Objectives: The integration of machine learning and radiomics in medical imaging has significantly advanced diagnostic and prognostic capabilities in healthcare. This study focuses on developing and validating an artificial intelligence (AI) model using U-Net architectures for the accurate detection and segmentation of spinal metastases from computed tomography (CT) images, addressing both osteolytic and osteoblastic lesions. Methods: Our methodology employs multiple variations of the U-Net architecture and utilizes two distinct datasets: one consisting of 115 polytrauma patients for vertebra segmentation and another comprising 38 patients with documented spinal metastases for lesion detection. Results: The model demonstrated strong performance in vertebra segmentation, achieving Dice Similarity Coefficient (DSC) values between 0.87 and 0.96. For metastasis segmentation, the model achieved a DSC of 0.71 and an F-beta score of 0.68 for lytic lesions but struggled with sclerotic lesions, obtaining a DSC of 0.61 and an F-beta score of 0.57, reflecting challenges in detecting dense, subtle bone alterations. Despite these limitations, the model successfully identified isolated metastatic lesions beyond the spine, such as in the sternum, indicating potential for broader skeletal metastasis detection. Conclusions: The study concludes that AI-based models can augment radiologists’ capabilities by providing reliable second-opinion tools, though further refinements and diverse training data are needed for optimal performance, particularly for sclerotic lesion segmentation. The annotated CT dataset produced and shared in this research serves as a valuable resource for future advancements.
The objective of this study was to develop and evaluate artificial intelligence (AI) models for the detection and instance segmentation of vertebrae and spinal metastases in computer tomography (CT) scans. The models were trained on datasets consisting of patients with polytrauma and relatively undamaged spines, as well as patients diagnosed with spinal metastases. Our results indicate that the models achieved high performance in vertebra segmentation, with F-beta scores ranging from 0.88 to 0.96 across all vertebra classes. For spinal metastases, the model attained F-beta scores of 0.68 for lytic type and 0.57 for sclerotic type metastases. Additionally, the models demonstrated the capability to detect isolated metastatic nodes in other bones, highlighting their robustness and potential for broader clinical application. Despite these promising results, the study faced limitations including a relatively small and homogeneous dataset, variability in segmentation mask quality, and the need for real-world clinical validation. Future clinical trials are necessary to evaluate the practical utility and effectiveness of these AI models in improving patient outcomes. Our study emphasizes the potential of AI-assisted detection and segmentation models in enhancing diagnostic accuracy and efficiency in clinical practice.
Understanding the environmental pathways and surface modification of beach sand grains is essential for reconstructing coastal dynamics and assessing the suitability of natural sands for engineering applications. This study applies a multiproxy approach—integrating grain roundness classification, SEM microtextural analysis, and XPS surface chemistry—to beach sediments from four coastal sectors of Latvia: Liepaja, Ventspils, Riga, and Salacgrīva. The results reveal clear spatial differences in grain maturity, abrasion signatures, biological imprinting, and nanoscale surface composition. Liepaja is characterised by sub-rounded to rounded grains with abundant percussion pits and abrasion surfaces, indicating prolonged high-energy wave reworking. Ventspils retains angular grains with fresh conchoidal fractures, reflecting rapid sediment renewal from glacial and coastal sources. Riga exhibits weak abrasion and hydrated particulate coatings typical of low-energy brackish environments. Salacgrīva displays strong fluvial influence, including persistent diatom and algal microtextural features and elevated oxygenated carbon and metal-associated XPS signals. These findings demonstrate strong coupling between grain-surface microtextures and surface chemistry and reveal distinct sedimentary fingerprints linked to environmental setting. The multiproxy framework presented here improves understanding of Baltic coastal sediment pathways and provides a preliminary basis for future evaluation of natural sands in filtration and other environmental engineering applications.
Titanium and its alloys are widely used in endoprostheses. The naturally formed titanium dioxide film on titanium surfaces improves chemical stability and enhances implant biocompatibility. However, oxidised titanium surfaces may also promote bacterial adhesion and biofilm formation, contributing to implant-associated infections. Therefore, surface modification represents a key strategy for controlling microbial–implant interactions. This article focuses widely used titanium alloy Ti-6Al-4V treated with a laser beam, which induces surface colour changes as a result of oxide formation. Laser processing enables controlled formation of micro- and nanoscale features, structural reconstructions, and defects that may influence the surface electrical charge and, consequently, cell immobilisation. Thus, the surface colour, electrical potential, and cell immobilisation capacity are likely interrelated. From a manufacturing perspective, titanium oxide colouring facilitates quality control and process reproducibility, as surface colour provides a rapid, non-destructive visual indicator of oxide thickness and treatment consistency. This study aims to identify correlations among surface colour, electrical potential, and cell immobilisation capacity on laser-treated titanium alloys. A relationship between the optical properties, electronic structure, and biological response of laser-processed titanium oxide films is established. Specifically, the blue colour saturation of the oxide film is inversely correlated with the electron work function. A more saturated blue corresponds to a lower work function, indicating a higher positive surface charge density. This shift is attributed to changes in electron affinity, likely resulting from laser-induced structural reconstruction and defect formation within the oxide layer. The proposed changes in electronic structure are supported by modifications in the electronic density of states, analysed using near-threshold photoelectron spectroscopy. The biological response is directly linked to these physical changes: enhanced immobilisation of yeast (Saccharomyces cerevisiae) cells on the treated alloy surface correlates with the electron work function. These results may assist in the development of controlled titanium oxide surfaces with enhanced biocompatibility.
Heavy metal contamination in coastal and ballast waters motivates the development of low-cost, environmentally compatible filtration media. This study investigates how 6 MeV electron-beam irradiation (0–100 Gy) modifies the surface electronic and chemical properties of quartz-rich Baltic Sea sands collected from four Latvian coastal locations (Riga, Salacgriva, Ventspils, and Liepaja), and how these modifications affect chromium removal from aqueous K2CrO4 solutions. Surface electronic behavior was evaluated by near-threshold photoelectron emission spectroscopy (PEES), including electron work function (EWF) and analysis of differentiated spectra, while irradiation-associated changes in near-surface chemistry were assessed by X-ray photoelectron spectroscopy (XPS). Filtration performance was quantified by UV–Vis absorbance of filtrates. Across all sands, EWF values remained within ~4.7–4.9 eV; however, irradiation effects were strongly site-dependent. Liepaja sand exhibited the most pronounced response, including an EWF increase at 40 Gy, a shift in the differentiated PEES peak toward higher photon energies at ≥40 Gy, and the largest integrated photoemission intensity across doses, consistent with an elevated relative photoemission response under identical acquisition and processing conditions. XPS trends for Liepaja were consistent with irradiation-driven modification of the Si–O environment, while other sites showed comparatively minor changes. Filtration results mirrored these observations: Liepaja sand demonstrated the clearest dose-dependent enhancement in chromium removal with a non-monotonic feature at 40 Gy, consistent with competing formation and transformation of oxygen-related surface-reactive centers. Overall, the results show that electron-beam irradiation can modestly enhance Cr(VI) removal by natural quartz sands, with the magnitude governed by site-specific near-surface electronic structure and its dose-dependent evolution.