Scholarlink

<p>This study investigates fluoride contamination in groundwater, the resulting health risks, and machine-learning-based prediction of groundwater fluoride contamination in Agra City (India). Fluoride concentrations varied from 0.53 to 5.93&#xa0;mg/L (mean 1.97&#xa0;mg/L) and in the majority (58.33%) of cases, levels exceeded acceptable limits as per the Bureau of Indian Standards and World Health Organization. The results reveal that Oral Chronic Daily Intake (CDI) values varied from 0.37 to 14.83&#xa0;mg/day over 48 locations. The average CDI values were 1.38&#xa0;mg/day for infants, 1.53&#xa0;mg/day for children, 3.93&#xa0;mg/day for teenagers and 4.91&#xa0;mg/day for adults at baseline. Risk assessment (non-carcinogenic) using Hazard Quotient (HQ) indicated that HQ mean values for infants were found to be highest (mean HQ = 3.06), followed by children (1.70), teenagers (1.31) and adults (1.05). We noted that the proportion of subjects with HQ values &gt; 1.0 reached 91.66% of infants, 45.83% of children, 41.66% of teenagers and 39.58% of adults, indicating a high risk to health especially in young populations. Four machine learning models were compared to predict fluoride concentrations, including Artificial Neural Network (ANN), Random Forest (RF), Support Vector Regression (SVR) and Extreme Gradient Boosting (XGB). The ANN showed the lowest RMSE (0.035), MSE (0.001), MAE (0.031) and highest R² (0.998), followed by the XGB but reduced performance was also observed using RF compared with the ANN. Results indicated significant differences among models (Friedman test <i>p</i> &lt; 0.05), and post hoc analyses showed that ANN produced greater statistics than other methods. The findings highlight an immediate need for measures to reduce risks and show that artificial neural networks (ANN) effectively predict fluoride concentrations in groundwater.</p> Graphical Abstract <p></p> <p>The graphical abstract demonstrates an integrated framework combining fluoride exposure, assessment of associated health risks, and application of machine learning models for the prediction of health risks due to fluoride exposure in Agra. The outer ring indicates fluoride concentration (0.53–5.93&#xa0;mg/L; mean = 1.97&#xa0;mg/L) for the analyzed samples, where &gt; 50% of the fluoride concentrations exceed the recommended maximum. The inner points (ages 0–Adult) indicate the different population groups used to determine chronic daily intake of fluoride by each population group (infants = 1.38, Children = 2.25, Adolescents = 3.07, and Adults = 4.91&#xa0;mg/day) with increasing chronic daily intake of fluoride exposure as a function of age representing a greater cumulative total of fluoride exposure in an individual over their lifetime. Hazard quotient (HQ) arrows reflect the level of non-carcinogenic risk resulting from daily exposure to fluoride with infants representing the segment of the population with the greatest vulnerability due to most HQ values exceeding safe limits (&gt; 1). The diagram also shows integration of the artificial intelligence models (ANN, SVR, RF and XGBoost) for predictive analysis, along with performance measures supporting the reliability of these model predictions, where artificial neural networks (ANN) provided the best performance. It can be concluded that the information presented in the figure (graphical abstract) demonstrates considerable public health risk and further supports the need for data-driven decision-making relative to fluoride management. A key novelty of this study lies in its integrated framework that simultaneously combines multi-age fluoride health risk assessment with a comparative evaluation of advanced machine learning models, identifying ANN as the most robust tool for accurate prediction and decision support.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Integrated Machine Learning Approach for Fluoride Contamination Prediction and Health Risk Assessment in a Semi-Arid Region of North India

  • 期刊论文

  • 发表日期：

    2026年5月29日

查看全文
  • Shahjad Ali,
  • Jitendra Kumar,
  • Salman Ahmed,
  • Raisul Islam,
  • Rajesh Kumar Deolia,
  • Mohammad Usama,
  • Azhar Shadab,
  • Kamlesh Deshmukh,
  • Jahangeer Jahangeer,
  • Mohammad Muqtada Ali Khan

摘要

This study investigates fluoride contamination in groundwater, the resulting health risks, and machine-learning-based prediction of groundwater fluoride contamination in Agra City (India). Fluoride concentrations varied from 0.53 to 5.93 mg/L (mean 1.97 mg/L) and in the majority (58.33%) of cases, levels exceeded acceptable limits as per the Bureau of Indian Standards and World Health Organization. The results reveal that Oral Chronic Daily Intake (CDI) values varied from 0.37 to 14.83 mg/day over 48 locations. The average CDI values were 1.38 mg/day for infants, 1.53 mg/day for children, 3.93 mg/day for teenagers and 4.91 mg/day for adults at baseline. Risk assessment (non-carcinogenic) using Hazard Quotient (HQ) indicated that HQ mean values for infants were found to be highest (mean HQ = 3.06), followed by children (1.70), teenagers (1.31) and adults (1.05). We noted that the proportion of subjects with HQ values > 1.0 reached 91.66% of infants, 45.83% of children, 41.66% of teenagers and 39.58% of adults, indicating a high risk to health especially in young populations. Four machine learning models were compared to predict fluoride concentrations, including Artificial Neural Network (ANN), Random Forest (RF), Support Vector Regression (SVR) and Extreme Gradient Boosting (XGB). The ANN showed the lowest RMSE (0.035), MSE (0.001), MAE (0.031) and highest R² (0.998), followed by the XGB but reduced performance was also observed using RF compared with the ANN. Results indicated significant differences among models (Friedman test p < 0.05), and post hoc analyses showed that ANN produced greater statistics than other methods. The findings highlight an immediate need for measures to reduce risks and show that artificial neural networks (ANN) effectively predict fluoride concentrations in groundwater.

Graphical Abstract

The graphical abstract demonstrates an integrated framework combining fluoride exposure, assessment of associated health risks, and application of machine learning models for the prediction of health risks due to fluoride exposure in Agra. The outer ring indicates fluoride concentration (0.53–5.93 mg/L; mean = 1.97 mg/L) for the analyzed samples, where > 50% of the fluoride concentrations exceed the recommended maximum. The inner points (ages 0–Adult) indicate the different population groups used to determine chronic daily intake of fluoride by each population group (infants = 1.38, Children = 2.25, Adolescents = 3.07, and Adults = 4.91 mg/day) with increasing chronic daily intake of fluoride exposure as a function of age representing a greater cumulative total of fluoride exposure in an individual over their lifetime. Hazard quotient (HQ) arrows reflect the level of non-carcinogenic risk resulting from daily exposure to fluoride with infants representing the segment of the population with the greatest vulnerability due to most HQ values exceeding safe limits (> 1). The diagram also shows integration of the artificial intelligence models (ANN, SVR, RF and XGBoost) for predictive analysis, along with performance measures supporting the reliability of these model predictions, where artificial neural networks (ANN) provided the best performance. It can be concluded that the information presented in the figure (graphical abstract) demonstrates considerable public health risk and further supports the need for data-driven decision-making relative to fluoride management. A key novelty of this study lies in its integrated framework that simultaneously combines multi-age fluoride health risk assessment with a comparative evaluation of advanced machine learning models, identifying ANN as the most robust tool for accurate prediction and decision support.