Clustering of the self-organizing map

Assessing coastal vulnerabilities impacting drinking water sources and sanitation: spatial, multivariate and ML approach in Satkhira, Bangladesh

With an emphasis on the effects of climate change, saltwater intrusion, and contaminated surface water in rural areas, this study investigates the susceptibility of sanitation and drinking water sources in Bangladesh's Satkhira District. Evaluating the sustainability and safety of drinking water sources more especially, groundwater, protected ponds, and pond sand filters was the main goal. To learn more about their water collection methods, reliance on various water sources, hygiene habits, and health concerns, we conducted a survey with 2000 residents. Furthermore, 24 water samples were gathered prior to winter and examined for hydrochemical parameters (pH, EC, TDS, salinity, temperature) and ions (NH₄⁺, NO₃⁻, PO₄3⁻, SO₄2⁻, and Cl⁻). According to the correlation heatmap, we found a significant positive correlation (0.97) between electrical conductivity and total dissolved solids. By highlighting the connections between these variables, Principal Component Analysis (PCA) was able to account for 53.6% of the overall variation in water quality. Water quality was categorized using a dendrogram according to soil, climate, and farming methods. Both national and international criteria were used to assess the quality of the water. The findings showed serious problems with water quality, such as ammonium levels that were higher than acceptable limits and chloride levels that were lower than the national average. According to the study, the main causes of waterborne illnesses, especially in children, were saltwater intrusion, poor sanitation, and water scarcity. Inverse Distance Weighting (IDW) analysis reveals high concentrations of pH, EC, TDS, salinity, and ions in the western and southern regions of Satkhira, influenced by human activities and local factors. The Self-Organizing Map (SOM) analysis reveals clear spatial patterns in water quality variables, highlighting the influence of pollution sources like agricultural runoff and industrial effluents on ion concentrations, salinity, and pH levels. The study highlights the necessity of better water management techniques and increased community engagement to guarantee that Satkhira District people have access to clean drinking water.

Application of machine learning in soil heavy metals pollution assessment in the southeastern Tibetan plateau

The Tibetan Plateau, a globally significant ecological region, is experiencing escalating pollution from heavy metals (HMs). This study applies a machine learning approach based on the self-organizing map hyper-clustering, alongside advanced methodologies such as Positive Matrix Factorization (PMF), Incremental Spatial Autocorrelation, and Bivariate Local Indicators of Spatial Association (BiLISA), to analyze the ecological risk of soil HMs in representative watersheds of the southeastern Tibetan Plateau, focusing on spatial pattern clustering, pollutant source identification, and interaction risk assessment. The results indicated higher HMs concentrations in the middle and downstream areas. A comprehensive ecological risk assessment integrating the Improved Potential Ecological Risk Index, Enrichment Factor, Contamination Factor, and Geo-accumulation Index identified Cd, Pb, and As as the primary pollutants of concern. By combining PMF with Mantel analysis, pollution was attributed to geological background, agricultural activities, traffic emissions, and atmospheric deposition. The BiLISA method revealed significant spatial interactions among HMs, with the composite pollution of As and Cd occupying the largest proportion in High (As)-High (Cd) aggregation zones, underscoring the need for integrated management strategies. This study offers novel insights into the spatial pollution patterns and source apportionment of soil HMs, providing an advanced analytical framework for their precise control and ecological restoration.

Spatial pattern of groundwater chemistry in a typical piedmont plain of Northern China driven by natural and anthropogenic forces

Groundwater is crucial for human society's development in piedmont plains, yet its hydrogeochemistry often exhibits complex spatial distributions due to the interplay of nature and human factors. Ninety-two phreatic groundwater samples were collected from a typical piedmont plain in northern China and analyzed using self-organizing map combined with hydrogeochemical simulation, diagrams, and the entropy-weighted water quality index. Groundwater samples were categorized into four clusters, demonstrating a gradual hydrogeochemical facies evolution from HCO3-Ca to Cl-Mg·Ca and Cl-Na, along with an increase in NO3- content in the order of clusters IV, II, III, and I. Natural processes, including silicates weathering and reverse cation-exchange, establish the natural fundamental framework of groundwater chemistry, which is furtherly sculptured by agricultural substances input. Groundwater quality was predominantly excellent or good, with entropy-weighted water quality index (EWQI) values below 100 at over 92% of the sampling sites. Groundwater quality is relatively poorer in the upstream areas near the mountains and along the Hutuo River, where the stratum permeability is high, but improves in the downstream areas where permeability is lower. Agricultural land use and spatial variation in aquifer permeability are responsible for the observed spatial variations in groundwater chemistry. Agricultural contaminants warrant attention for the protection of groundwater quality in piedmont plains that with long-term agricultural activities, especially in the upstream areas near the mountains. This research improves the understanding of the spatial distribution and variation of groundwater chemistry in piedmont plains, and provides scientific guidance for related groundwater development and management.

Disentangling Multiannual Air Quality Profiles Aided by Self-Organizing Map and Positive Matrix Factorization

The evaluation of air pollution is a critical concern due to its potential severe impacts on human health. Currently, vast quantities of data are collected at high frequencies, and researchers must navigate multiannual, multisite datasets trying to identify possible pollutant sources while addressing the presence of noise and sparse missing data. To address this challenge, multivariate data analysis is widely used with an increasing interest in neural networks and deep learning networks along with well-established chemometrics methods and receptor models. Here, we report a combined approach involving the Self-Organizing Map (SOM) algorithm, Hierarchical Clustering Analysis (HCA), and Positive Matrix Factorization (PMF) to disentangle multiannual, multisite data in a single elaboration without previously separating the sites and years. The approach proved to be valid, allowing us to detect the site peculiarities in terms of pollutant sources, the variation in pollutant profiles during years and the outliers, affording a reliable interpretation.

Global climate-driven sea surface temperature and chlorophyll dynamics

Herein we study long-term changes in global sea surface temperature (SST) and chlorophyll-a concentration (CHL) in order to evaluate possible effects of climate change on the global marine ecosystems. Our approach is to analyze multi-model ensemble-means from global numerical-simulations available through the Coupled Model Intercomparison Project Phase 6 (CMIP6). A 250-year span consisting of the 1850-2014 historical period and the 2015-2099 climate-change projection was considered, where the Shared Socioeconomic Pathways (SSPs) 2.45 and 5.85 were selected as the projected climate-change scenarios. In the historical period, global linear trends show an SST increasing at 0.0024 °C year-1 and a CHL decreasing at -2.35x10-5 mg m-3 year-1, but by the last years (1985-2014) these changes become more abrupt: SST rising at 0.0146 °C year-1 and CHL declining at -1.49x10-4 mg m-3 year-1. During the intense climate-change scenario (SSP-5.85), SST increases at 0.0341 °C year-1 and CHL decreases at -0.0002 mg m-3 year-1, but in the last years (2070-2099) the warming is stronger (0.045 °C year-1) and the CHL decline is weaker (-0.0001 mg m-3 year-1). Additionally, Empirical Orthogonal Function (EOF) and dual Self-Organizing Maps (SOM) analyses on the model-data ensembles show: 1) significant correlations between SST and CHL patterns and climate teleconnection indices, 2) contracting polar and high latitude seascapes, 3) rising SST range (both high and low temperatures), 4) declining CHL in warming tropical seascapes, and 5) global expansion of low CHL levels. On the other hand, recent (2022-2023) global observed-SST anomalies mirror end-of-century projections, with extreme anomalies in tropical and subtropical regions and significant changes in near-polar regions. Thus, our findings emphasize the need to curb fossil fuel emissions in order to avoid irreparable consequences for the marine environment.

Learning model combined with data clustering and dimensionality reduction for short-term electricity load forecasting

Electric load forecasting is crucial in the planning and operating electric power companies. It has evolved from statistical methods to artificial intelligence-based techniques that use machine learning models. In this study, we investigate short-term load forecasting (STLF) for large-scale electricity usage datasets. We propose a new prediction model for STLF that combines data clustering and dimensionality reduction schemes to handle large-scale electricity usage data effectively. Here, we adapt k-means clustering for data clustering, kernel principal component analysis (kernel PCA), universal manifold approximation and projection (UMAP), and t-stochastic nearest neighbor (t-SNE) for dimensionality reduction. To verify the effectiveness of the proposed model, we extensively apply it to neural network-based models. We compare and analyze the performance of the proposed model with the comparisons using actual electricity usage data for 4710 households. Experimental results demonstrate that data clustering with dimensionality reduction can improve the performance of baseline models. As a result, the prediction accuracy of the proposed method outperforms those of the existing methods by 1.01-1.76 times for summer data and by 1.03-1.36 times for winter data in terms of mean absolute percentage error (MAPE).

Integrating spatial clustering and multi-source geospatial data for comprehensive geological hazard modeling in Hunan Province

This study presents an integrated framework that combines spatial clustering techniques and multi-source geospatial data to comprehensively assess and understand geological hazards in Hunan Province, China. The research integrates self-organizing map (SOM) and geo-self-organizing map (Geo-SOM) to explore the relationships between environmental factors and the occurrence of various geological hazards, including landslides, slope failures, collapses, ground subsidence, and debris flows. The key findings reveal that annual average precipitation (Pre), profile curvature (Pro_cur), and slope (Slo) are the primary factors influencing the composite geological hazard index (GI) across the province. Importantly, the relationships between these key factors and GI exhibit spatial variability, as evidenced by the random intercept and slope models, highlighting the need for customized mitigation strategies. Additionally, the study demonstrates that land use patterns and stratigraphic stratum lithology significantly impact the cluster-specific relationships between the key factors and GI, emphasizing the importance of natural resource management for effective geological hazard mitigation. The proposed integrated framework provides valuable insights for policymakers and resource managers to develop spatially-aware strategies for geological hazard risk reduction and climate change adaptation.

Unsupervised self-organising map classification of Raman spectra from prostate cell lines uncovers substratified prostate cancer disease states

Prostate cancer is a disease which poses an interesting clinical question: Should it be treated? Only a small subset of prostate cancers are aggressive and require removal and treatment to prevent metastatic spread. However, conventional diagnostics remain challenged to risk-stratify such patients; hence, new methods of approach to biomolecularly sub-classify the disease are needed. Here we use an unsupervised self-organising map approach to analyse live-cell Raman spectroscopy data obtained from prostate cell-lines; our aim is to exemplify this method to sub-stratify, at the single-cell-level, the cancer disease state using high-dimensional datasets with minimal preprocessing. The results demonstrate a new sub-clustering of the prostate cancer cell-line into two groups-protein-rich and lipid-rich sub-cellular components-which we believe to be mechanistically linked. This finding shows the potential for unsupervised machine learning to discover distinct disease-state features for more accurate characterisation of highly heterogeneous prostate cancer. Applications may lead to more targeted diagnoses, prognoses and clinical treatment decisions via molecularly-informed stratification that would benefit patients. A method that could discover distinct disease-state features that are mechanistically linked could also assist in the development of more effective broad-spectrum treatments that simultaneously target linked disease-state processes.

Cluster analysis of social determinants of health and HIV/AIDS knowledge among Peruvian youths using Kohonen's self-organized maps: a data-exploration study based on a Demographic and health survey

BACKGROUND

Human immunodeficiency virus (HIV) and acquired immunodeficiency syndrome (AIDS) have evolved into a global development burden, with nearly 40 million infections and 25 million deaths. Compared to other age groups, youth have increased risks of contracting the disease due to social and health structural factors; thus, additional efforts are needed to effectively tackle the challenges associated with this age group. Epidemiological studies employing unsupervised learning techniques are essential for shaping public health policies.

OBJECTIVE

This study aimed to describe the Peruvian youth population based on their sociodemographic, health, and economic characteristics using an unsupervised learning approach through the development of a neural network model based on Kohonen's self-organizing maps (SOMs), allowing the identification of social profiles in the study population.

METHODS

This quantitative study used data from the 2019 Peruvian Demographic and Family Health Survey. An SOM network model for clustering individuals with similar attributes and clustering prototype vectors based on the agglomerative hierarchical clustering (AHC) method and their visualization on an SOM was applied to the study sample.

RESULTS

Clustering of prototype vectors yielded four clusters, each of which represented a profile of Peruvian youths based on their knowledge of HIV/AIDS and structural health determinants.

CONCLUSIONS

Kohonen's neural networks allowed the identification of patterns and behaviors among youths in Peru, quantifying and characterizing the four social clusters regarding HIV/AIDS and their social determinants. Kohonen's maps may benefit healthcare professionals and policymakers by offering a useful method for tailoring interventions and policies based on the detected profiles, thereby enhancing the visibility of these focal points at the national level.

Hydrochemical characteristics, control factors and health risk assessment of groundwater in typical arid region Hotan Area, Chinese Xinjiang

The Hotan region of Xinjiang is an arid region in northwest China, where water resources are scarce, and groundwater is the main water supply. In this study, a self-organizing map (SOM), positive matrix factorization (PMF), hydrochemical diagrams, and health risk assessment model were used to analyze the sources and controlling factors of groundwater chemistry, and evaluate health risks of nitrate and fluoride. The results showed that the evaporation process and water-rock interaction were the main factors influencing groundwater chemistry in the region. Based on the SOM, 239 groundwater samples were divided into six clusters. The main hydrochemical types were Cl-Na, HCO3-Na, and SO4-Ca. Natural factors such as evaporation, water-rock interaction and cation exchange play important roles in Cluster 1-2 and 4-6, while Cluster 3 is mainly polluted by nitrate. Fluoride pollution, primarily caused by geological processes, and nitrate pollution, caused by human activities, cannot be ignored. Attention should be paid to the high non-carcinogenic risk of fluoride and nitrate exposure through drinking water, especially for children. These results provide a theoretical basis for the rational development and utilization of local water resources and ecological environmental protection. The study suggested that the combined method of the SOM and PMF provides a reliable approach for interpreting nonlinear and high-dimensional hydrochemical data.

Machine learning-driven assessment of heavy metal contamination in the impounded lakes of China's South-to-North Water Diversion Project: Identifying spatiotemporal patterns and ecological risks

The Eastern Route of China's South-to-North Water Diversion Project (SNWDP-ER) traverses through impounded lakes that are potentially vulnerable to heavy metals (HMs) contamination although the understanding remains elusive. This study employed machine learning approaches, including super-clustering of Self-Organizing Map (SOM) and Robust Principal Component Analysis (RPCA), to elucidate the spatiotemporal patterns and assess ecological risks associated with HMs in the surface sediments of Gao-Bao-Shaobo Lake (GBSL) and Dongping Lake (DPL). We collected 184 surface sediments from 47 stations across the two important impounded lakes over four seasons. The results revealed higher HMs concentrations in the south-central GBSL and west-central DPL, with a notable increase in contamination in autumn. The comprehensive risk assessment, utilizing various indicators such as the Sediment Quality Guidelines (SQGs), Improved Potential Ecological Risk Index (IPERI), Geo-accumulation Index (Igeo), Contamination Factor (CF), and Enrichment Factor (EF), identified arsenic (As), cadmium (Cd), nickel (Ni), and chromium (Cr) as primary contaminants of concern. Positive Matrix Factorization (PMF) model, coupled with Spearman analysis, attributed over 70 % of HMs pollution to anthropogenic activities. This research provides a nuanced understanding of HMs pollution in the context of large-scale water diversion projects and offers a scientific basis for targeted pollution mitigation strategies.

Water quality classification using self-organizing maps and cluster analysis: Case of Meknes-El Hajeb Springs, Morocco

The Ouislane sub-watershed is currently experiencing severe water shortages and is highly dependent on its water supply. The sub-watershed spans two communes: Meknes to the north and El Hajeb to the south. It serves as the primary water source for irrigation and drinking purposes for the local population. Consequently, it is crucial to assess the spatio-temporal variations of water quality to identify and address potential gaps; these focused on effective monitoring systems to detect contaminants, pollutants and health risks. This research project aims on the application of self-organizing map (SOM) techniques combined with cluster analysis to classify water quality in springs for drinking and irrigation purposes. The present study evaluates the water quality variations using physicochemical parameters of twelve water springs, collected during the wet and dry seasons of 2022. For this purpose, the water quality index (WQI), self-organizing map (SOM), hierarchical cluster analysis (HCA), and principal component analysis (PCA) are used as evaluation and classification methods. As a result, the SOM algorithm with a size of 5 × 5 units identified as the most suitable, based on the minimum quantization error (QE) and topographic error (TE), yielding a QE of 0.379 and a TE of 0.000. It grouped the water quality data into five distinct clusters, Cluster I represented 37.5% of the total samples, while cluster II represented 25%. Cluster III and IV each accounted for 8.33% of the samples, while 20.83% of the sampling water are classified in cluster V. Clusters I, II, and IV indicate good water suitable for drinking. However, cluster V had the highest WQI, suggesting very high contamination due to increased levels of the 10 studied physicochemical parameters. The water quality in this region (cluster V) is influenced by natural processes, such as precipitation intensity, weathering and vegetation cover, as well as anthropogenic factors like agriculture and urban concentration. PCA confirmed the clustering results obtained by SOM. However, SOM provides a more detailed classification and additional insights into the dominant variables influencing the classification processes. The results of this study suggest that SOM was an effective tool for gaining a better understanding of the patterns and processes driving water quality in the Ouislane sub-watershed and provides valuable avenues for further research to establish and monitor water quality for effective management of water resources.

Assessment of groundwater quality variation characteristics and influencing factors in an intensified agricultural area: An integrated hydrochemical and machine learning approach

The decline in groundwater quality in intensive agricultural areas in recent years, driven by environmental change and intensified human activity, poses a significant threat to agricultural production and public health, requiring attention and effective management. However, distinguishing the specific impacts of various factors on groundwater quality remains challenging, which hinders the effective management and prevention of groundwater pollution. This research integrates a hydrochemical analysis with the Entropy-weighted Water Quality Index, Self-Organizing Map (SOM) approach, and Boruta algorithm to investigate groundwater chemical variations and their influencing factors in the Sanjiang Plain, an important grain-producing region in China. The findings reveal that, compared to 2012, the deep groundwater quality has improved, while the shallow groundwater quality has markedly deteriorated. This decline in shallow groundwater quality is primarily attributable to human activities and is characterized by elevated levels of chloride, sulfate, and nitrate and a shift in the groundwater hydrochemical facies from an HCO3-Ca·Mg type to a mixed HCO3-Ca·Mg and SO4·Cl-Ca·Mg type. The SOM results suggested that land use type significantly affects shallow groundwater quality. Further analysis with the Boruta algorithm identified increased sewage and manure emissions from expanding livestock operations as well as enhanced pollutant leakage from the expansion of paddy fields as the primary contributors to the decline in shallow groundwater quality. These findings offer new insights into the mechanisms of groundwater quality changes in agriculturally intensive regions and provide a foundation for improved groundwater pollution management in the Sanjiang Plain and similar areas.

Predicting groundwater phosphate levels in coastal multi-aquifers: A geostatistical and data-driven approach

The groundwater (GW) resource plays a central role in securing water supply in the coastal region of Bangladesh and therefore the future sustainability of this valuable resource is crucial for the area. However, there is limited research on the driving factors and prediction of phosphate concentration in groundwater. In this work, geostatistical modeling, self-organizing maps (SOM) and data-driven algorithms were combined to determine the driving factors and predict GW phosphate content in coastal multi-aquifers in southern Bangladesh. The SOM analysis identified three distinct spatial patterns: K+Na+pH, Ca2+Mg2+NO₃-, and HCO₃-SO₄2-PO43-F-. Four data-driven algorithms, including CatBoost, Gradient Boosting Machine (GBM), Long Short-Term Memory (LSTM), and Support Vector Regression (SVR) were used to predict phosphate concentration in GW using 380 samples and 15 prediction parameters. Forecasting accuracy was evaluated using RMSE, R2, RAE, CC, and MAE. Phosphate dissolution and saltwater intrusion, along with phosphorus fertilizers, increase PO43- content in GW. Using input parameters selected by multicollinearity and SOM, the CatBoost model showed exceptional performance in both training (RMSE = 0.002, MAE = 0.001, R2 = 0.999, RAE = 0.057, CC = 1.00) and testing (RMSE = 0.001, MAE = 0.002, R2 = 0.989, RAE = 0.057, CC = 0.998). Na+, K+, and Mg2+ significantly influenced prediction accuracy. The uncertainty study revealed a low standard error for the CatBoost model, indicating robustness and consistency. Semi-variogram models confirmed that the most influential attributes showed weak dependence, suggesting that agricultural runoff increases the heterogeneity of PO43- distribution in GW. These findings are crucial for developing conservation and strategic plans for sustainable utilization of coastal GW resources.

PM10-bound trace elements in pan-European urban atmosphere

Although many studies have discussed the impact of Europe's air quality, very limited research focused on the detailed phenomenology of ambient trace elements (TEs) in PM10 in urban atmosphere. This study compiled long-term (2013-2022) measurements of speciation of ambient urban PM10 from 55 sites of 7 countries (Switzerland, Spain, France, Greece, Italy, Portugal, UK), aiming to elucidate the phenomenology of 20 TEs in PM10 in urban Europe. The monitoring sites comprised urban background (UB, n = 26), traffic (TR, n = 10), industrial (IN, n = 5), suburban background (SUB, n = 7), and rural background (RB, n = 7) types. The sampling campaigns were conducted using standardized protocols to ensure data comparability. In each country, PM10 samples were collected over a fixed period using high-volume air samplers. The analysis encompassed the spatio-temporal distribution of TEs, and relationships between TEs at each site. Results indicated an annual average for the sum of 20 TEs of 90 ± 65 ng/m3, with TR and IN sites exhibiting the highest concentrations (130 ± 66 and 131 ± 80 ng/m3, respectively). Seasonal variability in TEs concentrations, influenced by emission sources and meteorology, revealed significant differences (p < 0.05) across all monitoring sites. Estimation of TE concentrations highlighted distinct ratios between non-carcinogenic and carcinogenic metals, with Zn (40 ± 49 ng/m3), Ti (21 ± 29 ng/m3), and Cu (23 ± 35 ng/m3) dominating non-carcinogenic TEs, while Cr (5 ± 7 ng/m3), and Ni (2 ± 6 ng/m3) were prominent among carcinogenic ones. Correlations between TEs across diverse locations and seasons varied, in agreement with differences in emission sources and meteorological conditions. This study provides valuable insights into TEs in pan-European urban atmosphere, contributing to a comprehensive dataset for future environmental protection policies.

Thoron, radon and microbial community as supportive indicators of seismic activity in groundwater

Earthquakes have a significant impact on groundwater environments as well as human life. However, identifying active and affected zones from seismic events using isotopic and microbial diversity indicators remains a challenging frontier. To validate the applicability of this coupled method for real-time analysis, we analyzed thoron (220Rn), radon (222Rn), microbial community compositions, and hydrochemistry in groundwater samples during the 2017 Pohang earthquake for the first time. We observed the detection of 220Rn in groundwater right before the aftershocks, with a high correlation to 222Rn concentrations. This indicates that 220Rn and 222Rn can serve as reliable seismic indicators for real-time analysis. The microbial data can assist in identifying affected groundwater zones, particularly when real-time detection of 220Rn is not feasible. At the phylum level, Peregrinibacteria and Firmicutes were only found in samples with detected thoron. At the genus level, hydrogen-oxidizing or sulfur-oxidizing bacteria could serve as indicators of active zones. Two statistical analyses, self-organizing map (SOM) and principal component analysis (PCA) using hydrochemical parameters, also correlated with the results from these coupled indicators. This study demonstrates the theoretical and practical applicability of 220Rn, 222Rn, and microbial community compositions as new multi-faceted ecological indicators, whether for real-time analysis or otherwise.

Self-Organizing Maps for Secondary Ion Mass Spectrometry

Secondary ion mass spectrometry (SIMS) is a powerful analytical technique for characterizing the molecular and elemental composition of surfaces. Individual mass spectra can provide information about the mean surface composition, while spatial mapping can elucidate the spatial distributions of molecular species in 2D and 3D with no prior labeling of molecular targets. The data sets produced by SIMS techniques are large and inherently complex, often containing subtle relationships between spatial and molecular features. Machine learning algorithms are well suited to exploring this complexity, making them ideal for data analysis, interpretation, and visualization of SIMS data sets. One such algorithm, the self-organizing map (SOM), is particularly well suited to clustering similar samples and reducing the dimensionality of hyperspectral data sets. Here, we present an introduction to the SOM, a concise mathematical description, and recent examples of its use in SIMS and other related mass spectrometry techniques. These examples demonstrate how SOMs may be used to interpret high volumes of individual mass spectra, imaging, or depth profiling data sets. This review will be useful for specialists in SIMS and other mass spectral techniques seeking to explore self-organizing maps for data analysis.

A regional examination of the footprint of agriculture and urban cover on stream water quality

Freshwater systems in cold regions, including the Laurentian Great Lakes, are threatened by both eutrophication and salinization, due to excess nitrogen (N), phosphorus (P) and chloride (Cl-) delivered in agricultural and urban runoff. However, identifying the relative contribution of urban vs. agricultural development to water quality impairment is challenging in watersheds with mixed land cover, which typify most developed regions. In this study, a self-organizing map (SOM) analysis was used to evaluate the contributions of various forms of land cover to water quality impairment in southern Ontario, a population-dense, yet highly agricultural region in the Laurentian Great Lakes basin where urban expansion and agricultural intensification have been associated with continued water quality impairment. Watersheds were classified into eight spatial clusters, representing four categories of agriculture, one urban, one natural, and two mixed land use clusters. All four agricultural clusters had high nitrate-N concentrations, but levels were especially high in watersheds with extensive corn and soybean cultivation, where exceedances of the 3 mg L-1 water quality objective dramatically increased above a threshold of ‍∼30 % watershed row crop cover. Maximum P concentrations also occurred in the most heavily tile-drained cash crop watersheds, but associations between P and land use were not as clear as for N. The most urbanized watersheds had the highest Cl- concentrations and expansions in urban area were mostly at the expense of surrounding agricultural land cover, which may drive intensification of remaining agricultural lands. Expansions in tile-drained corn and soybean area, often at the expense of mixed, lower intensity agriculture are not unique to this area and suggest that river nitrate-N levels will continue to increase in the future. The SOM approach provides a powerful means of simplifying heterogeneous land cover characteristics that can be associated with water quality patterns and identify problem areas to target management.

Unraveling the hydrogeochemical characteristics and pollution sources of groundwater in an intensive industrial area, East China

It is challenging to interpret hydrogeochemical datasets with complex natural and anthropogenic genesis in intensive industrial areas. This paper elucidates the hydrogeochemical characteristics and pollution sources of groundwater in an industrial park, East China, combining the self-organizing map (SOM), hydrochemical graphs, and correlation analysis. The results show that the total dissolved solids of groundwater range from 73.45 to 997.92 mg/L and can be regarded as freshwater. The pH varies greatly from 6.44 to 9.90, most of samples belonging to weakly acidic-weakly alkaline. The groundwater can be classified into five clusters by SOM, representing the non- or least-polluted groundwater (cluster D), high salt groundwater (cluster A), high NH4+-N and HCO3- groundwater (cluster B), high Fe and Mn groundwater (cluster C), and high pH groundwater (cluster E), which were contaminated by industrial salts, historical agriculture activity, industrial reducing substances, and industrial alkali, respectively. The natural evolution of groundwater (cluster D) in the study area is mainly controlled by mineral weathering/dissolution. The contributions of calcite, dolomite, gypsum, halite, and silicate mineral to groundwater solute are 55.8-66.3%, 15.1-18.0%, 9.0-10.7%, 2.5-10.1%, and 2.3-9.4%, respectively, based on the mass conservation. The contaminated groundwaters (all other clusters except for cluster D) have different hydrochemical characteristics associated with the pollution sources. In addition, the relatively reductive environment in quaternary flu-lacustrine sediments favored the formation of high level of Fe, Mn, and NH4+-N in groundwater. This study provides a new insight into the characteristic contaminants and their distributions in groundwater and the associated pollution sources based on the large datasets in an intensive industrial area. The data evaluation methods and results of this study could be useful to the groundwater usage management and pollution control in this area and other industrial areas.

Biopsychosocial phenotypes in people with HIV in the CHARTER cohort

Neuropsychiatric complications such as neurocognitive impairment and depression are common in people with HIV despite viral suppression on antiretroviral therapy, but these conditions are heterogeneous in their clinical presentations and associated disability. Identifying novel biopsychosocial phenotypes that account for neurocognitive performance and depressive and functional symptoms will better reflect the complexities encountered in clinical practice and may have pathological and therapeutic implications. We classified 1580 people with HIV based on 17 features, including 7 cognitive domains, 4 subscales of the Beck depression inventory-II, 5 components of the patient's assessment of own functioning inventory, and dependence in instrumental and basic activities of daily living. A two-stage clustering procedure consisting of dimension reduction with self-organizing maps and Mahalanobis distance-based k-means clustering algorithms was applied to cross-sectional data. Baseline demographic and clinical characteristics were compared between the phenotypes, and their prediction on the biopsychosocial phenotypes was evaluated using multinomial logistic regression. Four distinct phenotypes were identified. Participants in Phenotype 1 overall did well in all domains. Phenotype 2 had mild-to-moderate depressive symptoms and the most substance use disorders. Phenotype 3 had mild-to-moderate cognitive impairment, moderate depressive symptoms, and the worst daily functioning; they also had the highest proportion of females and non-HIV conditions that could affect cognition. Phenotype 4 had mild-to-moderate cognitive impairment but with relatively good mood, and daily functioning. Multivariable analysis showed that demographic characteristics, medical conditions, lifetime cocaine use disorder, triglycerides, and non-antiretroviral therapy medications were important variables associated with biopsychosocial phenotype. We found complex, multidimensional biopsychosocial profiles in people with HIV that were associated with different risk patterns. Future longitudinal work should determine the stability of these phenotypes, assess factors that influence transitions from one phenotype to another, and characterize their biological associations.

Groundwater hydrochemistry, source identification and health assessment based on self-organizing map in an intensive mining area in Shanxi, China

The Changzhi Basin in Shanxi is renowned for its extensive mining activities. It's crucial to comprehend the spatial distribution and geochemical factors influencing its water quality to uphold water security and safeguard the ecosystem. However, the complexity inherent in hydrogeochemical data presents challenges for linear data analysis methods. This study utilizes a combined approach of self-organizing maps (SOM) and K-means clustering to investigate the hydrogeochemical sources of shallow groundwater in the Changzhi Basin and the associated human health risks. The results showed that the groundwater chemical characteristics were categorized into 48 neurons grouped into six clusters (C1-C6) representing different groundwater types with different contamination characteristics. C1, C3, and C5 represent uncontaminated or minimally contaminated groundwater (Ca-HCO3 type), while C2 signifies mixed-contaminated groundwater (HCO3-Ca type, Mixed Cl-Mg-Ca type, and CaSO4 type). C4 samples exhibit impacts from agricultural activities (Mixed Cl-Mg-Ca), and C6 reflects high Ca and NO3- groundwater. Anthropogenic activities, especially agriculture, have resulted in elevated NO3- levels in shallow groundwater. Notably, heightened non-carcinogenic risks linked to NO3-, Pb, F-, and Mn exposure through drinking water, particularly impacting children, warrant significant attention. This research contributes valuable insights into sustainable groundwater resource development, pollution mitigation strategies, and effective ecosystem protection within intensive mining regions like the Changzhi Basin. It serves as a vital reference for similar areas worldwide, offering guidance for groundwater management, pollution prevention, and control.

Exploring the Association Between Structural Racism and Mental Health: Geospatial and Machine Learning Analysis

BACKGROUND

Structural racism produces mental health disparities. While studies have examined the impact of individual factors such as poverty and education, the collective contribution of these elements, as manifestations of structural racism, has been less explored. Milwaukee County, Wisconsin, with its racial and socioeconomic diversity, provides a unique context for this multifactorial investigation.

OBJECTIVE

This research aimed to delineate the association between structural racism and mental health disparities in Milwaukee County, using a combination of geospatial and deep learning techniques. We used secondary data sets where all data were aggregated and anonymized before being released by federal agencies.

METHODS

We compiled 217 georeferenced explanatory variables across domains, initially deliberately excluding race-based factors to focus on nonracial determinants. This approach was designed to reveal the underlying patterns of risk factors contributing to poor mental health, subsequently reintegrating race to assess the effects of racism quantitatively. The variable selection combined tree-based methods (random forest) and conventional techniques, supported by variance inflation factor and Pearson correlation analysis for multicollinearity mitigation. The geographically weighted random forest model was used to investigate spatial heterogeneity and dependence. Self-organizing maps, combined with K-means clustering, were used to analyze data from Milwaukee communities, focusing on quantifying the impact of structural racism on the prevalence of poor mental health.

RESULTS

While 12 influential factors collectively accounted for 95.11% of the variability in mental health across communities, the top 6 factors-smoking, poverty, insufficient sleep, lack of health insurance, employment, and age-were particularly impactful. Predominantly, African American neighborhoods were disproportionately affected, which is 2.23 times more likely to encounter high-risk clusters for poor mental health.

CONCLUSIONS

The findings demonstrate that structural racism shapes mental health disparities, with Black community members disproportionately impacted. The multifaceted methodological approach underscores the value of integrating geospatial analysis and deep learning to understand complex social determinants of mental health. These insights highlight the need for targeted interventions, addressing both individual and systemic factors to mitigate mental health disparities rooted in structural racism.

Prediction of sulfate concentrations in groundwater in areas with complex hydrogeological conditions based on machine learning

The persistent and increasing levels of sulfate due to a variety of human activities over the last decades present a widely concerning environmental issue. Understanding the controlling factors of groundwater sulfate and predicting sulfate concentration is critical for governments or managers to provide information on groundwater protection. In this study, the integration of self-organizing map (SOM) approach and machine learning (ML) modeling offers the potential to determine the factors and predict sulfate concentrations in the Huaibei Plain, where groundwater is enriched with sulfate and the areas have complex hydrogeological conditions. The SOM calculation was used to illustrate groundwater hydrochemistry and analyze the correlations among the hydrochemical parameters. Three ML algorithms including random forest (RF), support vector machine (SVM), and back propagation neural network (BPNN) were adopted to predict sulfate levels in groundwater by using 501 groundwater samples and 8 predictor variables. The prediction performance was evaluated through statistical metrics (R2, MSE and MAE). Mine drainage mainly facilitated increase in groundwater SO42- while gypsum dissolution and pyrite oxidation were found another two potential sources. The major water chemistry type was Ca-HCO3. The dominant cation was Na+ while the dominant anion was HCO3-. There was an intuitive correlation between groundwater sulfate and total dissolved solids (TDS), Cl-, and Na+. By using input variables identified by the SOM method, the evaluation results of ML algorithms showed that the R2, MSE and MAE of RF, SVM, BPNN were 0.43-0.70, 0.16-0.49 and 0.25-0.44. Overall, BPNN showed the best prediction performance and had higher R2 values and lower error indices. TDS and Na+ had a high contribution to the prediction accuracy. These findings are crucial for developing groundwater protection and remediation policies, enabling more sustainable management.

Ecological amplitude and indication potential of mining bees (Andrena spp.): a case study from the post-agricultural area of the Kampinos National Park (Poland)

The mining bee (Andrena spp.) play a key role in ensuring plant and animal diversity. The present study examines their diversity in a post-agricultural landscape exemplified by the Kampinos National Park (KNP), a UNESCO Biosphere Reserve in Poland. The following hypotheses were addressed: (H1) the mining bees demonstrate a narrow ecological amplitude, (H2) there are no indicator species for particular habitats, and (H3) the studied mining bees have the same ecological preferences to those presented in the literature. A total of 40 catch per unit effort samples (CPUE) were collected across various habitats with different soil humidity. Forty-six species were recorded, representing 46% of mining bees and approximately 10% of the known Polish bee fauna. Nineteen of the recorded species (41%) were assigned to CR-NT threat categories, indicating that the national park plays a significant role in preserving mining bee species diversity and their conservation. None of the hypotheses (H1, H2, H3) were confirmed. The mining bees were found to demonstrate a wide ecological amplitude. Surprisingly, habitats located in dry and wet soils were both characterised by high abundance and species richness. Seventeen indicators were distinguished among the dominant and rarer species. Our findings suggest that Andrena nigroaenea and A. ventralis (lower humidity), as well as A. alfkenella and A. minutuloides (higher humidity), have different significant relationships with habitat soil humidity to those reported in the literature.

Spatial response of water level and quality shows more significant heterogeneity during dry seasons in large river-connected lakes

The spatial response mechanism of hydrology and water quality of large river-connected lakes is very complicated. In this study, we developed a spatial response analysis method that couples wavelet correlation analysis (WTC) with self-organizing maps (SOM), revealing the spatial response and variation of water level and water quality in Poyang Lake, China's largest river-connected lake, over the past decade. The results show that: (1) there was significant spatial heterogeneity in water level and quality during the dry seasons (2010-2018) compared to other hydrological stages. (2) We identified a more pronounced difference in response of water level and quality between northern and southern parts of Poyang Lake. As the distance increases from the northern lake outlet, the impact of rising water levels on water quality deterioration intensified during the dry seasons. (3) The complex spatial heterogeneity of water level and quality response in the dry seasons is primarily influenced by water level fluctuations from the northern region and the cumulative pollutant entering the lake from the south, which particularly leads to the reversal of the response in the central area of Poyang Lake. The results of this study can contribute to scientific decision-making regarding water environment zoning management in large river-connected lakes amidst complex environment conditions.

WITHDRAWN: Trace metals translocation from soil to plants: Health risk assessment via consumption of vegetables in the urban sprawl of a developing country

This article has been withdrawn at the request of the author(s) and/or editor. The Publisher apologizes for any inconvenience this may cause. The full Elsevier Policy on Article Withdrawal can be found at https://www.elsevier.com/about/policies/article-withdrawal.

Evaluation of photobiomodulation therapy (PBMT) on salivary flow and composition in head and neck cancer patients undergoing radiation therapy

OBJECTIVE

Assess the impact of photobiomodulation therapy (PBMT) on xerostomia, salivary flow rate (SFR) and composition in patients undergoing radiotherapy (RT) for head and neck cancer (HNC).

STUDY DESIGN

Thirty patients undergoing RT (65 Gy) for HNC were enrolled. Saliva and xerostomia evaluations collected pre- and post-PBMT-RT. PBMT involved irradiation of extra and intraoral points, 15-20 sessions, 2-3 times/week. SFR, trace elements, total protein, alkaline phosphatase, xerostomia, and pH were analyzed.

RESULTS

The average age was 60.7 years. After treatment, there was not a significant reduction in SFR and there was no difference on xerostomia. Significant reductions in Al, Cd, Fe, Ni, P, and Sb concentrations were observed, along with a significant increase in Mg concentration. Sample data were organized into 3 groups based on a self-organizing map. Low concentrations of Al, As, Co, Cr, Cu, Fe, Mn, Mo, S, Sr, and Zn were the primary discriminatory factors for group A, while group B consisted of post-PBMT-RT samples with high concentrations of Ca, K, Mg, Na, and S.

CONCLUSIONS

PBMT prevented a significant reduction in SFR and xerostomia induced by radiation therapy. These findings suggest that PBMT prevents salivary gland damage minimizing the decline in salivary flow.

A global clustering of terrestrial food production systems

Food production is at the heart of global sustainability challenges, with unsustainable practices being a major driver of biodiversity loss, emissions and land degradation. The concept of foodscapes, defined as the characteristics of food production along biophysical and socio-economic gradients, could be a way addressing those challenges. By identifying homologues foodscapes classes possible interventions and leverage points for more sustainable agriculture could be identified. Here we provide a globally consistent approximation of the world's foodscape classes. We integrate global data on biophysical and socio-economic factors to identify a minimum set of emergent clusters and evaluate their characteristics, vulnerabilities and risks with regards to global change factors. Overall, we find food production globally to be highly concentrated in a few areas. Worryingly, we find particularly intensively cultivated or irrigated foodscape classes to be under considerable climatic and degradation risks. Our work can serve as baseline for global-scale zoning and gap analyses, while also revealing homologous areas for possible agricultural interventions.

Investigating the flexural behavior of nanomodified multi-delaminated composites using acoustic emission technique

The formation of multiple delaminations is a frequently observed damage mechanism in composite materials, exerting a more pronounced influence on their strength properties compared to single delaminations. To tackle this issue, the incorporation of nanoparticles has been investigated as a means to enhance composite materials. This study aims to examine the effects of nano-additives, specifically carbon nanotubes and nanosilica, on the flexural behavior of glass/epoxy composites containing multiple embedded delaminations. The acoustic emission technique is employed to gain deeper insights into the damage mechanisms associated with flexural failure. Artificial delaminations of varying sizes, arranged in a triangular pattern, were introduced into four interlayers of a [(0/90)2]s oriented glass/epoxy composite. The findings reveal a notable reduction in flexural properties due to the presence of multiple delaminations. However, the addition of nanoparticles demonstrates a significant improvement in the flexural behavior of the multi-delaminated specimens. The most substantial enhancement is observed in the composite incorporating 0.3 wt% nanosilica + 0.5 wt% carbon nanotubes. Furthermore, genetic K-means and hierarchical clustering techniques are employed to classify different damage mechanisms based on the peak frequency and amplitude of the acoustic emission signals. The results indicate that the hierarchical clustering method outperforms the genetic K-means method in accurately clustering the acoustic emission signals. Moreover, the incorporation of nanoparticles' impact on the occurrence of distinct damage mechanisms is evaluated through the analysis of acoustic signals using Wavelet Packet Transform. By investigating the flexural behavior of nanomodified multi-delaminated composites and employing the acoustic emission technique, this study offers valuable insights into the role of nanoparticles in enhancing the mechanical properties and monitoring the damage mechanisms of composite materials.

An efficient and straightforward online vector quantization method for a data stream through remove-birth updating

The growth of network-connected devices has led to an exponential increase in data generation, creating significant challenges for efficient data analysis. This data is generated continuously, creating a dynamic flow known as a data stream. The characteristics of a data stream may change dynamically, and this change is known as concept drift. Consequently, a method for handling data streams must efficiently reduce their volume while dynamically adapting to these changing characteristics. This article proposes a simple online vector quantization method for concept drift. The proposed method identifies and replaces units with low win probability through remove-birth updating, thus achieving a rapid adaptation to concept drift. Furthermore, the results of this study show that the proposed method can generate minimal dead units even in the presence of concept drift. This study also suggests that some metrics calculated from the proposed method will be helpful for drift detection.

Vascular plants and mosses as bioindicators of variability of the coastal pine forest (Empetro nigri-Pinetum)

Empetro nigri-Pinetum is a unique sea coast plant community developing along the Baltic Sea from Germany to Lithuania. Our detailed field research of bryophytes and vascular plants has highlighted the regional diversity of the Empetro nigri-Pinetum typicum plant community throughout its range in Central Europe. Our study indicated that vascular plants and mosses effectively discriminate against the described phytocoenoses, thus it was possible to distinguish three variants of the coastal forest: Calluna-Deschampsia (from Germany), Vaccinium vitis-idaea (from Poland) and Melampyrum-Deschampsia (from Lithuania). Redundancy analysis indicated that the division is related to the habitat conditions of the analyzed areas, with humidity having the greatest impact on this differentiation. Kohonen's artificial neural network (i.e. self-organising map, SOM) confirmed the heterogeneous nature of the studied phytocenoses, and combined with the IndVal index enabled identification of indicator species for respective studied patches: Deschampsia flexuosa for Calluna-Deschampsia group; Aulacomnium palustre, Calluna vulgaris, Carex nigra, Dicranum polysetum, Erica tetralix, Oxycoccus palustris, Sphagnum capillifolium, Vaccinium uliginosum and Vaccinium vitis-idaea for Vaccinium vitis-idaea group; and young specimens of Betula pendula, Lycopodium annotinum, Melampyrum pratense and Orthilia secunda for Melampyrum-Deschampsia group. Thereby, our study showed that individual groups of species can be very good bioindicators for each of the studied phytocoenoses.

Deterministic factors modulating assembly of groundwater microbial community in a nitrogen-contaminated and hydraulically-connected river-lake-floodplain ecosystem

The river-lake-floodplain system (RLFS) undergoes intensive surface-groundwater mass and energy exchanges. Some freshwater lakes are groundwater flow-through systems, serving as sinks for nitrogen (N) entering the lake. Despite the threat of cross-nitrogen contamination, the assembly of the microbial communities in the RLFS was poorly understood. Herein, the distribution, co-occurrence, and assembly pattern of microbial community were investigated in a nitrogen-contaminated and hydraulically-connected RLFS. The results showed that nitrate was widely distributed with greater accumulation on the south than on the north side, and ammonia was accumulated in the groundwater discharge area (estuary and lakeshore). The heterotrophic nitrifying bacteria and aerobic denitrifying bacteria were distributed across the entire area. In estuary and lakeshore with low levels of oxidation-reduction potential (ORP) and high levels of total organic carbon (TOC) and ammonia, dissimilatory nitrate reduction to ammonium (DNRA) bacteria were enriched. The bacterial community had close cooperative relationships, and keystone taxa harbored nitrate reduction potentials. Combined with multivariable statistics and self-organizing map (SOM) results, ammonia, TOC, and ORP acted as drivers in the spatial evolution of the bacterial community, coincidence with the predominant deterministic processes and unique niche breadth for microbial assembly. This study provides novel insight into the traits and assembly of bacterial communities and potential nitrogen cycling capacities in RLFS groundwater.

Disentangling Jenny's equation by machine learning

The so-called soil-landscape model is the central paradigm which relates soil types to their forming factors through the visionary Jenny's equation. This is a formal mathematical expression that would permit to infer which soil should be found in a specific geographical location if the involved relationship was sufficiently known. Unfortunately, Jenny's is only a conceptual expression, where the intervening variables are of qualitative nature, not being then possible to work it out with standard mathematical tools. In this work, we take a first step to unlock this expression, showing how Machine Learning can be used to predictably relate soil types and environmental factors. Our method outperforms other conventional statistical analyses that can be carried out on the same forming factors defined by measurable environmental variables.

Geological context and human exposures to element mixtures in mining and agricultural settings in Colombia

Anthropogenic and natural sources contribute to chemical mixtures in air, water, and soil, posing potential risks to the environment and human health. To understand the interplay between element profiles in the human body, geographical location, and associated economic activities, we carried out an observational analytic cross-sectional study. The study recruited 199 participants from three municipalities, two of which had gold-mining as their primary economic activity, while the other was dedicated to agricultural and other local activities not related to mining. The concentrations of a total of 30 elements in human hair samples and 21 elements in environmental soil samples were measured using various spectrometry techniques. Unsupervised clustering analysis using Self-Organizing Maps was applied to human hair samples to analyze element concentrations. Distinct clusters of individuals were identified based on their hair element profiles, which were mapped to geographical location and economic activities. While higher levels of heavy metals (Ag, As, Hg, and Pb) were observed in individuals engaged in mining activities in certain clusters, individuals in agricultural areas show higher concentrations of elements found in pesticides (Ba and Sr). However, the elemental composition of hair is influenced not only by the anthropogenic activities but also by the inherent geological context where people live. Our findings highlight the significance of accounting for environmental factors when evaluating human health risks, as the intricate mixture of elements can yield valuable insights for targeted health interventions.

A Survey of Hardware Self-Organizing Maps

Self-organizing feature maps (SOMs) are commonly used technique for clustering and data dimensionality reduction in many application fields. Indeed, their inherent property of topology preservation and unsupervised learning of processed data without any prior knowledge put them in the front of candidates for data reduction in the Internet of Things (IoT) and big data (BD) technologies. However, the high computational cost of SOMs limits their use to offline approaches and makes the online real-time high-performance SOM processing more challenging and mostly reserved to specific hardware implementations. In this article, we present a survey of hardware (HW) SOM implementations found in the literature so far: the most widely used computing blocks, architectures, design choices, adaptation, and optimization techniques that have been reported in the field of hardware SOMs. Moreover, we give an overview of main challenges and trends for their ubiquitous adoption as hardware accelerators in many application fields. This article is expected to be useful for researchers in the areas of artificial intelligence, hardware architecture, and system design.

A comparative study of predicting the availability of power line communication nodes using machine learning

Power Line Communication technology uses power cables to transmit data. Knowing whether a node is working in advance without testing saves time and resources, leading to the proposed model. The model has been trained on three dominant features, which are SNR (Signal to Noise Ratio), RSSI (Received Signal Strength Indicator), and CINR (Carrier to Interference plus Noise Ratio). The dataset consisted of 1000 readings, with 90% in the training set and 10% in the testing set. In addition, 50% of the dataset is for class 1, which indicates whether the node readings are optimum. The model is trained with multi-layer perception, K-Nearest Neighbors, Support Vector Machine with linear and non-linear kernels, Random Forest, and adaptive boosting (ADA) algorithms to compare between statistical, vector-based, regression, decision, and predictive algorithms. ADA boost has achieved the best accuracy, F-score, precision, and recall, which are 87%, 0.86613, 0.9, 0.8646, respectively.

Seasonal effects on hydrochemistry, microbial diversity, and human health risks in radon-contaminated groundwater areas

Groundwater is an important human resource. Daejeon in South Korea faces severe water quality issues, including radon, uranium, and fluoride pollution, all of which pose health risks to humans. With climate change, threats to potable water, such as heavy rain and typhoons, have become common. Therefore, examining the seasonal effects on groundwater quality and resultant health risks is important for understanding the mechanisms of different hydroclimatological conditions to enable the implementation of sustainable management plans in radon-contaminated groundwater areas. However, this issue has not yet been studied. To bridge this gap, in this study, major ions and microbial community structures were employed and groundwater quality index (GWQI) were calculated with hazard index based on limits set by the World Health Organization (WHO) to investigate the hydrochemical characterization and to assess pollution levels. The results showed that the rainy season had distinct hydrochemical characteristics with high correlations between radon and fluoride, and most groundwater samples collected after the typhoon had characteristics similar to those collected during the dry season, owing to the flow path. Furthermore, the microbial diversity and hazard quotient (HQ) values of fluoride revealed that pollution worsened during the dry season. All of the calculated effective dose values of radon exceeded the threshold limit set by the WHO, despite the low GWQI. Infants and children were particularly susceptible to radon-contaminated groundwater. The statistical results of self-organizing map (SOM) suggested that radon analysis was sufficient for public health intervention in the rainy season; however, in the dry season, combined analyses of radon, fluoride, and microbial diversity played important roles in health risk assessment. Our study presents a comprehensive understanding of radon-contaminated groundwater characteristics under seasonal effects and can serve as a reference for other similar zones to provide significant insights into the effective management of radon contamination.

Applying Artificial Neural Networks to Oxidative Stress Biomarkers in Forager Honey Bees (Apis mellifera) for Ecological Assessment

Insect pollinators provide an important ecosystem service that supports global biodiversity and environmental health. The study investigates the effects of the environmental matrix on six oxidative stress biomarkers in the honey bee Apis mellifera. Thirty-five apiaries located in urban, forested, and agricultural areas in Central Italy were sampled during the summer season. Enzyme activities in forager bees were analyzed using an artificial neural network, allowing the identification and representation of the apiary patterns in a Self-Organizing Map. The SOM nodes were correlated with the environmental parameters and tissue levels of eight heavy metals. The results indicated that the apiaries were not clustered according to their spatial distribution. Superoxide dismutase expressed a positive correlation with Cr and Mn concentrations; catalase with Zn, Mn, Fe, and daily maximum air temperature; glutathione S-transferase with Cr, Fe, and daily maximal air temperature; and glutathione reductase showed a negative correlation to Ni and Fe exposure. This study highlights the importance of exploring how environmental stressors affect these insects and the role of oxidative stress biomarkers. Artificial neural networks proved to be a powerful approach to untangle the complex relationships between the environment and oxidative stress biomarkers in honey bees. The application of SOM modeling offers a valuable means of assessing the potential effects of environmental pressures on honey bee populations.

Potential Global Invasion Risk of Scale Insect Pests Based on a Self-Organizing Map

In the present study, a global presence/absence dataset including 2486 scale insect species in 157 countries was extracted to assess the establishment risk of potential invasive species based on a self-organizing map (SOM). According to the similarities in species assemblages, a risk list of scale insects for each country was generated. Meanwhile, all countries in the dataset were divided into five clusters, each of which has high similarities of species assemblages. For those countries in the same neuron of the SOM output, they may pose the greatest threats to each other as the sources of potential invasive scale insect species, and therefore, require more attention from quarantine departments. In addition, normalized ζ_i values were used to measure the uncertainty of the SOM output. In total, 9 out of 63 neurons obtained high uncertainty with very low species counts, indicating that more investigation of scale insects should be undertaken in some parts of Africa, Asia and Northern Europe.

Self-organizing map algorithm for assessing spatial and temporal patterns of pollutants in environmental compartments: A review

The evaluation of the spatial and temporal distribution of pollutants is a crucial issue to assess the anthropogenic burden on the environment. Numerous chemometric approaches are available for data exploration and they have been applied for environmental health assessment purposes. Among the unsupervised methods, Self-Organizing Map (SOM) is an artificial neural network able to handle non-linear problems that can be used for exploratory data analysis, pattern recognition, and variable relationship assessment. Much more interpretation ability is gained when the SOM-based model is merged with clustering algorithms. This review comprises: (i) a description of the algorithm operation principle with a focus on the key parameters used for the SOM initialization; (ii) a description of the SOM output features and how they can be used for data mining; (iii) a list of available software tools for performing calculations; (iv) an overview of the SOM application for obtaining spatial and temporal pollution patterns in the environmental compartments with focus on model training and result visualization; (v) advice on reporting SOM model details in a paper to attain comparability and reproducibility among published papers as well as advice for extracting valuable information from the model results is presented.

Spatial distribution and driving factors of groundwater chemistry and pollution in an oil production region in the Northwest China

Concerns have been raised on the deterioration of groundwater quality associated with anthropogenic impacts such as oil extraction and overuse of fertilizers. However, it is still difficult to identify groundwater chemistry/pollution and driving forces in regional scale since both natural and anthropogenic factors are spatially complex. This study, combining self-organizing map (SOM, combined with K-means algorithm) and principal component analysis (PCA), attempted to characterize the spatial variability and driving factors of shallow groundwater hydrochemistry in Yan'an area of Northwest China where diverse land use types (e.g., various oil production sites and agriculture lands) coexist. Based on the major and trace elements (e.g., Ba, Sr, Br, Li) and total petroleum hydrocarbons (TPH), groundwater samples were classified into four clusters with obvious geographical and hydrochemical characteristics by using SOM - K-means clustering: heavily oil-contaminated groundwater (Cluster 1), slightly oil-contaminated groundwater (Cluster 2), least-polluted groundwater (Cluster 3) and NO₃^- contaminated groundwater (Cluster 4). Noteworthily, Cluster 1, located in a river valley with long-term oil exploitation, had the highest levels of TPH and potentially toxic elements (Ba, Sr). Multivariate analysis combined with ion ratios analysis were used to determine the causes of these clusters. The results revealed that the hydrochemical compositions in Cluster 1 were mainly caused by the oil-related produced water intrusion into the upper aquifer. The elevated NO₃^- concentrations in Cluster 4 were induced by agricultural activities. Water-rock interactions (e.g., carbonate as well as silicate dissolution and precipitation) also shaped the chemical constituents of groundwater in clusters 2, 3, and 4. In addition, SO₄^2--related processes (redox, precipitation of sulfate minerals) also affected groundwater chemical compositions in Cluster 1. This work provides the insight into the driving factors of groundwater chemistry and pollution which could contribute to groundwater sustainable management and protection in this area and other oil extraction areas.

Picoplankton diversity in an oligotrophic and high salinity environment in the central Adriatic Sea

By combining qualitative 16S metabarcoding and quantitative CARD-FISH methods with neural gas analysis, different patterns of the picoplankton community were revealed at finer taxonomic levels in response to changing environmental conditions in the Adriatic Sea. We present the results of a one-year study carried out in an oligotrophic environment where increased salinity was recently observed. We have shown that the initial state of community structure changes according to environmental conditions and is expressed as qualitative and quantitative changes. A general pattern of increasing diversity under harsh environmental conditions, particularly under the influence of increasing salinity at the expense of community abundance was observed. Considering the trend of changing seawater characteristics due to climate change, this study helps in understanding a possible structural change in the microbial community of the Adriatic Sea that could affect higher levels of the marine food web.

Both natural and anthropogenic factors control surface water and groundwater chemistry and quality in the Ningtiaota coalfield of Ordos Basin, Northwestern China

An understanding of the vertical variations in hydrogeochemical processes in various aquifers and quality suitability assessment is crucial for the utilization of groundwater in the Ningtiaota coalfield of Ordos Basin, Northwestern China. Based on 39 water samples collected from surface water (SW), Quaternary pore water (QW), weathered fissure water (WW), and mine water (MW), we conducted self-organizing maps (SOM) algorithm, multivariate statistical analysis (MSA), and classical graphical methods to elucidate the mechanisms controlling the vertical spatial variations in SW and groundwater chemistry and conducted a health risk assessment. The findings indicated that the hydrogeochemical type showed a transition from the HCO₃^--Na⁺ type in SW to the HCO₃^--Ca²⁺ type in QW, then to the SO₄^2--Mg²⁺ type in WW, and back to HCO₃^--Na⁺ type in MW. Water-rock interaction, silicate dissolution, and cation exchange were the main hydrogeochemical processes in the study area. Additionally, groundwater residence time and mining operations were critical external factors that affect water chemistry. Contrary to phreatic aquifers, confined aquifers featured greater circulation depth, water-rock interactions, and external interventions leading to worse quality and higher health risks. Water quality surrounding the coalfield was poor, causing it to be undrinkable, with excessive SO₄^2-, arsenic (As), and F^-, etc. Approximately 61.54% of SW, all of QW, 75% of WW, and 35.71% of MW can be used for irrigation.

Potential risks, source apportionment, and health risk assessment of dissolved heavy metals in Zhoushan fishing ground, China

Dissolved heavy metal pollution in the ocean is one of the most severe environmental concerns; however, the potential sources of heavy metals and the resulting health risks are not fully understood. To explore the distribution characteristics, source apportionment, and health risks of dissolved heavy metals (As, Cd, Cu, Hg, Pb, and Zn) in the Zhoushan fishing ground, this study analyzed heavy metals in surface seawater during the wet and dry seasons. The concentrations of heavy metals varied greatly between seasons, and the mean concentration in the wet season was generally higher than that in the dry season. A positive matrix factorization model coupled with correlation analysis was applied to identify promising sources of heavy metals. Four potential sources (agricultural, industrial, traffic, atmospheric deposition, and natural sources) were identified as the determinants of the accumulation of heavy metals. The health risk assessment results revealed that non-carcinogenic risk (NCR) for adults and children were acceptable (HI < 1), and carcinogenic risk (CR) were at a low level (1 × 10^-6 < TCR ≤ 1 × 10^-4). The source-oriented risk assessment indicated that industrial and traffic sources were the main sources of pollution, contributing 40.7 % of NCR and 27.4 % of CR, respectively. This study proposes forming reasonable, effective policies to control industrial pollution and improve the ecological environment of Zhoushan fishing grounds.

Both fine-grained and coarse-grained spatial patterns of neural activity measured by functional MRI show preferential encoding of pain in the human brain

How pain emerges from human brain remains an unresolved question in pain neuroscience. Neuroimaging studies have suggested that all brain areas activated by painful stimuli were also activated by tactile stimuli, and vice versa. Nonetheless, pain-preferential spatial patterns of voxel-level activation in the brain have been observed when distinguishing painful and tactile brain activations using multivariate pattern analysis (MVPA). According to two hypotheses, the neural activity pattern preferentially encoding pain could exist at a global, coarse-grained, regional level, corresponding to the "pain connectome" hypothesis proposing that pain-preferential information may be encoded by the synchronized activity across multiple distant brain regions, and/or exist at a local, fine-grained, voxel level, corresponding to the "intermingled specialized/preferential neurons" hypothesis proposing that neurons responding specially or preferentially to pain could be present and intermingled with non-pain neurons within a voxel. Here, we systematically investigated the spatial scales of pain-distinguishing information in the human brain measured by fMRI using machine learning techniques, and found that pain-distinguishing information could be detected at both coarse-grained spatial scales across widely distributed brain regions and fine-grained spatial scales within many local areas. Importantly, the spatial distribution of pain-distinguishing information in the brain varies across individuals and such inter-individual variations may be related to a person's trait about pain perception, particularly the pain vigilance and awareness. These results provide new insights into the long-standing question of how pain is represented in the human brain and help the identification of characteristic neuroimaging measurements of pain.

Natural-human driving factors of groundwater salinization in a long-term irrigation area

Salinization of groundwater is a major challenge for groundwater management in long-term irrigation areas, decoupling its complex influencing factors can provide insights for the sustainable development of irrigation areas. In this study, the natural-human driving factors of groundwater salinization in the Yinchuan Plain, a typical irrigated area, were identified using isotope analysis, information entropy, and self-organizing map. Results show that groundwater in the study area is seriously salinized with obvious spatial heterogeneity. Multiple natural conditions and frequent human activities complicate the salinization characteristics of groundwater. On this basis, four typical natural influence units of groundwater were identified, namely, an evaporation and upward leakage zone, a runoff zone, an evaporation zone, and a runoff and upward leakage zone. Information entropy was proposed to quantify the complexity of groundwater resulting from human activities: The complexity difference between densely populated areas and natural dominant areas is mainly reflected in Na⁺, SO₄^2-, and Cl^-. Multiple human-made drivers of complex water environment were further separated into three patterns by the SOM model: blockage-evaporation type, leakage-evaporation type, and irrigation type. The blockage of drainage ditches and obstruction of salt discharge has the highest impact on the salinization of groundwater, followed by irrigation activities and transportation losses. Water excessive stagnation caused by blockage or irrigation is the root cause of groundwater salinization in the irrigated area, and its impact is greater than that of the traditional understanding of groundwater level rise. Based on the evaluation of irrigation water quality, management initiatives for irrigated areas should prioritize dredging and maintaining a healthy soil and groundwater environment in tandem.

Studying Pregnancy Outcome Risk in Patients with Systemic Lupus Erythematosus Based on Cluster Analysis

Background. Pregnancy in systemic lupus erythematosus (SLE) patients is a challenge due to the potential maternal and fetal complications. Therefore, a multidisciplinary assessment of disease risk before and during pregnancy is essential to improve pregnancy outcomes. Objectives. Our purpose was to (i) define clusters of patients with similar history and laboratory features and determine the associative maternal and perinatal outcomes and (ii) evaluate the risk spectrum of maternal and perinatal outcomes of pregnancy in SLE patients, represented by our established risk-assessment chart. Methods. Medical records of 119 patients in China were analyzed retrospectively. Significant variables with $p<0.05$ were selected. The self-organizing map was used for clustering the data based on historical background and laboratory features. Results. Clustering was conducted using 21 maternal and perinatal features. Five clusters were recognized, and their prominent maternal manifestations were as follows: cluster 1 (including 27.73% of all patients): preeclampsia and lupus nephritis; cluster 2 (22.69%): oligohydramnios, uterus scar, and femoral head necrosis; cluster 3 (13.45%): upper respiratory tract infection; cluster 4 (15.97%): premature membrane rupture; and cluster 5 (20.17%): no problem. Conclusion. Pregnancy outcomes in SLE women fell into three categories, namely high risk, moderate risk, and low risk. Present manifestations, besides the medical records, are a potential assessment means for better management of pregnant SLE patients.