Due to the pathogenic nature of asbestos, a statutory ban on asbestos-containing products has been in place in Poland since 1997. In order to protect human health and the environment, it is crucial to estimate the quantity of asbestos–cement products in use. It has been evaluated that about 90% of them are roof coverings. Different methods are used to estimate the amount of asbestos–cement products, such as the use of indicators, field inventory, remote sensing data, and multi- and hyperspectral images; the latter are used for relatively small areas. Other methods are sought for the reliable estimation of the quantity of asbestos-containing products, as well as their spatial distribution. The objective of this paper is to present the use of convolutional neural networks for the identification of asbestos–cement roofing on aerial photographs in natural color (RGB) and color infrared (CIR) compositions. The study was conducted for the Chęciny commune. Aerial photographs, each with the spatial resolution of 25 cm in RGB and CIR compositions, were used, and field studies were conducted to verify data and to develop a database for Convolutional Neural Networks (CNNs) training. Network training was carried out using the TensorFlow and R-Keras libraries in the R programming environment. The classification was carried out using a convolutional neural network consisting of two convolutional blocks, a spatial dropout layer, and two blocks of fully connected perceptrons. Asbestos–cement roofing products were classified with the producer’s accuracy of 89% and overall accuracy of 87% and 89%, depending on the image composition used. Attempts have been made at the identification of asbestos–cement roofing. They focus primarily on the use of hyperspectral data and multispectral imagery. The following classification algorithms were usually employed: Spectral Angle Mapper, Support Vector Machine, object classification, Spectral Feature Fitting, and decision trees. Previous studies undertaken by other researchers showed that low spectral resolution only allowed for a rough classification of roofing materials. The use of one coherent method would allow data comparison between regions. Determining the amount of asbestos–cement products in use is important for assessing environmental exposure to asbestos fibres, determining patterns of disease, and ultimately modelling potential solutions to counteract threats.

Exposure to asbestos causes a wide range of diseases, such as asbestosis, malignant mesothelioma (MM) and other types of cancer. Many European countries have reduced production and use of asbestos and some have banned it altogether. Based on data derived from the World Health Organisation (WHO) Cancer Mortality Database, we investigated whether some regions in Europe could have a higher relative risk of MM incidence than others. The data were compared, including the number of MM deaths per million inhabitants and aged-standardized mortality rates. Applying Moran’s I and Getis-Ord Gi statistic on the agedstandardized mortality rates of MM cases assisted the spatial analysis of the occurrence of health events leading to an assessment of the heterogeneity of distribution and cluster detection of MM. We found a statistically significant positive autocorrelation for the male population and also the general population, while there was no statistically significant positive one for the female population. Hotspots of relative risk of developing MM were found in northwestern Europe. For the general population, Great Britain and the Netherlands stood out with high levels at the 99% and 95% confidence levels, respectively. For the male population, the results were similar, but with addition of risk also in Belgium and Switzerland. However, in many European countries with high asbestos use per capita, the MM incidence was found to still be low. The reasons for this are not yet clear, but part of the problem is certainly due to incomplete data in registers and databases. The latency time can be longer than 40 years and is related to the intensity and time of exposure (occupational, para-occupational and environmental). In Europe, even though peak production occurred in the 1960s and 1970s, a significant decrease in production did not occur until 25 years later, which means that the impact will continue for as late as The mid 2030s.

Asbestos is carcinogenic to humans; the exposure to asbestos causes a wide range of diseases. Aim: Malignant mesothelioma (MM) is unique for asbestos exposure. Methods: Based on the physical inventory of asbestos-cement roofing, the social-economic situation of communes, the proximity of asbestos manufacturing plants, the land use data referring to the surface of the built-up area, and the historical data on the annexations, the amount of asbestos-containing products in use was estimated by computing best Random Forest models. Per capita asbestos use is an indicator to compare the state of asbestos use among countries. MM cases in the local administrative units (provinces) were tested by the application of Moran’s I and Getis and Ord statistic. Results: The total amount of asbestos roofing in Poland was estimated at 738,068,000 m² (8.2 million tons). In total there were 28 plants in Poland located in 11 provinces throughout the country. The amount of asbestos-cement roofing in use is correlated primarily with the measurements of asbestos concentration fibers (r_s = 0.597). MM raw morbidity rate was calculated, stratified by province, and classified into five groups with respect to incidence. Hotspots of MM cases are in the southern part of Poland. Conclusions: MM cases are concentrated in the same geographical areas, which may indicate an increasing impact of environmental exposure. The results of the local and global autocorrelation clearly indicate a statistically significant relationship between incidences of MM in provinces. Poland and other Eastern European countries are among countries with low MM incidence rate. Detailed investigation is desirable since the current MM morbidity rate in Poland seems to be underestimated.

Asbestos is harmful to human health; exposure to asbestos causes a wide range of asbestos-related diseases. Aim: Malignant mesothelioma (MM) is unique to occupational and environmental asbestos exposure. Methods: Environmental asbestos exposure was examined in relation to asbestos use and manufacturing, the quantity of the asbestos-containing products still in use, the concentrations of asbestos fibres in the air and the number of MM cases diagnosed each year per county. Results: The correlation coefficient of the measurements of the asbestos fibre concentrations in the air and the quantity of asbestos-cement products in use is high and amounts to 0.68. Meanwhile, the correlation coefficient of the measurements of asbestos fibre concentrations in air and MM morbidity rate resulting from environmental exposure calculated for particular counties in provinces is low and amounts to 0.37. The highest MM morbidity rate was observed for Małopolskie and Śląskie, a typical industrial area of Poland. Conclusions: There are MM cases which are still attributable to occupational asbestos exposure, although MM cases resulting from environmental exposure to asbestos have an increased MM risk. Poland is among those countries with a low MM incidence rate, which seems to be an underestimation of environmental asbestos exposure. As long as asbestos-cement products are used in the environment, actions should be undertaken to protect public health.

Asbestos and asbestos containing products are harmful to human health, and therefore its use has been legally forbidden in the EU. Since there is no adequate data on the amount of asbestos-cement roofing in Poland, the objective of this study was to map asbestos-cement roofing with the use of hyperspectral APEX data (288 bands at the spatial resolution of 2.7 m) in the Karpacz area (southwest Poland). A field survey constituted the basis for training and verification polygons in the classification process. A SAM classification method was performed with the following classification results: 62% producer’s accuracy, 73% user’s accuracy and an overall accuracy of 95%. The asbestos-cement roofing for buildings may be discriminated with a high classification accuracy with the use of hyperspectral imagery. The vast majority of the classified buildings were characterised by their small area (i.e. residential type buildings), which reduced the overall accuracy of the classification.