detection, this model significantly improved the detection of aggregated fibers and nonfibers by 71.8 ± 19.5% and 89.2 ± 24.1%, respectively, and of discontinuous fibers and nonfibers by 90.2 ± 14.7% and 98.4 ± 4.4%, respectively. The proposed computational method demonstrated superior performance compared to the instrument-based approach, achieving significantly higher recall and F1 scores. Quantitative validation revealed exceptional alignment with ground-truth measurements, exhibiting low relative errors in particle number (≤19.1%), length (≤20.2%), and mass (≤12.4%), representing improvements over the instrumental approach of 31.0-, 3.1-, and 8.8-fold, respectively. Overall, the established approach can accurately obtain microplastic concentrations and multiparameters based on instrumental imaging, indicating its usefulness in the efficient detection and rapid monitoring of environmental microplastics.
More Papers Like This
Advancing MicroplasticMonitoring: Automatic Correctionof the Aggregation and Discontinuity Issues Based on Instrument Imaging
AI summary Read the abstract
Researchers developed an automated microplastic analysis method using the diffluent amodal instance segmentation former (DAISF) model trained on over 130,000 labeled particles, which uses Gaussian-Laplacian techniques to automatically correct aggregation and discontinuity errors in instrumental imaging and substantially improve segmentation accuracy for microplastic characterization.
Microplastic Binary Segmentation with Resolution Fusion and Large Convolution Kernels
AI summary Read the abstract
Researchers developed an improved machine-learning model to automatically detect and segment microplastic particles in images, achieving better accuracy than previous approaches by combining multi-resolution image analysis with large convolution kernels. Reliable automated detection tools are essential for scaling up microplastic monitoring, since manual identification is too slow and inconsistent for the volumes of environmental samples that need to be processed.
A Deep Learning Approach for Microplastic Segmentation in Microscopic Images
AI summary Read the abstract
Researchers developed a deep learning model for automated segmentation and classification of microplastics in microscopic images, identifying five distinct categories including fibers, fragments, spheres, foam, and film. The model achieved high accuracy while maintaining low computational requirements, making it suitable for high-throughput deployment in environmental monitoring. The study offers a tool that could help overcome the measurement bottleneck in microplastic characterization for toxicological and risk assessment studies.
Detection of Microplastics Using Machine Learning
AI summary Read the abstract
Researchers reviewed and demonstrated machine learning approaches for detecting and classifying microplastics in environmental samples, finding that automated image analysis and spectral classification methods can improve the speed and accuracy of microplastic monitoring compared to manual methods.
Instance Segmentation for the Quantification of Microplastic Fiber Images
AI summary Read the abstract
Researchers applied deep learning instance segmentation to automatically count and measure microplastic fibers in microscope images, replacing tedious manual analysis. The automated method achieved high accuracy and could significantly accelerate microplastic quantification workflows in research and monitoring programs.
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.