Spectral Classification of Large-Scale Blended (Micro)Plastics Using FT-IR Raw Spectra and Image-Based Machine Learning

Quantitative analysis of plastic blends based on virtual mid-infrared spectroscopy combined with chemometric methods

Developing efficient and accurate quantitative analysis methods for plastic blends holds significant value for resource recycling and environmental monitoring. Mid-infrared (MIR) spectroscopy, combined with chemometric techniques, has demonstrated excellent performance in plastic blend quantification. However, obtaining mid-infrared spectral data for a large number of plastic blends to calibrate the model remains challenging. This study proposes an innovative approach that utilizes pure plastic MIR spectra and the Beer-Lambert law to generate virtual plastic blend spectra. Four experimental groups (A-D) were designed, incorporating both real and virtual spectra to systematically evaluate the method's effectiveness. Experiment group A established a baseline model using real spectra, while experiment groups B, C, and D respectively validated the applicability and generalization capability of models based on virtual spectra, as well as their potential applications in MIR hyperspectral imaging (MIR-HSI), respectively. The study further explores feature band selection, model construction, evaluation, and interpretation. The results demonstrate that this method can efficiently predict the mass percentages of components in ternary plastic blends. In experimental group C, partial least squares regression (PLSR), one-dimensional convolutional neural network (CNN1D), and two-dimensional convolutional neural network based on Gramian Angular Field (GAF-CNN2D) models-trained on 208 virtual plastic blend spectra-were employed to predict the mass percentages of 66 ternary plastic blends composed of polyethylene (PE), polypropylene (PP), and polystyrene (PS). The prediction coefficients of determination (RT2) reached 0.9872, 0.9879, and 0.9944, respectively, indicating exceptional predictive accuracy. Experimental group D further demonstrated that, even under Gaussian noise interference and limited spectral range, the strategy of fusing mid-wave infrared and long-wave infrared bands allowed the PLSR and GAF-CNN2D models to maintain high performance in predicting the mass percentages of 66 ternary blends of PE, PP, and PS, with RT2 values of 0.9852 and 0.9895, respectively. This suggests that the proposed method holds potential for applications in MIR-HSI and is promising for real-time online analysis. Finally, this study proposes a more widely applicable and optimized quantitative analysis application scheme based on virtual plastic blend spectra, aiming to enable rapid and precise determination of unknown plastic blends.

Identification of marine microplastics by laser-induced fluorescence spectroscopy: 1-Dimensional convolutional neural network and continuous convolutional model

Marine microplastic pollution is a serious threat to ecosystems and human health, and its identification is of great significance for determining the source and extent of pollution. Conventional methods such as Fourier transform infrared (FTIR) spectroscopy and Raman spectroscopy are effective but they are time-consuming and their equipment is expensive. Laser induced fluorescence can reflect the molecular structure through the fluorescence characteristics of aromatic groups and hydrocarbon chains. This method has the characteristics of non-destructive, rapid and efficient, which can be used for the identification of microplastics. This study investigated 2400 LIF spectra of six types of marine microplastics excited by a 405 nm laser. A 1-dimensional convolutional neural network (1D-CNN) and an optimized continuous convolution (Cont-conv) model were used for classification. The accuracy of 1D-CNN is 97.5 %, demonstrating good performance, while the accuracy of the Cont-conv model can reach up to 99.5 %. The results show that the Cont-conv model effectively enhances the model's ability to extract features through continuous convolution operations and achieves faster convergence. CNN models trained on commercial microplastic samples were applied to the identification of field-collected marine microplastics, and also achieved good results. This study presents an innovative and efficient automated classification method for the detection of marine MPs, which offers the potential for integration with portable devices.

How small is big enough? Big data-driven machine learning predictions for a full-scale wastewater treatment plant

Wastewater treatment plants (WWTPs) generate vast amounts of water quality, operational, and biological data. The potential of these big data, particularly through machine learning (ML), to improve WWTP management is increasingly recognized. However, the costs associated with data collection and processing can rise sharply as datasets grow larger, and research on determining the optimal data volume for effective ML application remains limited. In this study, we comprehensively analyzed water quality, operational, and biological data collected from a full-scale WWTP over 970 days. Our results demonstrate that ML models can predict not only operational and water quality parameters (concentrations of dissolved oxygen and effluent chemical oxygen demand) but also the abundances of functional bacteria. Notably, we discovered that increasing data volume does not always improve model performance, and that data collection intervals do not need to be excessively small, as moderate intervals can still yield reliable predictions. These findings suggest that excessively large datasets may not be necessary for effective ML predictions in WWTPs. Overall, this study underscores the importance of optimizing dataset size to balance computation efficiency and prediction accuracy, providing valuable insights into data management strategies that can enhance the operational efficiency and sustainability of WWTPs.

Automatic microplastic classification using dual-modality spectral and image data for enhanced accuracy

The development of an automatic microplastic (MPs) classification system using spectra is crucial due to the time-consuming and error-prone nature of analyzing individual spectra, especially with a large quantity of MPs. This study presents a classification system using a dual-modality dataset from micro-Fourier Transform Infrared Spectroscopy (μFTIR) for five common polymer types: polypropylene, polystyrene, polyethylene terephthalate, polyethylene, and polyamide. A comparison of machine learning models, including Decision Tree (DT), Extremely Randomized Trees (ET), Support Vector Classifier (SVC), and Multiclass Logistic Regression (LR), is conducted using features extracted by AlexNet, ResNet18, and Vision Transformer (ViT). Notably, the AlexNet with Logistic Regression (AlexNet-LR) model demonstrated exceptional performance, achieving a validation accuracy of 99.03 % and nearly perfect test scores of 99.99 %. However, ResNet18-LR was selected for web deployment due to its shorter training and inference times compared to AlexNet-LR, while still achieving 99 % validation and test accuracy. This highlights the effectiveness of using a dual-modality dataset for precise microplastic classification. MPsSpecClassify, a web-based application, was developed to enable users to efficiently identify MPs and improve microplastic pollution management.

Exploring Environmental Behaviors and Health Impacts of Biodegradable Microplastics

Biodegradable plastics (BPs) are promoted as eco-friendly alternatives to conventional plastics. However, compared to conventional microplastics (MPs), they degrade rapidly into biodegradable microplastics (BMPs), which may lead to a more significant accumulation of BMPs in the environment. This review systematically compares BMPs and MPs, summarizes current knowledge on their environmental behaviors and impacts on ecosystems and human health, and offers recommendations for future research. BMPs are detected in water, sediments, indoor dust, food, marine organisms, and human samples. Compared to MPs, BMPs are more prone to environmental transformations, such as photodegradation and biodegradation, which results in a shorter migration distance across different matrices. Like MPs, BMPs can adsorb pollutants and transport them into organisms, enhancing toxicity and health risks through the Trojan horse effect. Studies indicate that BMPs may negatively impact terrestrial and aquatic ecosystems more than MPs by disrupting nutrient cycling and inhibiting plant and animal growth. In vivo and in vitro research also shows that BMP degradation products increase bioavailability, exacerbating neurotoxicity and overall toxicity. However, findings on BMPs' environmental and health effects remain inconsistent. Further evaluation of the trade-offs between BMP risks and their biodegradability is needed to address these uncertainties.

Advancements and challenges in microplastic detection and risk assessment: Integrating AI and standardized methods

Microplastics (MPs) pose significant threats to ecosystems and human health due to their persistence and widespread distribution. This paper provides a comprehensive review of sampling methods for MPs in aquatic environments, soils, and biological samples, assessing pre-treatment procedures like digestion and separation. It examines the application and limitations of identification techniques, including microscopic observation, spectroscopic analysis, and thermal analysis. The review highlights the potential of AI technology to enhance detection efficiency and precision. It underscores the necessity of standardized protocols for consistent sampling and detection, and the importance of systematic risk assessment methodologies for managing environmental and health risks associated with MPs. The paper concludes with recommendations for future research, emphasizing the standardization of methods, advancement of detection technologies, integration of AI, and comprehensive health risk assessments. This review will be helpful for researchers to comprehensively understand the current main detection technologies and risk assessment methods of the MP, and to accelerate the establishment of an artificial intelligence regulatory framework for MPs.

Advancing micro-nano supramolecular assembly mechanisms of natural organic matter by machine learning for unveiling environmental geochemical processes

The nano-self-assembly of natural organic matter (NOM) profoundly influences the occurrence and fate of NOM and pollutants in large-scale complex environments. Machine learning (ML) offers a promising and robust tool for interpreting and predicting the processes, structures and environmental effects of NOM self-assembly. This review seeks to provide a tutorial-like compilation of data source determination, algorithm selection, model construction, interpretability analyses, applications and challenges for big-data-based ML aiming at elucidating NOM self-assembly mechanisms in environments. The results from advanced nano-submicron-scale spatial chemical analytical technologies are suggested as input data which provide the combined information of molecular interactions and structural visualization. The existing ML algorithms need to handle multi-scale and multi-modal data, necessitating the development of new algorithmic frameworks. Interpretable supervised models are crucial owing to their strong capacity of quantifying the structure-property-effect relationships and bridging the gap between simply data-driven ML and complicated NOM assembly practice. Then, the necessity and challenges are discussed and emphasized on adopting ML to understand the geochemical behaviors and bioavailability of pollutants as well as the elemental cycling processes in environments resulting from the NOM self-assembly patterns. Finally, a research framework integrating ML, experiments and theoretical simulation is proposed for comprehensively and efficiently understanding the NOM self-assembly-involved environmental issues.

Open-set deep learning-enabled single-cell Raman spectroscopy for rapid identification of airborne pathogens in real-world environments

Pathogenic bioaerosols are critical for outbreaks of airborne disease; however, rapidly and accurately identifying pathogens directly from complex air environments remains highly challenging. We present an advanced method that combines open-set deep learning (OSDL) with single-cell Raman spectroscopy to identify pathogens in real-world air containing diverse unknown indigenous bacteria that cannot be fully included in training sets. To test and further enhance identification, we constructed the Raman datasets of aerosolized bacteria. Through optimizing OSDL algorithms and training strategies, Raman-OSDL achieves 93% accuracy for five target airborne pathogens, 84% accuracy for untrained air bacteria, and 36% reduction in false positive rates compared to conventional close-set algorithms. It offers a high detection sensitivity down to 1:1000. When applied to real air containing >4600 bacterial species, our method accurately identifies single or multiple pathogens simultaneously within an hour. This single-cell tool advances rapidly surveilling pathogens in complex environments to prevent infection transmission.

Organophosphorus Pesticides Management Strategies: Prohibition and Restriction Multi-Category Multi-Class Models, Environmental Transformation Risks, and Special Attention List

Organophosphorus pesticides (OPs) have become one of the most widely used pesticides in Chinese agriculture; however, methods to identify potential restrictions on OPs molecules are lacking. Therefore, this study retrieved the OPs restriction list and constructed eight multi-class, multi-category machine learning models for OPs restrictions. Among these, the random forest (RF) model demonstrated excellent predictive performance, as it was successfully validated and applied. Potential environmental transformation products of OPs were obtained using EAWAG-BBD software, while toxicity indicators for the parent OPs and their transformation products were predicted with ADMETlab 3.0 software. This study found that unrestricted OPs, such as phorate, parathion, and chlorpyrifos, exhibited a high probability of toxicity. Additionally, the environmental transformation products of OPs posed similar comprehensive toxicity risks as the parent compounds. A special attention list for OPs was created based on the toxicity risks of unrestricted parent OPs and their transformation products, using standard deviation classification. Phorate and parathion were identified as OPs requiring special attention. This paper aims to provide an effective method for identifying the potential restriction levels of OPs and to propose an evaluation system that comprehensively considers the health risk, thereby supporting the improvement and optimization of management and usage strategies for OPs.

Artificial intelligence in microplastic detection and pollution control

The rising prevalence of microplastics (MPs) in various ecosystems has increased the demand for advanced detection and mitigation strategies. This review examines the integration of artificial intelligence (AI) with environmental science to improve microplastic detection. Focusing on image processing, Fourier transform infrared spectroscopy (FTIR), Raman spectroscopy, and hyperspectral imaging (HSI), the review highlights how AI enhances the efficiency and accuracy of these techniques. AI-driven image processing automates the identification and quantification of MPs, significantly reducing the need for manual analysis. FTIR and Raman spectroscopy accurately distinguish MP types by analyzing their unique spectral features, while HSI captures extensive spatial and spectral data, facilitating detection in complex environmental matrices. Furthermore, AI algorithms integrate data from these methods, enabling real-time monitoring, traceability prediction, and pollution hotspot identification. The synergy between AI and spectral imaging technologies represents a transformative approach to environmental monitoring and emphasizes the need to adopt innovative tools for protecting ecosystem health.

Machine learning based workflow for (micro)plastic spectral reconstruction and classification

With the advancement of artificial intelligence, it is foreseeable that computer-assisted identification of microplastics (MPs) will become increasingly widespread. Therefore, exploring a machine learning-based workflow to facilitate the identification of MPs is both meaningful and practically significant. However, interferences present in MPs spectra often compromise identification accuracy, making the improvement of spectral quality a critical prerequisite for precise identification. This study developed a fully machine learning-based workflow that combines spectral reconstruction and identification of MPs. To enhance the quality of MPs spectra, two reconstruction models named autoencoders (AE) and V-like convolutional neural networks (VCNN) were employed. Then, four classification models including decision tree, random forest, linear support vector machines (LSVM) and 1D convolutional neural networks were developed to accurately identify MPs. In terms of reconstruction, VCNN outperformed AE with a higher R2 value of 0.965, while both models outperformed conventional widely used Savitzky-Golay algorithm. For classification, LSVM exhibited the best performance with an overall accuracy of 91.35% on the original dataset and 98.00% on the VCNN-reconstructed dataset. When applied to real environmental datasets, a slight decrease in performance was observed, but a maximum top-1 accuracy of 71.43% and top-3 accuracy of >90% was still practically significant, indicating that the combined workflow has great potential for spectral reconstruction and identification of MPs.

Rapid assessment of heavy metal accumulation capability of Sedum alfredii using hyperspectral imaging and deep learning

Hyperaccumulators are the material basis and key to the phytoremediation of heavy metal contaminated soils. Conventional methods for screening hyperaccumulators are highly dependent on the time- and labor-consuming sampling and chemical analysis. In this study, a novel spectral approach assisted with multi-task deep learning was proposed to streamline accumulating ecotype screening, heavy metal stress discrimination, and heavy metals quantification in plants. The significant Cd/Zn co-hyperaccumulator Sedum alfredii and its non-accumulating ecotype were stressed by Cd, Zn, and Pb. Spectral images of leaves were rapidly acquired by hyperspectral imaging. The self-designed deep learning architecture was composed of a shallow network (ENet) for accumulating ecotype identification, and a multi-task network (HMNet) for heavy metal stress type and accumulation prediction simultaneously. To further assess the robustness of the networks, they were compared with conventional machine learning models (i.e., partial least squares (PLS) and support vector machine (SVM)) on a series of evaluation metrics of classification, multi-label classification, and regression. S. alfredii with heavy metals accumulation capability was identified by ENet with 100 % accuracy. HMNet reduced overfitting and outperformed machine learning models with the average exact match ratio (EMR) of heavy metal stress discrimination increased by 7.46 %, and residual prediction deviations (RPD) of heavy metal concentrations prediction increased by 53.59 %. The method succeeded in rapidly and accurately discriminating heavy metal stress with EMRs over 91 % and accuracies over 96 %, and in predicting heavy metals accumulation with an average RPD of 3.29 for Zn, 2.57 for Cd, and 2.53 for Pb, indicating the satisfactory practicability and potential for sensing heavy metals accumulation. This study provides a relatively novel spectral method to facilitate hyperaccumulator screening and heavy metals accumulation prediction in the phytoremediation process.

Novel approach for identifying VOC emission characteristics based on mobile monitoring platform data and deep learning: Application of source apportionment in a chemical industrial park

Refined volatile organic compound (VOC) emission characteristics are crucial for accurate source apportionment in chemical industrial parks. The data from mobile monitoring platforms in chemical industrial parks contain pollution information that is not intuitively displayed, requiring further excavation. A novel approach was proposed to identify VOC emission characteristics using the class activation map (CAM) technology of convolutional neural network (CNN), which was applied on the mobile monitoring platform data (MD) derived from a typical fine chemical industrial park. It converts a large amount of monitoring data with high spatiotemporal complexity into simple and interpretable characteristic maps, effectively improving the identification effect of VOC emission characteristics, supporting more accurate source apportionment of VOC pollution around the park. Using this method, the VOC emission characteristics of eight key factories were identified. VOC source apportionment in the park was conducted for one day using a positive matrix factorization (PMF) model and seven combined factor profiles (CFPs) were calculated. Based on the identified VOC emission characteristics, the main pollution sources and their contributions to surrounding schools and residential areas were determined, revealing that one pesticide factory (named LKA) had the highest contribution ratio. The source apportionment results indicated that the impact of the chemical industrial park on the surrounding areas varied from morning to afternoon, which to some extent reflected the intermittent production methods employed for fine chemicals.

Identification and quantification of nanoplastics in different crops using pyrolysis gas chromatography-mass spectrometry

Quantitative studies of nanoplastics (NPs) abundance on agricultural crops are crucial for understanding the environmental impact and potential health risks of NPs. However, the actual extent of NP contamination in different crops remains unclear, and therefore insufficient quantitative data are available for adequate exposure assessments. Herein, a method with nitric acid digestion, multiple organic extraction combined with pyrolysis gas chromatography-mass spectrometry (Py-GC/MS) quantification was used to determine the chemical composition and mass concentration of NPs in different crops (cowpea, flowering cabbage, rutabagas, and chieh-qua). Recoveries of 74.2-109.3% were obtained for different NPs in standard products (N = 6, RSD <9.6%). The limit of detection (LOD) and the limit of quantitation (LOQ) were 0.02-0.5 μg and 0.06-1.5 μg, respectively. The detection method for NPs exhibited good external calibration curves and linearity with 0.99. The results showed that poly (vinylchloride) (PVC), poly (ethylene terephthalate) (PET), polyethylene (PE), and polyadiohexylenediamine (PA66) NPs could be detected in crop samples, although the accumulation levels associated with the various crops varied significantly. PVC (N.D.-954.3 mg kg-1, dry weight (DW)) and PE (101.3-462.9 mg kg-1, DW) NPs were the dominant components in the samples of all four crop species, while high levels of PET (414.3-1430.1 mg kg-1, DW) NPs were detected in cowpea samples. Furthermore, there were notable differences in the accumulation levels of various edible crop parts, such as stems (60.2%) > leaves (39.8%) in flowering cabbage samples and peas (58.8%) > pods (41.2%) in cowpea samples. This study revealed the actual extent of NP contamination in different types of crops and provided crucial reference data for future research.

Prediction of molecular-specific mutagenic alerts and related mechanisms of chemicals by a convolutional neural network (CNN) model based on SMILES split

Structural alerts (SAs) are essential to identify chemicals for toxicity evaluation and health risk assessment. We constructed a novel SMILES split-based deep learning model (SSDL) that was trained and verified with 5850 chemicals from the ISSSTY database and 384 external test chemicals from published papers. The training accuracy was above 0.90 and the evaluation metrics (precision, recall and F1-score) all reached 0.78 or above on both internal and external test chemicals. In this model, the molecular-specific fragment importance of chemicals was first quantified independently. Then, the SA identification method based on the importance of these fragments was statistically analyzed and verified with the ISSSTY test and external test chemicals containing one of 28 typical SAs, and most of the performances were better than that of expert rules. Furthermore, a mutagenicity mechanism prediction method was developed using 237 chemicals with four known mutagenic mechanisms based on molecular similarity calibrated by the SSDL method and fragment importance, which significantly improved accuracy in three mechanisms and had comparable accuracy in the other one compared to traditional methods. Overall, the SSDL model quantifying fragment toxicity within molecules would be a novel potentially powerful tool in the determination and visualization of molecular-specific SAs and the prediction of mutagenicity mechanisms for environmental or industrial compounds and drugs.

Enhancing spatial resolution in Fourier transform infrared spectral image via machine learning algorithms

Owing to the intrinsic signal noise in the characterization of chemical structures through Fourier transform infrared (FT-IR) spectroscopy, the determination of the signal-to-noise ratio (SNR) depends on the level of the concentration of the chemical structures. In situations characterized by limited concentrations of chemical structures, the traditional approach involves mitigating the resulting low SNR by superimposing repetitive measurements. In this study, we achieved comparable high-quality results to data scanned 64 times and superimposed by employing machine learning algorithms such as the principal component analysis and non-negative matrix factorization, which perform the dimensionality reduction, on FT-IR spectral image data that was only scanned once. Furthermore, the spatial resolution of the mapping images correlated to each chemical structure was enhanced by applying both the machine learning algorithms and the Gaussian fitting simultaneously. Significantly, our investigation demonstrated that the spatial resolution of the mapping images acquired through relative intensity is further improved by employing dimensionality reduction techniques. Collectively, our findings imply that by optimizing research data through noise reduction enhancing spatial resolution using the machine learning algorithms, research processes can be more efficient, for instance by reducing redundant physical measurements.

Prediction of biological nutrients removal in full-scale wastewater treatment plants using H2O automated machine learning and back propagation artificial neural network model: Optimization and comparison

The effective control of total nitrogen (ETN) and total phosphorus (ETP) in effluent is challenging for wastewater treatment plants (WWTPs). In this work, automated machine learning (AutoML) (mean square error = 0.4200 ∼ 3.8245, R2 = 0.5699 ∼ 0.6219) and back propagation artificial neural network (BPANN) model (mean square error = 0.0012 ∼ 6.9067, R2 = 0.4326 ∼ 0.8908) were used to predict and analyze biological nutrients removal in full-scale WWTPs. Interestingly, BPANN model presented high prediction performance and general applicability for WWTPs with different biological treatment units. However, the AutoML candidate models were more interpretable, and the results showed that electricity carbon emission dominated the prediction. Meanwhile, increasing data volume and types of WWTP hardly affected the interpretable results, demonstrating its wide applicability. This study demonstrated the validity and the specific advantages of predicting ETN and ETP using H2O AutoML and BPANN model, which provided guidance on the prediction and improvement of biological nutrients removal in WWTPs.

Supervised classification combined with genetic algorithm variable selection for a fast identification of polymeric microdebris using infrared reflectance

Pollution caused by plastics and, in particular, microplastics has become a source of environmental concern for Society. Their ubiquity, with millions of tons of plastic debris spilled in both land and sea, requires efficient technological improvements in the ways residues are collected, handled, characterized and recycled. For reliable decision-making, dependable chemical information is essential to assess both the nature of the plastics found in the environment and their fate. In this work an efficient method to identify the polymeric composition of microplastic fragments is proposed. It combines infrared reflectance spectra and chemometric methods. A breakthrough result is that the models include polymers weathered under both dry (shoreline) and submerged (in sea water) conditions and, hence, they are very promising as a starting point for eventual practical applications. In addition, no spectral processing is required after the initial measurement. SYNOPSIS: This approach to identify microplastics in aquatic environments combines infrared measurements and multivariate data analysis to fight against (micro)plastic pollution.