Application of a Hybrid Fusion Classification Process for Identification of Microplastics Based on Fourier Transform Infrared Spectroscopy

Abstract

Microplastic research is an emerging field. Consistent accurate identification of microplastic polymer composition is vital for understanding the effect of microplastic pollution in the environment. Fourier transform infrared (FT-IR) spectroscopy is becoming commonplace for identifying microplastics. Conventional spectral identification is based on library searching, a process that utilizes a search algorithm against digital databases containing single spectra of pristine reference plastics. Several conditions on environmental microplastic particles such as weathering, additives, and residues cause spectral alterations relative to pristine reference library spectra. Thus, library searching is vulnerable to misidentification of microplastic samples. While a classification process (classifier) based on a collection of spectra can alleviate misidentification problems, optimization of each classifier (tuning parameter) is required. Additionally, erratic results relative to the particular optimized tuning parameter can occur when microplastic samples originate from new environmental or biological conditions than those defining the class. Presented in this study is a process that utilizes spectroscopic measurements in a hybrid fusion algorithm that depending on the user preference, simultaneously combines high-level fusion with low- and mid-level fusion based on an ensemble of non-optimized classifiers to assign microplastic samples into specific plastic categories (classes). The approach is demonstrated with 17 classifiers using FT-IR for binary classification of polyethylene terephthalate (PET) and high-density polyethylene (HDPE) microplastic samples from environmental sources. Other microplastic types are evaluated for non-class PET and HDPE membership. Results show that high accuracy, sensitivity, and specificity are obtained thereby reducing the risk of misidentifying microplastics.

First evidence of in vitro cytotoxic effects of marine microlitter on Merluccius merluccius and Mullus barbatus, two Mediterranean commercial fish species

“…Rios-Fuster et al, 2019;Savoca et al, 2019). Here we confirmed that both our target species ingest plastic materials that range in size from 250 to 3000 μm and are represented by polypropylene, high density polyethylene and polyamide items (Chabuka and Kalivas, 2020;Primpke et al, 2018). Benthic macrolitter in the area was recently described quali-quantitatively, and the most abundant categories were attributed to plastic (Mancini et al, 2021).…”

Distribution Patterns of Floating Microplastics in Open and Coastal Waters of the Eastern Mediterranean Sea (Ionian, Aegean, and Levantine Seas)

“…Cross-reference of spectra retrieved from the ATR-FTIR instrument was performed with the open source database Open Specy (www.openspecy.org; Cowger et al, 2021). The Open Specy tool includes 636 spectra of 276 materials from three libraries of pure polymers, and materials relevant to microparticles and fibers found in the environment (Primpke et al, 2018;Chabuka and Kalivas, 2020;Suja Sukumaran, Thermo Fisher Scientific).…”

Contributions of Fourier transform infrared spectroscopy in microplastic pollution research: A review

“…Moreover, training the classifier can increase the analysis speed substantially when dealing with large datasets of FTIR spectra. For example, automated identification methods were tested based on hierarchical cluster analysis (Primpke et al, 2018), shortwave infrared imaging (Schmidt et al, 2018), identification of the most relevant bands (Renner et al, 2017;Renner, Nellessen, et al, 2019), random decision forest method (Hufnagl et al, 2019), modified chemometric identification concept (Renner, Sauerbier, et al, 2019), machine learning method (Kedzierski et al, 2019), Python based lFTIR mapping (Renner et al, 2020) and Hybrid fusion method (Chabuka & Kalivas, 2020). The analysis of FTIR spectra is time-consuming as often it is needed to compare the spectra one by one with the reference spectra.…”