Deep learning assisted XRF spectra classification.

EDXRF spectrometry is a well-established and often-used analytical technique in examining materials from which cultural heritage objects are made. The analytical results are traditionally subjected to additional multivariate analysis for archaeometry studies to reduce the initial data's dimensionality based on informative features. Nowadays, artificial intelligence (AI) techniques are used more for this purpose. Different soft computing techniques are used to improve speed and accuracy. Choosing the most suitable AI method can increase the sustainability of the analytical process and postprocessing activities. An autoencoder neural network has been designed and used as a dimension reduction tool of initial [Formula: see text] data collected in the raw EDXRF spectra, containing information about the selected points' elemental composition on the canvas paintings' surface. The autoencoder network design enables the best possible reconstruction of the original EDXRF spectrum and the most informative feature extraction, which has been used for dimension reduction. Such configuration allows for efficient classification algorithms and their performances. The autoencoder neural network approach is more sustainable, especially in processing time consumption and experts' manual work.

Verification of the sampling procedure for radioactivity analysis of waste and industrial material.

In accordance with ISO/IEC 17025:2017, the testing laboratory needs to verify its ability to properly perform methods. For radiological testing, the sampling does not affect the results directly, but it has to ensure that sample adequately represents the tested material. To verify the procedure, a sampling of red mud and bauxite ore was conducted. All samples were measured in identical geometry by HPGe spectrometer. The counting rates per unit mass in the recorded spectra were compared. For each measurement series, the mean and standard deviations of respected peaks were calculated, as well as average and standard deviation of all series. The obtained results of each individual series were considered satisfactory, i.e. sampling procedure ensures the representativeness of the bulk material, if it is within ± two standard deviations of the average of the mean values.

Toxicity of high uranium doses in broilers and protection with mineral adsorbents.

The aim of this study was to determine the uranium distribution and histopathological changes in broiler organs (kidney, liver, and brain) and muscle after 7 days of contamination with high doses of uranyl nitrate hexahydrate (UN), and the protective efficiency of three different mineral adsorbents (organobentonite, organozeolite, and sepiolite). During the 7 days, the UN administration was 50 mg per day, and administration of adsorbents was 2 g per day immediately after UN. In control group where broilers received only UN, histopathological changes such as necrosis of intestinal villi, oedema, vacuolisation and abruption of epithelial cells in renal tubules, oedema and vacuolisation of the cytoplasm of hepatocytes, and dystrophic changes in the neurons of the medulla oblongata were observed. In contrast, when the adsorbents organobentonite, organozeolite, and sepiolite were administered, no histopathological changes were observed in liver and brain. The investigated adsorbents showed the highest protective effects in liver (80-92%), compared to the kidney (77-86%), brain (37-64%), and meat (31-63%).

Natural and anthropogenic radioactivity in the environment of Kopaonik mountain, Serbia.

To evaluate the state of the environment in Kopaonik, a mountain in Serbia, the activity concentrations of (4) K, (226)Ra, (232)Th and (137)Cs in five different types of environmental samples are determined by gamma ray spectrometry, and radiological hazard due to terrestrial radionuclides is calculated. The mean activity concentrations of natural radionuclides in the soil are higher than the global average. However, with an exception of two sampling locations, the external radiation hazard index is below one, implying an insignificant radiation hazard. Apart from (40)K, content of the natural radionuclides is predominantly below minimum detectable activities in grass and cow milk, but not in mosses. Although (137)Cs is present in the soil, grass, mosses and herbal plants, its specific activity in cow milk is below minimum detectable activity. Amongst the investigated herbal plants, Vaccinium myrtillus L. shows accumulating properties, as a high content of (137)Cs is detected therein. Therefore, moderation is advised in consuming Vaccinium myrtillus L. tea.

Uranium distribution in broiler organs and possibilities for protection.

The aim of the present study was to investigate the distribution of uranium (uranyl nitrate hexahydrate, UN) in muscle and organs (kidney, liver, and brain) of broilers, after a 7-day contamination with UN and administration of two different adsorbents (organobentonite and organozeolite). The birds were contaminated during 7 days with 25 mg/UN per day. Adsorbents were given via gastric tube, immediately after contamination with UN. In group 1 that did not receive any adsorbents, histopathological changes in the contaminated broilers were observed in small intestine, liver, and kidney in the form of necrosis of intestinal villi, oedema and cytoplasmic vacuolation of hepatocytes, and dystrophic changes in the kidney tubules epithelium. Organobentonite administered via gastric tube (group 2) reduced uranium distribution by 66 % in kidney, 81 % in liver, and 34 % in brain. In group 3, administration of organozeolite reduced uranium distribution by 67 % in kidney, 68 % in liver, and 49 % in brain. In groups 2 and 3, where the broilers received adsorbents immediately after the UN contamination, no histopathological lesions were observed.

Non-destructive characterisation and classification of ceramic artefacts using pEDXRF and statistical pattern recognition.

BACKGROUND: Portable energy dispersive X-ray fluorescence (pEDXRF) spectrometry analysis was applied for the characterisation of archaeological ceramic findings from three Neolithic sites in Serbia. Two dimension reduction techniques, principal component analysis (PCA) and scattering matrices-based dimension reduction were used to examine the possible classification of those findings, and to extract the most discriminant features. RESULTS: A decision-making procedure is proposed, whose goal is to classify unknown ceramic findings based on their elemental compositions derived by pEDXRF spectrometry. As a major part of decision-making procedure, the possibilities of two dimension reduction methods were tested. Scattering matrices-based dimension reduction was found to be the more efficient method for the purpose. Linear classifiers designed based on the desired output allowed for 7 of 8 unknown samples from the test set to be correctly classified. CONCLUSIONS: Based on the results, the conclusion is that despite the constraints typical of the applied analytical technique, the elemental composition can be considered as viable information in provenience studies. With a fully-developed procedure, ceramic artefacts can be classified based on their elemental composition and well-known provenance.