We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Numerical clustering model generation for the identification of microplastics collected in Parques Nacionales Naturales Corales de Profundidad y Corales del Rosario y San Bernardo, using Fourier Transform Infrared Spectroscopy with Attenuated Total Reflection (FTIR-ATR)
Summary
This Colombian study developed machine learning clustering models to classify microplastics collected from Colombian marine protected areas using FTIR-ATR spectroscopy. Automated identification of plastic types in environmental samples helps researchers efficiently characterize microplastic pollution in sensitive coral reef ecosystems.
Se proponen dos modelos de agrupamiento para elementos sólidos, con tamaño menor a 5 mm, en muestras de agua de mar, colectadas durante la temporada seca (Diciembre 2021 - Abril 2022) en doce estaciones de monitoreo localizadas en los Parques Nacionales Naturales Corales de Profundidad y Corales del Rosario y San Bernardo, a partir de sus espectros de absorbancia en el infrarrojo medio, con número de onda en el rango 4000 cm^-1 – 500 cm^-1, usando FTIR-ATR. A partir de 818 espectros FTIR-ATR, previamente normalizados y expresados en sus primeras siete (7) componentes principales (PCA por sus siglas en inglés, Principal Component Analysis), se implementó un modelo de cúmulos mediante el algoritmo k-means, el cual sugirió cinco (5) grupos, atendiendo a indicadores de validación intrínsecos (Dunn, Davies-Bouldin y Silueta). De manera similar, se desarrolló un dendrograma, que sintetiza el agrupamiento jerárquico. Los espectros promedios de los grupos sugeridos por los modelos se comparan con espectros de referencia de la base de datos comercial (KnowItAll, Bio-Rad/Wiley). La comparación con la base de datos permitió identificar minerales típicos de la descomposición de los corales y algunos polímeros, a saber: polietileno, poliéster, polietileno tereftalato (PET), polipropileno. Dado que el tamaño de la muestra es menor a 5mm, se trata de microplásticos. Así, este estudio evidencia la presencia de estos elementos en estos sensibles ecosistemas.
Sign in to start a discussion.
More Papers Like This
An ensemble machine learning method for microplastics identification with FTIR spectrum
Researchers developed an ensemble machine learning method to automatically identify microplastics using Fourier transform infrared (FTIR) spectroscopy data. The approach combines multiple classification algorithms to improve accuracy over individual methods for detecting and categorizing microplastic particles. The study suggests this automated approach could help standardize and accelerate microplastic monitoring in marine environments.
A machine learning algorithm for high throughput identification of FTIR spectra: Application on microplastics collected in the Mediterranean Sea
Researchers developed a machine learning method to automatically identify the chemical composition of microplastics from FTIR spectroscopy data collected during the Tara Mediterranean expedition. The algorithm performed well for common polymers like polyethylene and was applied to classify over 4,000 unidentified microplastic spectra. The study demonstrates that automated identification tools can significantly speed up large-scale microplastic pollution surveys while maintaining acceptable accuracy.
A Comparative Study of Machine Learning and Deep Learning Models for Microplastic Classification using FTIR Spectra
Researchers compared machine learning and deep learning models for classifying microplastics using FTIR spectra, evaluating multiple algorithmic approaches against standardised spectral datasets. The study assessed classification accuracy and computational efficiency, identifying which model architectures best discriminate between polymer types in environmental microplastic samples.
Robust Automatic Identification of Microplastics in Environmental Samples Using FTIR Microscopy
Researchers developed a robust automated method for identifying microplastics in environmental samples using FTIR microscopy combined with machine learning-based spectral matching, improving the consistency and efficiency of microplastic identification compared to manual evaluation.
Machine learning-driven microplastics identification using ensemble stacking with Extra Tree meta-models from FTIR data
Researchers applied ensemble stacking machine learning to ATR-FTIR spectra for microplastic identification, finding that combining multiple classifier outputs improved polymer classification accuracy beyond any single model, particularly for weathered plastics with degraded spectral signatures.