The PCM (phase contrast microscopy) method for asbestos counting needs special sample treatments, hence it is time consuming and rather expensive. As an alternative, we implemented a deep learning procedure on images directly acquired from the untreated airborne samples using standard Mixed Cellulose Ester (MCE) filters. Several samples with a mix of chrysotile and crocidolite with different concentration loads have been prepared. Using a 20x objective lens coupled with a backlight illumination system a number of 140 images were collected from these samples, which along with additional 13 highly fibre loaded artificial images constituted the database. About 7500 fibres were manually recognised and annotated following the National Institute for Occupational Safety and Health (NIOSH) fibre counting Method 7400 as input for the training and validation of the model. The best trained model provides a total precision of 0.84 with F1-Score of 0.77 at a confidence of 0.64. A further post-detection refinement to ignore detected fibres < 5 µm in length improves the final precision. This method can be considered as a reliable and competent alternative to conventional PCM.

Asbestos has been widely used in many applications for its technical properties (i.e. resistance to abrasion, heat and chemicals). Despite its properties, asbestos is recognized as a hazardous material to human health. In this paper a study, based on multivariate analysis, was carried out to verify the possibilities to utilize the hyperspectral imaging (HSI), working in the short-wave infrared range (SWIR: 1000-2500 nm), to detect the presence of asbestos-containing materials (ACM) in construction and demolition waste (CDW). Multivariate classification methods including classification and regression tree (CART), partial least squares-discriminant analysis (PLS-DA) and correcting output coding with support vector machines (ECOC-SVM), were adopted to perform the recognition/classification of ACM in respect of the other fibrous panels not containing asbestos, in order to verify and compare Efficiency and robustness of the classifiers. The correctness of classification results was confirmed by micro-X-ray fluorescence maps. The results demonstrate as SWIR technology, coupled with multivariate analysis modeling, is a quite promising approach to develop both “off-line” and “on-line” fast reliable and robust quality control strategies, finalized to perform a first evaluation of the presence of ACM.

Asbestos was largely used in the past by several countries all over the world. From 1900 to 1990 asbestos-containing materials (ACMs) were produced in large amounts and mainly utilized for the production of insulation, flame retardant materials, as well as to improve the mechanical and the chemical characteristics of construction materials. Its extensive use has therefore led to the presence of fibers in existing buildings and within the construction and demolition waste. A fast, reliable and accurate recognition of ACMs represents an important target to be reached. In this paper the use of micro X-ray fluorescence (micro-XRF) technique coupled with a statistical multivariate approach was applied and discussed with reference to ACMs characterization. Different elemental maps of the ACMs were preliminary acquired in order to evaluate distribution and composition of asbestos fibers, then samples energy spectra where collected and processed using chemometric methods to perform an automatic classification of the different typologies of asbestos fibers. Spectral data were analyzed using PLS-Toolbox™ (Eigenvector Research, Inc.) running into Matlab® (The Mathworks, Inc.) environment. An automatic classification model was then built and applied. Results showed that asbestos fibers were correctly identified and classified according to their chemical composition. The proposed approach, based on micro-XRF analysis combined with an automatic classification of the elemental maps, is not only effective and non-destructive, it is fast and it does not require the presence of a trained operator. The application of the developed methodology can help to correctly characterize and manage demolition waste where ACMs are present.

Asbestos fibers are well-known carcinogens, and their rapid detection is critical for ensuring safety, protecting public health, and promoting environmental sustainability. In this work, short-wave infrared (SWIR) spectroscopy, combined with machine learning (ML), was evaluated as an environmentally friendly analytical approach for simultaneously distinguishing the asbestos type, asbestos-containing materials in various forms, asbestos-contaminated/-uncontaminated soil, and asbestos-contaminated/-uncontaminated cement, simultaneously. This approach offers a noninvasive and efficient alternative to traditional laboratory methods, aligning with sustainable practices by reducing hazardous waste generation and enabling in situ testing. Different chemometrics techniques were applied to discriminate the material classes. In more detail, partial least squares discriminant analysis (PLS-DA), principal component analysis-based discriminant analysis (PCA-DA), principal component analysis-based K-nearest neighbors classification (PCA-KNN), classification and regression trees (CART), and error-correcting output-coding support vector machine (ECOC SVM) classifiers were tested. The tested classifiers showed different performances in discriminating between the analyzed samples. CART and ECOC SVM performed best (${Recall}_M$ and ${Accuracy}_M$ equal to 1.00), followed by PCA-KNN (${Recall}_M$ of 0.98–1.00 and ${Accuracy}_M$ equal to 1.00). Poorer performances were obtained by PLS-DA (${Recall}_M$ of 0.68–0.72 and ${Accuracy}_M$ equal to 0.95) and PCA-DA (${Recall}_M$ of 0.66–0.70 and ${Accuracy}_M$ equal to 0.95). This research aligns with the United Nations’ Sustainable Development Goals (SDGs), particularly SDG 3 (Good Health and Well-Being), by enhancing human health protection through advanced asbestos detection methods, and SDG 12 (Responsible Consumption and Production), by promoting sustainable, low-waste testing methodologies.