Environmental variable importance for under-five mortality in Malaysia: A random forest approach

Random forest analysis of the relative importance of meteorological indicators for heatstroke cases in Japan based on the degree of severity and place of occurrence

Heatstroke is a serious health concern in Japan. To reduce heatstroke risk, the government of Japan implemented the "Heatstroke Alert" nationwide in 2021, employing the wet bulb globe temperature (WBGT) as a criterion. Although the WBGT is a useful meteorological indicator for assessing the risk of heatstroke, other important meteorological indicators must also be investigated. Therefore, using a random forest approach, this study analyzed the relative importance of several meteorological indicators, including those representing heat acclimatization, for each of the 47 Japanese prefectures. Using the generalized linear model, important meteorological indicators were employed as explanatory variables in the heatstroke prediction model to determine the predictive meteorological indicator. Heatstroke cases were evaluated separately by the degree of severity and the place of occurrence. The results showed that the relative temperature (RelTemp), which represents heat acclimatization and was calculated considering past temperature history, was the most predictive (i.e., provided the best goodness of fit) concerning the degree of severity, place of occurrence, and prefectures. RelTemp can be a complementary indicator of WBGT in countries and regions such as Japan, where seasonal differences in heat acclimatization must be considered. In addition, the findings of this study contribute to the development of a more accurate assessment of heatstroke risk.

Effects of Various Heavy Metal Exposures on Insulin Resistance in Non-diabetic Populations: Interpretability Analysis from Machine Learning Modeling Perspective

Increasing and compelling evidence has been proved that heavy metal exposure is involved in the development of insulin resistance (IR). We trained an interpretable predictive machine learning (ML) model for IR in the non-diabetic populations based on levels of heavy metal exposure. A total of 4354 participants from the NHANES (2003-2020) with complete information were randomly divided into a training set and a test set. Twelve ML algorithms, including random forest (RF), XGBoost (XGB), logistic regression (LR), GaussianNB (GNB), ridge regression (RR), support vector machine (SVM), multilayer perceptron (MLP), decision tree (DT), AdaBoost (AB), Gradient Boosting Decision Tree (GBDT), Voting Classifier (VC), and K-Nearest Neighbour (KNN), were constructed for IR prediction using the training set. Among these models, the RF algorithm had the best predictive performance, showing an accuracy of 80.14%, an AUC of 0.856, and an F1 score of 0.74 in the test set. We embedded three interpretable methods, the permutation feature importance analysis, partial dependence plot (PDP), and Shapley additive explanations (SHAP) in RF model for model interpretation. Urinary Ba, urinary Mo, blood Pb, and blood Cd levels were identified as the main influencers of IR. Within a specific range, urinary Ba (0.56-3.56 µg/L) and urinary Mo (1.06-20.25 µg/L) levels exhibited the most pronounced upwards trend with the risk of IR, while blood Pb (0.05-2.81 µg/dL) and blood Cd (0.24-0.65 µg/L) levels showed a declining trend with IR. The findings on the synergistic effects demonstrated that controlling urinary Ba levels might be more crucial for the management of IR. The SHAP decision plot offered personalized care for IR based on heavy metal control. In conclusion, by utilizing interpretable ML approaches, we emphasize the predictive value of heavy metals for IR, especially Ba, Mo, Pb, and Cd.

Identifying cardiovascular disease risk in the U.S. population using environmental volatile organic compounds exposure: A machine learning predictive model based on the SHAP methodology

BACKGROUND

Cardiovascular disease (CVD) remains a leading cause of mortality globally. Environmental pollutants, specifically volatile organic compounds (VOCs), have been identified as significant risk factors. This study aims to develop a machine learning (ML) model to predict CVD risk based on VOC exposure and demographic data using SHapley Additive exPlanations (SHAP) for interpretability.

METHODS

We utilized data from the National Health and Nutrition Examination Survey (NHANES) from 2011 to 2018, comprising 5098 participants. VOC exposure was assessed through 15 urinary metabolite metrics. The dataset was split into a training set (70 %) and a test set (30 %). Six ML models were developed, including Random Forest (RF), Light Gradient Boosting Machine (LightGBM), Decision Tree (DT), Extreme Gradient Boosting (XGBoost), Multi-Layer Perceptron (MLP), and Support Vector Machines (SVM). Model performance was evaluated using the Area Under the Receiver Operating Characteristic Curve (AUROC), accuracy, balanced accuracy, F1 score, J-index, kappa, Matthew's correlation coefficient (MCC), positive predictive value (PPV), negative predictive value (NPV), sensitivity (sens), specificity (spec) and SHAP was applied to interpret the best-performing model.

RESULTS

The RF model exhibited the highest predictive performance with an ROC of 0.8143. SHAP analysis identified age and ATCA as the most significant predictors, with ATCA showing a protective effect against CVD, particularly in older adults and those with hypertension. The study found a significant interaction between ATCA levels and age, indicating that the protective effect of ATCA is more pronounced in older individuals due to increased oxidative stress and inflammatory responses associated with aging. E-values analysis suggested robustness to unmeasured confounders.

CONCLUSIONS

This study is the first to utilize VOC exposure data to construct an ML model for predicting CVD risk. The findings highlight the potential of combining environmental exposure data with demographic information to enhance CVD risk prediction, supporting the development of personalized prevention and intervention strategies.

Construction of metal interpretable scoring system and identification of tungsten as a novel risk factor in COPD

Numerous studies have highlighted the correlation between metal intake and deteriorated pulmonary function, emphasizing its pivotal role in the progression of Chronic Obstructive Pulmonary Disease (COPD). However, the efficacy of traditional models is often compromised due to overfitting and high bias in datasets with low-level exposure, rendering them ineffective in delineating the contemporary risk trends associated with pulmonary diseases. To address these limitations, we embarked on developing advanced, interpretable models, crucial for elucidating the intricate mechanisms of metal toxicity and enriching the domain knowledge embedded in toxicity models. In this endeavor, we scrutinized extensive, long-term metal exposure datasets from NHANES to explore the interplay between metal and pulmonary functionality. Employing a variety of machine-learning approaches, we opted for the "Mixer of Experts" model for its proficiency in identifying a myriad of toxicological trends and sensitivities. We conceptualized and illustrated the TSAP (Toxicity Score at Population-level), a metal interpretable scoring system offering performance nearly equivalent to the amalgamation of standard interpretable methods addressing the "black box" conundrum. This streamlined, bifurcated procedural analysis proved instrumental in discerning established risk factors, thereby uncovering Tungsten as a novel contributor to COPD risk. SYNOPSIS: TSAP achieved satisfied performance with transparent interpretability, suggesting tungsten intake need further action for COPD prevention.

Exploring the relationship between heavy metals and diabetic retinopathy: a machine learning modeling approach

Diabetic retinopathy (DR) is one of the leading causes of adult blindness in the United States. Although studies applying traditional statistical methods have revealed that heavy metals may be essential environmental risk factors for diabetic retinopathy, there is a lack of analyses based on machine learning (ML) methods to adequately explain the complex relationship between heavy metals and DR and the interactions between variables. Based on characteristic variables of participants with and without DR and heavy metal exposure data obtained from the NHANES database (2003-2010), a ML model was developed for effective prediction of DR. The best predictive model for DR was selected from 11 models by receiver operating characteristic curve (ROC) analysis. Further permutation feature importance (PFI) analysis, partial dependence plots (PDP) analysis, and SHapley Additive exPlanations (SHAP) analysis were used to assess the model capability and key influencing factors. A total of 1042 eligible individuals were randomly assigned to two groups for training and testing set of the prediction model. ROC analysis showed that the k-nearest neighbour (KNN) model had the highest prediction performance, achieving close to 100% accuracy in the testing set. Urinary Sb level was identified as the critical heavy metal affecting the predicted risk of DR, with a contribution weight of 1.730632 ± 1.791722, which was much higher than that of other heavy metals and baseline variables. The results of the PDP analysis and the SHAP analysis also indicated that antimony (Sb) had a more significant effect on DR. The interaction between age and Sb was more significant compared to other variables and metal pairs. We found that Sb could serve as a potential predictor of DR and that Sb may influence the development of DR by mediating cellular and systemic senescence. The study revealed that monitoring urinary Sb levels can be useful for early non-invasive screening and intervention in DR development, and also highlighted the important role of constructed ML models in explaining the effects of heavy metal exposure on DR.

Preliminary construction of a microecological evaluation model for uranium-contaminated soil

With the extensive development of nuclear energy, soil uranium contamination has become an increasingly prominent problem. The development of evaluation systems for various uranium contamination levels and soil microhabitats is critical. In this study, the effects of uranium contamination on the carbon source metabolic capacity and microbial community structure of soil microbial communities were investigated using Biolog microplate technology and high-throughput sequencing, and the responses of soil biochemical properties to uranium were also analyzed. Then, ten key biological indicators as reliable input variables, including arylsulfatase, biomass nitrogen, metabolic entropy, microbial entropy, Simpson, Shannon, McIntosh, Nocardioides, Lysobacter, and Mycoleptodisus, were screened by random forest (RF), Boruta, and grey relational analysis (GRA). The optimal uranium-contaminated soil microbiological evaluation model was obtained by comparing the performance of three evaluation methods: partial least squares regression (PLS), support vector regression (SVR), and improved particle algorithm (IPSO-SVR). Consequently, partial least squares regression (PLS) has a higher R2 (0.932) and a lower RMSE value (0.214) compared to the other. This research provides a new evaluation method to describe the relationship between soil ecological effects and biological indicators under nuclear contamination.

Heatwaves in Peninsular Malaysia: a spatiotemporal analysis

One of the direct and unavoidable consequences of global warming-induced rising temperatures is the more recurrent and severe heatwaves. In recent years, even countries like Malaysia seldom had some mild to severe heatwaves. As the Earth's average temperature continues to rise, heatwaves in Malaysia will undoubtedly worsen in the future. It is crucial to characterize and monitor heat events across time to effectively prepare for and implement preventative actions to lessen heatwave's social and economic effects. This study proposes heatwave-related indices that take into account both daily maximum (Tmax) and daily lowest (Tmin) temperatures to evaluate shifts in heatwave features in Peninsular Malaysia (PM). Daily ERA5 temperature dataset with a geographical resolution of 0.25° for the period 1950-2022 was used to analyze the changes in the frequency and severity of heat waves across PM, while the LandScan gridded population data from 2000 to 2020 was used to calculate the affected population to the heatwaves. This study also utilized Sen's slope for trend analysis of heatwave characteristics, which separates multi-decadal oscillatory fluctuations from secular trends. The findings demonstrated that the geographical pattern of heatwaves in PM could be reconstructed if daily Tmax is more than the 95th percentile for 3 or more days. The data indicated that the southwest was more prone to severe heatwaves. The PM experienced more heatwaves after 2000 than before. Overall, the heatwave-affected area in PM has increased by 8.98 km2/decade and its duration by 1.54 days/decade. The highest population affected was located in the central south region of PM. These findings provide valuable insights into the heatwaves pattern and impact.

Virtual sample generation empowers machine learning-based effluent prediction in constructed wetlands

The design of constructed wetlands (CWs) is critical to ensure effective wastewater treatment. However, limited availability of reliable data can hamper the accuracy of CW effluent predictions, thus increasing design costs and time. In this study, a novel effluent prediction framework for CWs is proposed, utilizing data dimensionality reduction and virtual sample generation. By using four the machine learning algorithms (Cubist, random forest, support vector regression, and extreme learning machine), important features of CW design are identified and used to build prediction models. The extreme learning machine algorithm achieved the highest determination coefficient and lowest error, identifying it as the most suitable algorithm for effluent prediction. A multi-distribution mega-trend-diffusion algorithm with particle swarm optimization was employed to generate virtual samples. These virtual samples were then combined with real samples to retrain the prediction model and verify the optimization effect. Comparative analysis demonstrated that the integration of virtual samples significantly improved the prediction accuracy for ammonium and chemical oxygen demand. The root mean square error decreased by averages of 60.5% and 42.1%, respectively, and the mean absolute percentage error by averages of 21.5% and 23.8%, respectively. Finally, a CW design process is proposed based on prediction models and virtual samples. This integrated forward prediction and reverse design tool can efficiently support CW design when sample sizes are limited, ultimately leading to more accurate and cost-effective design solutions.

A predictive model for determining the nitrite concentration in the effluent of an anammox reactor using ensemble regression tree algorithm

Anaerobic ammonium oxidation (anammox) is a cost-effective biological nitrogen removal method for treating wastewater. Nitrite has strong negative effect on microbial activity of anammox bacteria, while the conventional equitment available for determining nitrite on-line is challenging due to high price. By knowing the concentration of nitrite in the effluent, its concentration in the reactor can be controlled accordingly. To investigate this, an ensemble regression tree algorithm was used to establish the predictive model proposed in the current work. Moreover, the Bayesian algorithm was adopted to systematically optimize various parameters of machine learning algorithms. The predicted concentrations of nitrite were in good agreement with the observed values, and the coefficient of determination (R2) and root mean squared error (RMSE) values reached 0.91 and 4.81, respectively. Furthermore, the model established by the ensemble regression tree algorithm was compared with models established by commonly used machine learning algorithms. Finally, the established models were applied to another anammox reactor, and the predicted results of ensemble regression tree model were found to be in good agreement with the experimental values with R2 and RMSE values of 0.84 and 6.34, respectively.

Effects of heavy metal exposure on hypertension: A machine learning modeling approach

Heavy metal exposure is a common risk factor for hypertension. To develop an interpretable predictive machine learning (ML) model for hypertension based on levels of heavy metal exposure, data from the NHANES (2003-2016) were employed. Random forest (RF), support vector machine (SVM), decision tree (DT), multilayer perceptron (MLP), ridge regression (RR), AdaBoost (AB), gradient boosting decision tree (GBDT), voting classifier (VC), and K-nearest neighbour (KNN) algorithms were utilized to generate an optimal predictive model for hypertension. Three interpretable methods, the permutation feature importance analysis, partial dependence plot (PDP), and Shapley additive explanations (SHAP) methods, were integrated into a pipeline and embedded in ML for model interpretation. A total of 9005 eligible individuals were randomly allocated into two distinct sets for predictive model training and validation. The results showed that among the predictive models, the RF model demonstrated the highest performance, achieving an accuracy rate of 77.40% in the validation set. The AUC and F1 score for the model were 0.84 and 0.76, respectively. Blood Pb, urinary Cd, urinary Tl, and urinary Co levels were identified as the main influencers of hypertension, and their contribution weights were 0.0504 ± 0.0482, 0.0389 ± 0.0256, 0.0307 ± 0.0179, and 0.0296 ± 0.0162, respectively. Blood Pb (0.55-2.93 μg/dL) and urinary Cd (0.06-0.15 μg/L) levels exhibited the most pronounced upwards trend with the risk of hypertension within a specific value range, while urinary Tl (0.06-0.26 μg/L) and urinary Co (0.02-0.32 μg/L) levels demonstrated a declining trend with hypertension. The findings on the synergistic effects indicated that Pb and Cd were the primary determinants of hypertension. Our findings underscore the predictive value of heavy metals for hypertension. By utilizing interpretable methods, we discerned that Pb, Cd, Tl, and Co emerged as noteworthy contributors within the predictive model.

The presence of polycyclic aromatic hydrocarbons (PAHs) in air particles and estimation of the respiratory deposition flux

Polycyclic aromatic hydrocarbons (PAHs) in the atmospheric particles constitute a topic of growing health concern. This study aims to calculate PAH concentrations, identify the source, assess the health risk from exposure to carcinogenic PAHs, and the respiratory deposition flux. PM₁₀ and PM_2.5 were collected in September 2019 in the urban, semi-urban, and semi-urban-industrial areas of Kuala Lumpur, Batu Pahat, and Bukit Rambai, respectively. A total of 18 PAHs from PM₁₀ and 17 PAHs from PM_2.5 were extracted using dichloromethane and determined using gas chromatography coupled with a flame ionization detector (GC-FID). The health risk assessment (HRA) calculated included B[a]P equivalent (B[a]P_eq), lifetime lung cancer risk (LLCR), incremental lifetime cancer risk (ILCR), and respiratory deposition dose (RDD). The results show PAHs in PM₁₀ recorded in Kuala Lumpur (DBKL), Batu Pahat (UTHM), and Bukit Rambai are 9.91, 8.45, and 9.57 ng/m³, respectively. The average PAHs in PM_2.5 at the three sampling sites are 11.65, 9.68, and 9.37 ng/m³, respectively. The major source of PAHs obtained from the DRs indicates pyrogenic activities for both particle sizes. For PM₁₀, the total B[a]P_eq in DBKL, UTHM, and Bukit Rambai were 1.97, 1.82, and 2.32 ng/m^3, respectively. For PM_2.5 samples, the total B[a]P_eq in DBKL, UTHM, and Bukit Rambai were 2.80, 2.33, and 2.57 ng/m³_, respectively. The LLCR and ILCR show low to moderate risk for all age groups. The RDD of adults and adolescents is highest in both PM₁₀ and PM_2.5, followed by children, toddlers, and infants. Overall, we perceive that adults and adolescents living in the urban area of Kuala Lumpur are at the highest risk for respiratory health problems because of prolonged exposure to PAHs in PM₁₀ and PM_2.5, followed by children, toddlers, and infants.

Blood Biomarkers Panels for Screening of Colorectal Cancer and Adenoma on a Machine Learning-Assisted Detection Platform

OBJECTIVE

A mini-invasive and good-compliance program is critical to broaden colorectal cancer (CRC) screening and reduce CRC-related mortality. Blood testing combined with imaging examination has been proved to be feasible on screen for multicancer and guide intervention. The study aims to construct a machine learning-assisted detection platform with available multi-targets for CRC and colorectal adenoma (CRA) screening.

METHODS

This was a retrospective study that the blood test data from 204 CRCs, 384 CRAs, and 229 healthy controls was extracted. The classified models were constructed with 4 machine learning (ML) algorithms including support vector machine (SVM), random forest (RF), decision tree (DT), and eXtreme Gradient Boosting (XGB) based on the candidate biomarkers. The importance index was used by SHapely Adaptive exPlanations (SHAP) analysis to identify the dominant characteristics. The performance of classified models was evaluated. The most dominating features from the proposed panel were developed by logistic regression (LR) for identification CRC from control.

RESULTS

The candidate biomarkers consisted of 26 multi-targets panel including CEA, AFP, and so on. Among the 4 models, the SVM classifier for CRA yields the best predictive performance (the area under the receiver operating curve, AUC: .925, sensitivity: .904, and specificity: .771). As for CRC classification, the RF model with 26 candidate biomarkers provided the best predictive parameters (AUC: .941, sensitivity: .902, and specificity: .912). Compared with CEA and CA199, the predictive performance was significantly improved. The streamlined model with 6 biomarkers for CRC also obtained a good performance (AUC: .946, sensitivity: .885, and specificity: .913).

CONCLUSIONS

The predictive models consisting of 26 multi-targets panel would be used as a non-invasive, economical, and effective risk stratification platform, which was expected to be applied for auxiliary screening of CRA and CRC in clinical practice.

Machine learning identifies prominent factors associated with cardiovascular disease: findings from two million adults in the Kashgar Prospective Cohort Study (KPCS)

BACKGROUND

Identifying factors associated with cardiovascular disease (CVD) is critical for its prevention, but this topic is scarcely investigated in Kashgar prefecture, Xinjiang, northwestern China. We thus explored the CVD epidemiology and identified prominent factors associated with CVD in this region.

METHODS

A total of 1,887,710 adults at baseline (in 2017) of the Kashgar Prospective Cohort Study were included in the analysis. Sixteen candidate factors, including seven demographic factors, 4 lifestyle factors, and 5 clinical factors, were collected from a questionnaire and health examination records. CVD was defined according to International Clinical Diagnosis (ICD-10) codes. We first used logistic regression models to investigate the association between each of the candidate factors and CVD. Then, we employed 3 machine learning methods-Random Forest, Random Ferns, and Extreme Gradient Boosting-to rank and identify prominent factors associated with CVD. Stratification analyses by sex, ethnicity, education level, economic status, and residential setting were also performed to test the consistency of the ranking.

RESULTS

The prevalence of CVD in Kashgar prefecture was 8.1%. All the 16 candidate factors were confirmed to be significantly associated with CVD (odds ratios ranged from 1.03 to 2.99, all p values < 0.05) in logistic regression models. Further machine learning-based analysis suggested that age, occupation, hypertension, exercise frequency, and dietary pattern were the five most prominent factors associated with CVD. The ranking of relative importance for prominent factors in stratification analyses showed that the factor importance generally followed the same pattern as that in the overall sample.

CONCLUSIONS

CVD is a major public health concern in Kashgar prefecture. Age, occupation, hypertension, exercise frequency, and dietary pattern might be the prominent factors associated with CVD in this region.In the future, these factors should be given priority in preventing CVD in future.

Impacts of climate change and environmental degradation on children in Malaysia

The impacts of climate change and degradation are increasingly felt in Malaysia. While everyone is vulnerable to these impacts, the health and wellbeing of children are disproportionately affected. We carried out a study composed of two major components. The first component is an environmental epidemiology study comprised of three sub-studies: (i) a global climate model (GCM) simulating specific health-sector climate indices; (ii) a time-series study to estimate the risk of childhood respiratory disease attributable to ambient air pollution; and (iii) a case-crossover study to identify the association between haze and under-five mortality in Malaysia. The GCM found that Malaysia has been experiencing increasing rainfall intensity over the years, leading to increased incidences of other weather-related events. The time-series study revealed that air quality has worsened, while air pollution and haze have been linked to an increased risk of hospitalization for respiratory diseases among children. Although no clear association between haze and under-five mortality was found in the case-crossover study, the lag patterns suggested that health effects could be more acute if haze occurred over a longer duration and at a higher intensity. The second component consists of three community surveys on marginalized children conducted (i) among the island community of Pulau Gaya, Sabah; (ii) among the indigenous Temiar tribe in Pos Kuala Mu, Perak; and (iii) among an urban poor community (B40) in PPR Sg. Bonus, Kuala Lumpur. The community surveys are cross-sectional studies employing a socio-ecological approach using a standardized questionnaire. The community surveys revealed how children adapt to climate change and environmental degradation. An integrated model was established that consolidates our overall research processes and demonstrates the crucial interconnections between environmental challenges exacerbated by climate change. It is recommended that Malaysian schools adopt a climate-smart approach to education to instill awareness of the impending climate change and its cascading impact on children's health from early school age.