Next Article in Journal
Diverging Carbon Balance and Driving Mechanisms of Expanding and Shrinking Cities in Transitional China
Previous Article in Journal
Evaluation and Projection of Degree-Days and Degree-Days Categories in Southeast Europe Using EURO-CORDEX
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Systematic Review

Explainable Artificial Intelligence and Machine Learning for Air Pollution Risk Assessment and Respiratory Health Outcomes: A Systematic Review

by
Israel Edem Agbehadji
1,* and
Ibidun Christiana Obagbuwa
2
1
Centre for Global Change, Sol Plaatje University, Kimberley 8301, South Africa
2
Department of Computer Science and Information Technology, Faculty of Natural and Applied Sciences, Sol Plaatje University, Kimberly 8301, South Africa
*
Author to whom correspondence should be addressed.
Atmosphere 2025, 16(10), 1154; https://doi.org/10.3390/atmos16101154
Submission received: 26 August 2025 / Revised: 23 September 2025 / Accepted: 30 September 2025 / Published: 1 October 2025
(This article belongs to the Section Air Quality and Health)

Abstract

Air pollution is a leading environmental risk that causes respiratory morbidity and mortality. The increasing availability of high-resolution environmental data and air pollution-related health cases have accelerated the use of machine learning models (ML) to estimate environmental exposure–response relationships, forecast health risks and call for the needed policy and practical interventions. Unfortunately, ML models are opaque, in a sense that, it is unclear how these models combine various data inputs to make a concise decision. Thus, limiting its trust and use in clinical matters. Explainable artificial intelligence (xAI) models offer the necessary techniques to ensure transparent and interpretable models. This systematic review explores online data repositories through the lens of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guideline to synthesize articles from 2020 to 2025. Various inclusion and exclusion criteria were established to narrow the search to a final selection of 92 articles, which were thoroughly reviewed by independent researchers to reduce bias in article assessment. Equally, the ROBINS-I (Risk Of Bias In Non-randomized Studies of Interventions) domain strategy was helpful in further reducing any possible risk in the article assessment and its reproducibility. The findings reveal a growing adoption of ML techniques such as random forests, XGBoost, parallel lightweight diagnosis models and deep neural networks for health risk prediction, with SHAP (SHapley Additive exPlanations) emerging as the dominant technique for these models’ interpretability. The extremely randomized tree (ERT) technique demonstrated optimal performance but lacks explainability. Moreover, the limitations of these models include generalizability, data limitations and policy translation. This review’s outcome suggests limited research on the integration of LIME (Local Interpretable Model-Agnostic Explanations) in the current ML model; it recommends that future research could focus on causal-xAI-ML models. Again, the use of such models in respiratory health issues may be complemented with a medical professional’s opinion.

1. Introduction

Air pollution impacts people’s health globally, and it is considered one of the leading risk factors for premature death worldwide [1]. Again, it is a major environmental and public health challenge in many countries [2]. Air pollution can severely impact the lungs of children and adults; however, it is often not considered a priority in the public health sector by policymakers in some developing countries [3]. Sankar et al. [4] indicate that the decline in air quality affects major cities, which negatively impact human health. One common air pollutant is particulate matter (PM2.5), which suggests the need for an optimal and precise model that could be supported by a policy document.
Respiratory health outcomes refer to the state of the respiratory system, which involves the presence or absence of any disease, and it likely impacts an individual. Health outcome ranges from normal lung function to various respiratory illnesses and diseases, which are often measured using lung function tests, symptom reporting (like coughing and wheezing) and the presence of diagnosed respiratory conditions. Furthermore, respiratory illnesses and diseases can be in the form of asthma attacks, chronic obstructive pulmonary disease (COPD), respiratory infections and many more. Respiratory abnormality often puts a significant health burden on the individual, which requires prompt intervention from a health facility [5]. Notably, air pollution is a well-known contributor to asthma. Furthermore, air toxics are hazardous air pollutants with serious health effects [6].
Respiratory health outcome models are analytical models for quantifying the relationship between environmental exposures (especially air pollutants) and respiratory events. Respiratory health outcomes can significantly affect a person’s quality of life, impacting their ability to perform daily activities, exercise and participate in social events. Understanding respiratory health outcome is crucial to developing strategies to prevent respiratory illnesses and effectively manage existing conditions to improve the overall well-being of a person. Human exposure to air pollutants such as particulate matter, sulfur dioxide, ozone (O3) and carbon monoxide is known to contribute to an adverse health hazard, which results in respiratory diseases, cardiorespiratory diseases, cancers, and worst of all, it can lead to death [7].
With the acceleration of industrialization, industrial smoke particles containing complex chemical compositions and varying particle sizes pose a serious threat to the environment and human health [8]. While AI-based models are increasingly being used in assessing the risks of exposure to hazardous air pollutants, the quality of a model’s decision is a subject for open debate. Arguably, users of AI-based models are more interested in how and why an AI model made a particular decision relative to a respiratory health outcome. Understanding these dynamics would help in making the choice of an AI technique for environmental and health impact assessment systems and, in some instance, help understand the need to combine federated learning models in impact-based health assessment models. Due to the proliferation of AI models in different contexts, model generalization is problematic and this could lead to the lack of trust in those generalized model in practical health cases [9]. AI models for respiratory impact assessment could include some elements of environmental variables such as the air quality index (AQI) and other meteorological variables [7]. Unfortunately, the geographical positioning of a country implies different underlying weather conditions, which makes model generalization challenging.
AI models are becoming a common commodity in every sector, from health to manufacturing to education, mining and many more. Such AI-driven models embody machine learning (ML) and deep learning (DL), which are employed in robotic-assisted surgery, thus helping to optimize the clinical workflows and improve patient outcomes [10] in categories of health outcomes such as any health symptoms, asthma, allergy and flu-like symptoms [11]. Machine learning models play a key role in assessing and predicting these health-related outcomes [11]. Examples of such models include logistic regression (LR) and random forest (RF) models.
The current dispensation of the integration of xAI models into ML/AI models called for a thorough review of machine learning models in air pollution risk assessment and respiratory health outcomes. Such a review is beneficial to health policymakers and health-related AI practitioners because it suggests the most optimal model that equally explains the how and why behind AI models’ decision relative to air pollution exposure risk and respiratory health outcomes. Furthermore, it benefits clinicians by giving them an alternative strategy to confirm their clinical health outcomes, thus mitigating any possible clinical negligence that comes with legal consequences. Thus, this research aims to systematically review the extant literature on xAI integration with ML models for air pollution risk assessment for respiratory health outcomes. This review contributes to the development of optimal data-driven models for the health sector, and the foresight gained from different countries are shared with the least developed ones, in terms of respiratory health outcomes and air pollution assessment models. In so doing, machine learning models can be tailored and trained on different sets of data other than dataset on air pollution and meteorological factors. Thus, this study bridges the research gap by finding the most optimal machine learning model for air pollution risk assessment and respiratory outcomes.
The optimality in this context refers to the accuracy of a model’s prediction. More importantly, policymakers are concerned about the model’s prediction accuracy to avoid negative health effect outcomes that may result in respiratory diseases, cardiorespiratory diseases, cancers, and worst of all, it can lead to death. In this paper, we attempt to systematically review recent xAI and machine learning models for air pollution risk assessment and respiratory health outcomes. The sections are organized as follows: Section 2 describes the methods and materials used for this review, Section 3 describes the results and discussion, and Section 4 concludes the systematic review.

2. Materials and Methods

In recent years, researchers have applied different methodologies to examine the extant literature on different contexts of air pollution. It is imperative to understand these methodologies to know the gaps to fill in our study. Table 1 presents recent methodological frameworks and research focus.
As summarized in Table 1, prior reviews have focused on diverse but fragmented aspects of the field. Chadalavada et al. [12] concentrated on monitoring and prediction, while Masood and Ahmad [13] provided a broad methodological overview. Subramaniam et al. [15] connected forecasting to human health outcomes, and Liu et al. [16] emphasized bibliometric trends rather than methodological synthesis. Leivaditis, Maniatopoulos [10] narrowed the focus to applications in thoracic surgery, while Ye et al. [17] concentrated on greenspace measurement and health impacts. More recently, Vachon et al. [18] compared statistical and machine learning models for some selected air pollutants.
In contrast, our systematic review makes a distinct contribution by integrating these fragmented perspectives into a unified synthesis that explicitly connects methodological developments, application contexts, and performance outcomes. Unlike bibliometric or domain-specific reviews, our work (i) develops a new comparative framework across spatiotemporal and AI-based approaches, (ii) highlights underexplored intersections between pollution risk assessment, respiratory health-related outcomes, and transparent decision-making applications, and (iii) identifies three cross-cutting gaps in the literature: limited integration of xAI with spatiotemporal deep learning models, despite their promise for fine-grained air pollution forecasting; weak linkage between pollutant risk assessment and downstream decision support, such as exposure-aware routing, early warning systems or policy guidance—most reviews treat prediction and application separately; and lack of standardized evaluation protocols and cross-domain validation, making it difficult to compare models across regions, pollutants, and health outcomes.
By systematically addressing these gaps, our review not only consolidates existing knowledge but also extends beyond prior studies by proposing an integration of xAI in ML air pollution assessment models that bridges methodological advances with real-world implementation. Thus, our study adopts the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines to synthesis the most recent, relevant and related literature from 2020 to 2025. Furthermore, registration information is not applicable in this study. The inclusion criteria include the following: peer reviewed articles, conference papers, reviews, publication years and written in the English Language. The exclusion criteria include the following: book chapters, retracted papers, observational and animal studies. Figure 1 illustrates the PRISMA flow diagram detailing the systematic process: identification, screening and inclusion.
The following steps were undertaken to ensure quality assessment of the reviewed research article.

2.1. Research Questions

The research questions underpinning this systematic review are as follows:
  • What machine learning techniques are used for air pollutants risk assessment and respiratory health outcomes?
  • To what extent is xAI integrated into machine learning models for air pollution risk assessment and respiratory health outcomes?
While these research questions provided the hypothetical basis for the systematic review, an initial search keyword was conducted to explore the existing literature in the online repository.

2.2. Literature Search Strategy

This review followed the PRISMA guidelines, and the search strategy ensured a thorough article search through online data repositories: PubMed, ScienceDirect, Scopus and IEEE Explore. The search was carried out using the following keywords: “Air Pollution”, “Air pollution risk assessment”, “Respiratory Health Outcomes”, “Machine Learning” and “Explainable Artificial Intelligence”. The articles were extracted from each data repository on the 24 August 2025. The Boolean operators such as “AND” and “OR” were used in the search string to identify articles. Appendix A.1 (Supplementary S1) shows the search string used for the systematic review.

2.3. Risk of Bias Assessment and Reproducibility Strategy

The potential biases include publication bias and the lack of transparency in ML model reporting. To minimize risks of bias and enhance reproducibility, several strategies were adopted. Firstly, two independent academic researchers were engaged in the data collection process. The first researcher was responsible for data retrieval from the online data repositories, while the second merged the extracted datasets and removed duplicates using the EndNote reference manager. Discrepancies and uncertainties were resolved through group discussions, ensuring consistency and agreement on the final set of included studies.
Secondly, subject matter expert input was critical in confirming the relevance of studies included in the final analysis and to avoid bias in classification. Again, search keywords were predefined and agreed upon by the independent researchers to ensure topic alignment within the scope of the review.
Finally, to assess study quality and transparency, we drew upon established criteria (e.g., PRISMA guidelines, ROBINS-I (Risk Of Bias In Non-randomized Studies of Interventions)) for risk of bias in non-randomized studies and best practices in ML model reporting). These steps enhance the reproducibility of the review process and mitigate potential biases, including publication bias and selective reporting.
In terms of ROBINS-1, the domains relevant in this review focused on biases in confounding, reviewer selection, missing data and the selection of reported outcome. Firstly, bias due to confounding described the differences in contextual factors that could affect the review which includes ML, xAI and respiratory health outcomes. Secondly, expect view was crucial in avoiding misclassification of articles. Thirdly, bias due to missing data in terms of DOI was addressed by retrieving from Google Scholar, where missing author names were removed. Finally, bias in the selection of reported results ensured that all the specified outcomes were reported and that ML/xAI model performance metrics were transparently presented in the articles reviewed. Performance metrics, in terms of the model’s result accuracy, are measured with RMSE, MSE and R2 values. An optimal model refers to a model that guarantees high results accuracy.

2.4. Articles Extraction

Articles were saved in EndNote file type, where 177 articles were finally extracted, with their related articles downloaded for a thorough literature review. After removing 25 duplicate article titles, 153 papers with unique titles were screened further. In the second stage, documents that were not within the scope were excluded from the final document review. The final stage required a thorough review of the articles (92) to clearly understand the issues relating to the research questions. Because of the possibility of missing articles during extraction, Google Scholar helped to retrieve the missing articles/documents using the DOI, because they contain some useful information.

3. Results and Discussion

Air pollution is a global health hazard which affects the health of an individual in several ways [12]. Respiratory diseases ranging from severe acute respiratory syndrome (SARS), Middle East respiratory syndrome (MERS), HIV, influenza A (H1N1) and COVID-19 can cause lung infections [19]. Huang, Zhou [20] identified the air pollutant that primarily contributes to mortality risk. They also characterized individual and joint pollutants (i.e., multi-pollutants) and examined the association of long-term air pollutant exposures with cardiopulmonary mortality. Liu, Li [21] indicate that limited information exists on the interaction effects between air pollution and influenza vaccination on allergic respiratory diseases. They conducted a large population-based study to evaluate the interaction effects between influenza vaccinations. Atzeni et al. [22] acknowledged the complexity in managing asthma issues and the many environmental factors that play a critical role in asthma management. It is critical to monitor and assess personal exposure to air pollutants to examine its impact on health and respiratory outcomes aside from asthma conditions. Neo, Hasikin [7] conducted a systematic review on air quality impact on human health, focusing on the advancement of artificial intelligence (AI) and the use of federated learning.
Among the studies that meet the inclusion criteria but were excluded include the study by Rowland et al. [23] which lacks ML and xAI but focused on gut feelings relative to fatty acids for the regulation of respiratory health; the study by Renouf, Sutanto [24], though focused on asthma, was largely on laboratory research, lacking machine learning and/or xAI content.
By combining independent screening, keyword-based selection, expert input, and formal bias assessment with ROBINS-I, we strengthened both the transparency and reproducibility of our review, while mitigating risks such as publication bias and selective reporting. The following subsections present the results of the systematic literature review to address the key research questions.

3.1. Machine Learning for Air Pollution Risk Assessment and Respiratory Health Outcomes

Ye et al. [17] suggested health outcomes such as obesity/overweight, myopia, lung health, circulatory health, cognitive function and general health in children and adolescents. Vachon, Kerckhoffs [18] indicated that machine learning techniques such as tree-based methods are superior to linear non-regularized techniques for predicting the ambient concentrations of NO2, UFPs and BC. They also suggested that the most frequently used models are linear non-regularized methods and RF.
Geo-statistical models have been applied to assess fine-scaled air pollution exposures; however, they have not been used to assess non-regulated air pollutants, and this led to the use of a spatial model to assess both regulated and non-regulated pollutants (such as ultrafine particles, black carbon and benzene) [25]. This spatial model applied partial least squares regression (PLS) and a dimension reduction algorithm to predict pollution exposure. Upon comparison with other machine learning models, the RF and PLS emerged as the most effective models that can predict the short-term average concentration of toxic air pollutants. After evaluation, both models guaranteed a high cross-validation R2 score and the lowest root mean square error value.
Li et al. [26] indicated that understanding how a model estimates error can help determine its impact on model development. Unfortunately, their experiment suggested a negative impact on model development when the RF model was used to predict the personal PM2.5 exposure. Furthermore, an atmosphere-to-scenario-based ratio and linear regression model were used for the essential parameters of the scenario-based ratio exposure model. Their finding, however, suggested that personal PM2.5 exposure was higher than atmospheric PM2.5 concentration, with an R2 value between 0.65 and 0.93.
Maio et al. [27] examined the long-term effect of multi-pollutants on human health. The health effects that were examined include the prevalence of respiratory/allergic symptoms and diseases. During their study, questionnaires were administered to obtain the risk factors and their health outcomes from participants; thereafter, the association between air pollutants and health effects were assessed with machine learning models such as LR (for single pollutant) and principal component LR model (for multiple pollutants). The pollutants were PM2.5, PM10, NO2 and O3. Their findings suggested that PM10 and PM2.5 contributed to an increased health effects in different percentages: rhinitis (14–25%), asthma (23–34%) and night awakening (30–33%); NO2 was rhinitis (6–9%), asthma (7–8%) and night awakening (12%); and O3 contributed 37% increased odds for asthma attacks [27].
Tang et al. [28] studied the major factors that control emission of particle-bound hydrophobic organic compounds (HOCs) in lung fluids. They focused on particles from barbecue and smoking; then a machine learning model with a vitro method assessed the most significant factor that affected inhalation bio-accessibility of polycyclic aromatic hydrocarbons (PAHs), followed by the organic carbon and elemental carbon contents. Their study suggested significant particle deposition efficiency and bio-accessible fractions of HOCs in the risk assessment.
Yang et al. [29] indicated that multiple air pollutants often interact, which suggests that there is an inter-correlation of air pollutants, and as such, single pollutant estimation models often fall short in estimating this inter-correlation. This challenge was addressed via a joint multi-pollutant retrieval model (referred to as physics-informed multi-task deep neural network (phyMTDNN)). During the experiment, six air pollutants, namely PM2.5, PM10, SO2, NO2, CO and O3, were used and the evaluation with cross-validation R2 value suggested a variation from 0.7 to 0.90 [29].
Zhang et al. [30] opined that PM2.5 has a significant impact on health and climate. During their impact analysis, the inter-annual trends in particulate air pollutants and the response to anthropogenic emissions were examined using RF algorithm. This machine learning model quantified key drivers of PM2.5. Furthermore, the health-risk air quality index (HAQI) value appeared to have decreased with significant contribution from anthropogenic emission reduction. It was concluded that the contribution of PM2.5 to HAQI was reduced, and there was a significant contribution from O3 [30].
Zhou et al. [31] examined the joint effect of the indoor air pollution index and ambient particulate matter on fetal growth. Prenatal exposure to PM2.5 and PM1 was assessed using a machine learning technique, which further computed the indoor air pollution index. The finding suggested that indoor air pollution and ambient PM exposure had individual and joint negative effects on fetal growth [31]. Ahmadian, Rajabi [32] assessed outdoor air pollutants and their consequent health risks using machine learning models. The pollutants considered were PM2.5, SO2, NO2, and CO. During the assessment, the EPA health risk assessment model was used to understand the exposure to human health. In using the model, the mean hazard quotients (HQ) for PM2.5 and NO2 were established, where the HQ value exceeded 12 h exposure, which raises health risk concerns. The machine learning models used for the analysis are RF, K-Nearest Neighbors (KKN), decision tree (DT) and Naive Bayesian (NB). RF model emerged as the best prediction model with the highest accuracy of R2 value of 0.99.
Chen et al. [33] developed an RF model to predict the daily air pollutants such as PM2.5 and PM10, NO2 and O3 from multisource at an 8 h interval. This model was evaluated with out-of-bag R2, where PM2.5 (0.85), PM10 (0.77), O3 (0.85) and NO2 (0.85).
Guo et al. [34] examined the association of PM2.5 and acceleration of aging (AA) using a two-staged machine learning technique at one km2 resolution. The machine learning technique was based on the post-matching linear mixed model (LMM), where the determination of aging was based on the variation between Klemera–Doubal biological age and chronological age. The outcome suggested that long-term exposure to PM2.5 may increase the biological aging among the adult population, for instance, in China. Furthermore, when the adult population was exposed to an increased level of PM2.5, it also increased their aging process [34].
Jiang et al. [35] designed a navigation system to aid users in commuting at least PM2.5-exposed paths, and the system automatically sends a risk warning to its users. Their navigation system employs machine learning models, namely RF, LightGBM, Adaboost and XGBoost, where XGBoost outperformed the comparative models with R2 value exceeding 0.90 and with root mean square error (RMSE) (~15.74 μg/m3). Thus, commuters who engage in outdoor activities can avoid locations that are highly exposed to PM2.5 concentration [35].
Lv et al. [36] assessed the spatiotemporal dispersal of PM2.5 source-specific impacts in locations without PM2.5 compositions using machine learning models. XGBoost showed outstanding prediction performance for nitrogen dioxide (NO2), carbon oxide (CO) and elemental carbon (EC), with cross-validated R2 values ranging between 0.87 and 0.92 [36].
Masseran, Safari [37] classified the categories of air pollution events (that is, extreme and non-extreme) using a probabilistic machine learning technique, which is based on naïve Bayes technique. These events are grouped based on the magnitude of severity. The naïve Bayes model guaranteed high accuracy, specificity and sensitivity on the trained and test datasets [37].
Chen et al. [38] suggested that ultrafine particles (UFPs) (with a diameter ≤ 100 nm) are airborne particles which can pose a significant health risk, including cardiovascular disease, respiratory and cancer. Their study found that while experimental approaches are unable to capture the temporal fluctuation, computational models are also deficient in capturing the complex nature of the ultrafine particle depositions, thus making it challenging to understand the impact on human health. Some of the sources of UFPs include traffic, combustion and industrial processes [38].
Amini et al. [39] indicates that European cities (such as Copenhagen, Denmark) lack models to understand the long-term average particle size (APS), lung deposited surface area (LDSA) and particle number concentration (PNC). Thus, a machine learning model based on RF and bagged trees was suggested. Their model achieved a high R2 value (0.93) and low root mean square error [39].
Cao et al. [40] focused on both water and air pollution as they pose significant threats to public health. They used traditional biological methods, which are time consuming, costly and limited in scale. Subsequently, a machine learning model, that is, Knowledge-guided Pre-training of Graph (KPGT), was used as a pollution risk assessment model, which predicted the carcinogenicity of pollutants with an AUC (0.83) [40].
Chen et al. [41] indicates that air pollution and climate change are key challenges in recent times. Unfortunately, there is an inadequate assessment of the impact of extreme temperatures and air pollution from different regions. Air pollution includes daily PM2.5, PM10, NO2, and O3 concentrations. They design machine learning models to estimate the daily concentration within the required location-specific time-varying temperature threshold [41].
Gao et al. [42] examined the association between mixed pollutants in PM2.5 and reproductive health risk using the quantitative structure–activity relationship model (QSAR) integrated machine learning algorithm. In their study, the mixed reproductive health risks associated with phthalates (PAEs) and organophosphates (OPEs) exposure were assessed to understand how strongly the compounds bind to estrogen receptors (ER) and androgen receptors (AR). Thus, it will help gain insight into how the mixed reproductive exposure risk of PAE and OPE is associated with PM2.5 [42]. Phthalates (PAEs) from indoor dust impact human reproductive health [43].
Holloway et al. [44] indicated that collecting satellite data on PM2.5 and NO2 are beneficial in developing a health risk assessment model to track both long-term and short-term trends on air quality information and related health records.
Kuo et al. [45] proposed the machine learning (ML)-measurement model fusion (MMF) framework to quantify air pollutants using Chemical Transport Modeling (CTM) data and to assess the burden of disease (BD) from exposure to different sources of air pollutants. Among the air pollutants considered are PM2.5 and O3. The proposed ML-MMF successfully improves CTM-modeled for PM2.5 and O3 with R2 values, respectively, ranging from 0.41 to 0.86 and 0.48 to 0.82 [45].
Lee et al. [46] developed a comprehensively integrated air-risk index (CARI) to assess the adverse effect of hazardous air pollutants (HAP) such as particulate matter (PM). They suggest that including a machine learning model enhances the spatiotemporal resolution of the CARI model for large industrial locations.
Liu et al. [47] examined how atmospheric particulate matter (PM) affects human health using a Lung Performance Optimization-based XGBoost (LPO-XGBoost) model, which leverages adaptive optimization principles inspired by lung function to enhance classic PM source apportionment. Comparative analyses with models such as RF, Support Vector Machine (SVM), and their LPO-enhanced variants confirm that LPO-XGBoost provides the most reliable performance in estimating pollution source contributions, offering scalability and robustness ideal for high-time-resolution observational data and having a coefficient of determination R2 (0.88) [47].
Wenjie et al. [48] indicates that air pollution and tobacco smoke are one of the significant risk factors. The presence of Benzopyrene (BaP) in these pollutants was explored using a combination of different models, such as network toxicology, machine learning (Support Vector Machine, RF and LASSO regression), and molecular docking analyses. This resulted in the creation of a robust periodontitis risk assessment model to understand environmental pollutants on periodontal health [48].
Yang et al. [49] attempted to model the association between varied types of eye-level greenness and pediatric blood pressure using the deep machine learning model and generalized linear mixed-effects regression model. The mediation analysis results show that hypertension is significantly mitigated by lower levels of air pollutants, including PM2.5, PM10, SO2 and NO2. This outcome suggests that eye-level greenness, especially trees, was associated with lower prevalence of hypertension in children, with air pollution exhibiting mediating effects [49].
Zhao et al. [50] clearly identifies the source of atmospheric particulate matter concentration to help in planning the needed intervention to reduce the concentration, which can be harmful to human health. They further stated that, though traditional methods such as chemical mass balance models are effective, machine learning models are more efficient and effective. Such effective models include the source apportionment pipeline (RX model), which combines computer vision and machine learning to trace particle sources. This can help build particulate matter health risk assessment models for atmospheric pollution source analysis [50].
According to Paul et al. [51], assessing the acute and chronic health effects of biomass smoke exposure require reliable estimates of PM2.5 concentrations during the wildfire season and throughout. The RF machine learning model uses potential predictor variables integrated from multiple data sources and estimates daily mean (24 h) PM2.5 concentrations at a 5 km × 5 km spatial resolution [51].
Brooks et al. [52] examined how brick kilns contribute to fine particulate matter (PM2.5) and estimated its association with child asthma symptoms, chronic obstructive pulmonary disease (COPD) and general respiratory symptoms [52].
Hu et al. [53] examined prenatal and early-life exposure to air pollution and extreme temperatures and how they are associated with childhood asthma and wheeze. By using the distributed lag nonlinear models (DLNMs), they suggested that early-life air pollution exposure contributes to the development of childhood asthma and wheeze, while exposure to temperature showed mixed associations with respiratory outcomes.
The study by Okello, Nantanda [54] indicates that ambient air pollution (such as PM2.5) and weather have an impact on respiratory diseases. They applied generalized additive models (GAM) to adjust for rainfall, temperature and humidity, and also assessed the association between air pollution and healthcare facility events. The finding suggested that PM2.5 impacts healthcare facility and that their association are influenced by meteorological factors [54]. Bouma et al. [55] acknowledged the use of different risk exposure assessment methods for outdoor air pollution with asthma onset and lung function in children [55].
The study by Leão et al. [56] indicates that an increase in temperature will have impacts on air particulate matter and health outcomes [56]. Shams et al. [57] suggested the need to link deep learning-based air quality forecasting models with health and socioeconomic outcomes. They proposed the combination of best-performing models, such as Deep convolutional neural networks (Deep-CNN), Long Short-Term Memory (LSTM), and deep neural networks (DNN) to forecast the daily maximum O3 concentrations [57]. Many researchers have acknowledged the impact of air pollution on human health and have adopted different methods to assess the health risk outcomes [56,58,59,60]. Understanding the spatiotemporal dynamics of Population Exposure to PM2.5 (PEP) and its relationship with socioeconomic activity (SEA) is crucial to reducing exposure risks and health dangers [61].
Pei et al. [62] studied the environmental pollutant fluctuation and its potential health impact, particularly viral hepatitis. They focus on 2005 and 2020 viral hepatitis datasets and 2012 to 2020 pollutants datasets (CO, NO2, and O3). Three time series models, Holt–Winters model, seasonal auto-regressive integrated moving average (SARIMA) and generalized additive model (GAM), were applied for prediction and to assess the effect of pollutants on viral hepatitis. Furthermore, ML models were also used for pollutant analysis. While Holt–Winters was superior in predicting the incidence of viral hepatitis, SVM and GPR models were superior in analyzing pollutant data. The findings suggest that patients who are infected with HAV and HEV were influenced by PM10 and CO, whereas SO2 and PM2.5 impact each other significantly. Again, persons aged between 35 and 64 years are more susceptible to these pollutants.
VOC is one of the air pollutants that is very harmful to human beings, and it is associated with risks of acute and chronic disease, such as cancer and leukemia [63]. Given the complex nature of screening the biological particles for VOC exposures, machine learning models and bio-information techniques were found to be limited in validating the biomarkers and toxicity of VOC exposure.
Wang, Wang [64] measured the PM2.5 and PM10 concentrations both indoors and outdoors. While sensors provide accurate measurements, it suggests that the RF model is an effective technique to validate the measurement as compared to the multivariate linear model (ML), K-nearest neighbor model (KNN), support vector machine model (SVM), decision tree model (DT), neural network model (that is, Multi-Layer Perceptron), R2 and RMSE performance validation models. Appendix A.2 (Supplementary S1) provides a summary of the related literature on ML for air pollution risk assessment; Table 2 also presents model categories and related algorithms.
The heterogeneity of methods summarized in Table 2 highlights a key challenge for comparability across studies. Differences in modeling approaches (e.g., tree-based ensembles vs. deep neural networks), dataset sources (monitoring stations, satellite retrievals, clinical records) and validation strategies (cross-validation, temporal splits, rolling-origin evaluation) make it difficult to directly compare reported performance metrics. For example, deep learning models often require large-scale spatiotemporal datasets, while tree-based models perform reliably even with smaller, heterogeneous inputs. Similarly, temporal validation generally produces lower performance estimates compared to random K-fold cross-validation, which may inflate comparability. This methodological diversity underscores the need for standardized evaluation protocols and reporting practices to enable fair comparison and benchmarking of models for air pollution risk assessment and respiratory health outcomes.
In view of this literature, ML techniques have been extensively applied to air pollutants risk assessment and respiratory health outcomes. Machine learning models such as extremely randomized tree (ERT) proves to outperform both XGBoost and RF using the PAHs pollutant, thus leading to the exposure of lung cancer risks; the hybrid LUR and CNN performed optimally in predicting the impact of UFPs on mortality; RF performed optimally in PM2.5 Bound Mercury (PBM2.5) prediction which lead to the findings that population density and power generation stations contribute largely to PBM2.5 concentration. Furthermore, to a very limited extent, PBM2.5 in 2020 posed a non-carcinogenic health risk. This is supported by the hazard quotient (HQ), which remained well below the threshold of concern, with values mostly <0.02 during winter. Such levels indicate that the risk was negligible, meaning exposure did not reach levels typically associated with adverse non-carcinogenic health outcomes. Estimating PM2.5 using TDLPI + XGBoost to enhance the LUR model suggested a moderate health risk in winter. Again, a satellite-based ML model with a time-varying Cox hazard model proves to establish the association between mortality risks due to lung cancer with PM2.5.
To a significant extent, exposure to PM2.5 is associated with an elevated risk of cardiorespiratory mortality [65]. Epidemiological studies show that fine particulate matter can penetrate the lungs of an individual, trigger systemic inflammation, oxidative stress and vascular dysfunction. The risk may be more pronounced among vulnerable groups such as the elderly, children and individuals with some pre-existing conditions. However, the extent of risks varies depending on pollutant concentration, duration of exposure, location and population susceptibility. Therefore, while evidence strongly supports a causal link, the degree of mortality risk remains context-dependent, influenced by environmental and demographic factors.

3.2. Explainable AI with Machine Learning Models for Air Pollution Risk Assessment and Respiratory Health Outcomes

Explainable AI (xAI) helps with model interpretability and transparency. While most machine learning models have been widely used to predict air pollution concentrations, they lack interpretability [64]. Zhang, Yi [66] indicated that environmental pollution contributes to cancer etiology; however, the expensive nature of traditional experiments and their time-consuming nature suggest the need for an alternative approach, such as deep learning models. For instance, the interpretable DL model (CarcGC for chemical carcinogenicity) was proposed, which was based on graph convolutional neural network (GCN) that employed molecular structural graphs as input and for interpretability. Their study further indicated the association of air pollution with the incidence of lung cancers, and, therefore, accurate prediction was suggested.
Rajesh, Babu [67] developed a real-time air quality assessment and health risk prediction model for multi-locations. Their techniques were Gradient Boosting, RF, XGBoost and LSTM. These helped systems to predict air pollutant concentrations and then classify the air quality levels with high temporal accuracy. Furthermore, SHAP (Shapley Additive exPlanations) was used for interpretability of their model.
Razavi-Termeh, Sadeghi-Niaraki [68] examined the lack of interaction between asthma-prone locations and urban environmental factors and suggested the use of spatiotemporal modeling techniques with an xAI technique like SHAP. Their modeling technique leveraged XGBoost (eXtreme Gradient Boosting), optimized using Bat algorithm (BA). Upon evaluating the asthma-prone area map with Receiver Operating Characteristic (ROC), different accuracies were obtained for different seasons and percentages, such as spring (97.3%), summer (97.5%), autumn (97.8%) and winter (98.4%). This result suggested that rainfall and temperature have the most significant impact on asthma. Furthermore, air pollutants in certain seasons, such as PM2.5 (in spring), CO (in summer), O3 (in autumn) and PM10 (in winter) showed the most substantial effect among other factors of air pollution.
Agbehadji and Obagbuwa [69] examined how to enhance the noise in measuring the concentration of SO2 prediction. They indicate that the noise can increase the error of estimation in a dynamic time-varying air pollutant such as SO2. In this regard, an xAI technique using SHAP was integrated with an LSTM-based Adaptive Kalman Filter model. However, the aspect of respiratory health-related outcomes was not considered in their study. Furthermore, Agbehadji and Obagbuwa [70] leveraged xAI (e.g., LIME) in graph neural network that also include A* search strategy (GeoxAI-GCN-LSTM-A*) to find an optimal navigation path for predicted CO concentration in multi-dimension time step. However, the respiratory health-related outcome was not considered in their approach.
Kibria, Hossain [19] pointed out the common lung infection detection techniques like X-ray image analysis, computed tomography (CT) scans and reverse transcription–polymerase chain reaction (RT-PCR). These techniques have their limitations, thus leading to the proposition of parallel lightweight diagnosis model based on depth-wise separable CNN (LW-PDS-CovidNet) model. Additionally, SHAP and gradient-weighted class activation mapping (GRAD-CAM) were the xAI techniques that showed the most significant features. Their model achieved a high accuracy of 98.06% (for two-class), 97.43% (three-class) and 92% (four-class) classification with a low number of parameters. Finally, it was suggested that respiratory diseases, including severe acute respiratory syndrome (SARS), Middle East respiratory syndrome (MERS), HIV, influenza A (H1N1) and COVID-19, can cause lung infections.
Sunder, Tikkiwal [71] indicated that human beings are mostly exposed to fine PM2.5, which has adverse health impacts related to cardiovascular and respiratory-related diseases. They suggested that machine learning models proved to be a good predictor of air quality. To validate their suggestion, they provided a framework based on temporal-based prediction using RF, Gradient Boosting (GB) regression, and Tree-based Pipeline Optimization Tool (TP) Auto ML that worked with meta-heuristic methods such as genetic algorithm. TP produced Global Performance Index (GPI) values of 7.4, which was the highest GPI value for August 2016, while the lowest GPI value (−0.6) was recorded in June 2019. These outcomes were explained with xAI technique, which helped to investigate the fidelity of feature importance of machine learning models used to extract information on rhythmic shift in the PM2.5 concentration patterns.
Chang, Liu [72] demonstrated the use of machine learning to predict pediatric acute respiratory infections (ARIs), which are common in children. The machine learning model predicted six common respiratory pathogens, such as adenovirus, influenza virus (types A and B), parainfluenza virus (PIV), Mycoplasma pneumoniae (MP) and respiratory syncytial virus (RSV). Feature importance was assessed with SHAP values. The findings suggest that event patterns are useful for influenza virus prediction, and C-reactive protein had the highest SHAP value for adenovirus infections. Thus, such models could improve patient outcomes and reduce unplanned medical expenses when integrated into clinical workflow.
Chi, Moghadas-Dastjerdi [73] examined the diagnosis code associated with the highest mortality risk using a deep SHAP (D-SHAP) model. They indicated that in most instances, clinicians are more interested in learning how input (from raw dataset) affects prediction. The results suggest the top five most important diagnosis codes as respiratory failure, sepsis, pneumonia, shock, and acute kidney injury. Fortunately, these results were consistent with the opinion of the physicians. On the contrary, the diagnosis for urinary tract infection showed a discrepancy between D-SHAP values and clinical opinion, because of the low frequency of the disease and its outcome in combination with other comorbidities. Appendix A.3 (Supplementary S1) summarizes the xAI technique integrated with ML for air pollution risk assessment in the literature.
In conclusion, to some extent, xAI has been integrated into ML air pollution concentration prediction models. Among such models are XGBoost with SHAP for air pollutants such as UFPs and PM0.1; LightGBM-based Bayesian with SHAP for PM2.5, OM, NO3, NH4+, BC and SO42−; XGBoost-based NDVI with SHAP for PM2.5. The performance evaluation of the XGBoost with SHAP based on RMSE guaranteed 0.79, suggesting a need to enhance the model’s performance. Also, Table 3 presents the methodological heterogeneity of xAI-ML studies.
Table 3 highlights the methodological heterogeneity in how explainable AI has been integrated into machine learning applications for air pollution and respiratory health outcomes. While previous reviews have focused primarily on either model performance or broad applications of ML in environmental health, our synthesis uniquely maps model architectures, datasets, xAI techniques, and health/pollution outcomes in a comparative framework. This approach reveals not only what models were used, but also how interpretability was operationalized (e.g., SHAP, Grad-CAM, deep SHAP), in what contexts (spatiotemporal prediction, imaging, clinical diagnosis) and with what implications for respiratory health and policy translation. By presenting both a structured comparison and a critical appraisal, this systematic review provides clarity on where methodological diversity strengthens generalizability and where it limits comparability, offering actionable insights for researchers aiming to select, validate and apply interpretable ML models in further air pollution health studies. To some extent, on respiratory outcomes, high PM2.5 contributes significantly to an increased asthma risk. Again, a high urban heat island index (UHI) and high PM2.5 concentration levels also increase asthma risk. In the categorization of health risk outcomes [11], asthma was revealed to be the most featured in the literature to the neglect of others, such as any health symptoms, allergy and flu-like symptoms. PM2.5 has been identified as a significant air pollutant contributing to premature mortality [65]. From an epidemiological perspective, most studies establish associations but not precise causal thresholds; however, SHAP algorithm identifies the concentration thresholds at which the components of PM2.5 could affect mortality. Thus, indicating a long-term mortality and morbidity risk. Table 4 presents the reported performance metrics.
Although several studies report high performance metrics (e.g., R2 > 0.80, AUC > 0.90), the lack of standardized datasets, inconsistent validation approaches, and variable outcome measures preclude direct effect-size comparisons. For example, deep learning models often report superior accuracy in image-based classification, while tree-based ensembles achieve robust performance in spatiotemporal pollution prediction; however, these results are not directly comparable due to differences in input features and evaluation designs. Accordingly, our conclusions emphasize relative performance trends within contexts (e.g., tree-based models for tabular spatiotemporal data, CNNs for imaging) rather than identifying a universally “best” model. These findings suggest that systematic reviews would benefit from harmonized performance reporting and benchmark datasets, which would enable quantitative meta-analysis of ML/xAI models in this domain.

3.3. Challenges and Limitations

Though ML models have been applied within the context of respiratory health outcome models. It is evident that the focus mostly lies on asthma. One of the major challenges in environmental pollution assessment is that health consequences are difficult to evaluate scientifically and accurately [74]. Exposure to particulate air pollution is one of the greatest environmental risk factors for adverse human health outcomes [75].
The challenges in terms of machine learning models are as follows: Firstly, to some extent, models perform differently, which impacts the accuracy of prediction. In view of this, the model’s performance may be good in one location or air pollution dataset, and that same model might perform sub-optimally in another location within the context of air pollution risk analysis and its respiratory health-related outcomes. Thus, generalization of a model would be problematic. The second challenge is that these models are more focused on the prediction of air pollutant concentration and its health risk-related factors, including mortality risk [20]. Unfortunately, causal inference may limit real policy application in the health sector, particularly on respiratory health outcome issues. This, however, may be the reason for reliance on the traditional methods of respiratory health outcome analysis. In this regard, machine learning models could complement or validate expert opinion as well. This suggests a symbiotic relationship between ML models and expert opinion.
The methodological challenge stems from data collected from the online repositories for this systematic review. For instance, not all available data repositories are considered, and it is equally possible that there could be very useful information on those repositories. Another methodological limitation is that the research articles extracted were restricted to articles, review articles and conference paper, neglecting other equally important and relevant materials like book chapters and others.
Generally, most cited studies focus on developed countries including Asia, Europe and North America. Thus, there was limited research data coverage of low- and middle-income countries, where air pollution and respiratory risk is highest.
The challenge relating to the integration of xAI with ML is the possibility of discrepancy in results obtained from the xAI model (e.g., D-SHAP) and clinical results [73]. This suggests that the interpretation of results from ML models should be augmented with an expert view. Thus, clinical validation of xAI model’s outcome should be encouraged due to the generalization challenge of ML models. Furthermore, different air pollution and respiratory datasets could be used to validate these models to enhance model generalizability. While SHAP was the most frequently reported interpretability method in the reviewed studies, alternative xAI techniques such as LIME (Local Interpretable Model-Agnostic Explanations), Grad-CAM (Gradient-weighted Class Activation Mapping), counterfactual explanations and integrated gradients are increasingly applied. SHAP provides consistent, game-theoretic attributions that are generally more stable and globally interpretable than LIME, but it can be computationally expensive and may still struggle with correlated features. LIME, by contrast, is faster and more flexible but often produces explanations that are unstable across runs, which limits its reliability in high-stakes clinical settings. Grad-CAM and related saliency-based methods are particularly useful in image-based tasks (e.g., chest X-rays, CT scans), offering visual heatmaps of important regions, but they provide less quantitative interpretability compared to SHAP or LIME. Counterfactual methods have the advantage of generating “what-if” scenarios that can directly inform clinicians about actionable changes but remain less explored in environmental epidemiology.
From a clinical decision-making perspective, these trade-offs are critical. For example, SHAP’s robust feature attributions can support policymakers in identifying consistent air pollution drivers across populations, whereas LIME’s instability could undermine trust in regulatory applications. Conversely, in diagnostic imaging, Grad-CAM may be more intuitive for physicians since it visually overlays explanations on familiar radiological images. Table 5 presents the comparison of common xAI techniques for air pollution and respiratory health applications.

3.4. Uncertainty Quantification

A critical consideration for the translation of ML models into air quality regulation and respiratory health policy is the treatment of uncertainty. While many studies report performance metrics such as R2, RMSE or AUC, relatively few explicitly quantify predictive uncertainty. This omission limits the reliability of model-informed decision-making in regulatory settings where precautionary principles and health risk thresholds are central. Uncertainty in ML predictions arises from several sources: (1) data uncertainty includes measurement errors, missing data and sensor calibration issues; (2) model uncertainty includes structural choices, parameter variability and overfitting/under fitting); (3) outcome uncertainty includes inherent variability in respiratory health responses to pollutant exposures.
For policymakers, explicit uncertainty estimation methodological approach (e.g., bootstrap resampling and cross-validation, ensemble methods, among others) are crucial to balance precautionary measures, communicate risk transparently and prioritize interventions. For example, if an ML model predicts high respiratory risk from PM2.5 with wide uncertainty intervals, regulators may adopt stricter emission standards or reinforce monitoring in hotspot areas. Conversely, narrow and stable uncertainty bounds can justify targeted, cost-effective interventions. Therefore, integrating uncertainty quantification approach into xAI-ML-based air pollution assessments enhances not only scientific robustness but also regulatory credibility and public trust.

3.5. Implications of the Research Finding

The implication focused on three core areas: practical, policy and theoretical. The practical implication is to determine how the findings could be used in real-world practice. In this regard, the findings suggest methodological considerations to guide healthcare application developers to focus on the best-performing models such as ERT in addition to XGBoost and RF. Again, uncertainty estimation methods, when integrated into xAI-based ML models, provide some robustness in existing respiratory health outcome applications. Furthermore, healthcare sector manages volumes of patient records using legacy systems built with statistical and/or hybrid models, and integrating the optimal model identified in this review is beneficial. Furthermore, it helps health practitioners find alternative strategies aside from laboratory work to validate their laboratory results with machine learning methods. It is imperative for models (causal-xAI-ML/DL) that seek to find causal relationship with some explanation to be considered for practical implementation.
Policy implications seek regulatory reforms within the health sector that leverage environmental data and respiratory impact. In this regard, the findings suggest the adoption of hybrid models (ML/DL and xAI) in the work process of the health sector, which require the re-design of work structure and staff training on the use of new models for respiratory risk assessment and its outcomes. Again, policy that enforces uncertainty quantification in xAI-based ML/DL in addition to the current performance metrics is encouraged for AI models on health.
Theoretically, the findings are highly relevant and timely, given the increasing adoption of AI techniques in environmental epidemiology and the growing demand for interpretability in clinical and policy contexts.

3.6. Future Directions

With continued advancements and responsible integration, AI will play a pivotal role in shaping the next generation of precision thoracic surgery [10]. To some extent, the use of other explainable AI models, such as LIME could be explored within the context of air pollution risk assessment models for respiratory health outcomes. While Table 3 provides a comprehensive lens on methodological diversity (models, xAI methods, application domains), and it also highlights their gaps (e.g., limited direct integration with respiratory health outcomes, reliance on post hoc SHAP). Future research is focused on digital twin technology, federated learning, and xAI to improve AI interpretability, reliability, and accessibility.
Considering the lack of causal inference, future work should explore causal ML approaches (e.g., causal forests, Directed Acyclic Graphs (DAG)-based models). DAG models the complex causal structures and improve causal inference in various health-related scenarios. While causal forests identify effect heterogeneity with tree-based models, DAG methods offer a structural method to capture causal relationships before using an estimation technique. Furthermore, developing a hybrid causal-xAI-ML model offers interpretability in clinical context.
To enable future meta-analyses and more robust cross-study comparisons, this review recommends the development of a standardized benchmarking framework with the following elements. (1) Dataset standardization: establish benchmark datasets (e.g., multi-city, multi-season air quality and health records) that can serve as reference points for model evaluation, and encourage open-access repositories to facilitate reproducibility and fair comparison. (2) Harmonize performance metrics which require consistent reporting of core metrics (e.g., R2, RMSE/MAE for regression; AUC, sensitivity, specificity for classification), mandating clear definitions of metrics (e.g., temporal vs. random cross-validation splits) to avoid misinterpretation. (3) Validation protocols promote standardized validation strategies such as rolling-origin evaluation for time-series forecasting and nested cross-validation for small datasets, explicitly documenting training/validation/test splits, including temporal coverage, to improve reproducibility. (4) Transparent xAI reporting requires reporting of performance and interpretability outputs (e.g., SHAP values, feature rankings, counterfactual examples), and alignment of xAI explanations with domain expertise (e.g., clinicians, environmental scientists) is encouraged to validate interpretability claims.

4. Conclusions

In conclusion, this review provides valuable insight into machine learning and explainable models in air pollution risk assessment for respiratory health outcomes. This review highlights several issues including methodological issues relating to xAI integration in ERT algorithm, fusion of uncertainty quantification methods into best-performing AI models, federated learning and causal-xAI-ML models. These provide some level of assurance to policymakers within the arena of healthcare application. Furthermore, the challenges identified and the implication of the findings contribute greatly to helping AI-based healthcare application developers find alternative strategies aside from the traditional methods of risk and exposure–response assessment models within the respiratory health outcome arena. Ultimately, no single xAI technique is universally optimal; the choice depends on data modality, computational constraints and end-user needs. Future studies should aim to triangulate findings using multiple xAI methods and to validate interpretability outputs against expert judgment to enhance trustworthiness in clinical and policy translation. Again, future research should systematically benchmark uncertainty quantification techniques alongside performance metrics to enable evidence-based policymaking.

Supplementary Materials

The following supporting information can be downloaded at https://www.mdpi.com/article/10.3390/atmos16101154/s1.

Author Contributions

Conceptualization, I.E.A. and I.C.O.; methodology, I.E.A.; data curation, I.E.A. and I.C.O.; writing—original draft preparation, I.E.A. and I.C.O.; writing—review and editing, I.C.O.; supervision, I.C.O. All authors have read and agreed to the published version of the manuscript.

Funding

This work is funded by the Centre for Global Change, Sol Plaatje University, with National Research Foundation (NRF) (Number: 136097).

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

No new data were created or analyzed in this study.

Conflicts of Interest

The authors declare no conflicts of interest.

Abbreviations

The following abbreviations are used in this manuscript:
AbbreviationMeaning
AAAcceleration of Aging
WHO World Health Organization
ECElemental Carbon
AIArtificial Intelligence
xAIExplainable Artificial Intelligence
MLMachine Learning
PRISMAPreferred Reporting Items for Systematic Reviews and Meta-Analyses
OPEsorganophosphates
APSAverage Particle Size
KPGTKnowledge-guided Pre-training of Graph
LDSALung Deposited Surface Area
PNCParticle Number Concentration
PAEsPhthalates
RMSERoot-Mean-Square Error
MERSMiddle East respiratory syndrome
SARSSevere Acute Respiratory Syndrome
UFPsUltrafine Particles
BCBlack Carbon
BDBurden of Disease
DTDecision Tree
KKNK-Nearest Neighbors
NBNaive Bayesian
HAPHazardous Air Pollutants
AQIAir Quality Index
COPDChronic Obstructive Pulmonary Disease
LPO-XGBoostLung Performance Optimization-based XGBoost
PEPPopulation Exposure to PM2.5
CARIComprehensively Integrated Air-Risk Index
MPMycoplasma Pneumoniae
XGBoost eXtreme Gradient Boosting
GCNGraph Convolutional Neural Network
ERTExtremely Randomized Tree
SEASocioeconomic Activity
SARIMASeasonal Auto-Regressive Integrated Moving Average
DNNDeep Neural Networks
Deep-CNN Deep Convolutional Neural Networks
BaPBenzopyrene
DLNMsDistributed Lag Nonlinear Models
EPAEnvironmental Protection Agency
LECR Lung Excess Cancer Risk
VOCsVolatile Organic Compounds
ERAE-Waste Recycling Areas
PLSpartial least squares regression
phyMTDNNPhysics-informed multi-task deep neural network
CTComputed Tomography
HOCsHydrophobic Organic Compounds
ARAndrogen Receptors
CTMChemical Transport Modeling
MMFMeasurement Model Fusion
QSARQuantitative Structure–Activity Relationship
LIMELocal Interpretable Model-Agnostic Explanations
GAMGeneralized Additive Model
HAQIHealth-Risk Air Quality Index
HQHazard Quotients
RFRandom Forest
LRLogistic Regression
RT-PCRReverse transcription–polymerase chain reaction

Appendix A

Appendix A.1. Search String

(TITLE-ABS-KEY (“Machine learning”) AND TITLE-ABS-KEY (“Explainable Artificial Intelligence”) AND TITLE-ABS-KEY (“air pollution”) OR TITLE-ABS-KEY (“Air pollution risk assessment”) OR TITLE-ABS-KEY (“Respiratory Health Outcomes”)) AND PUBYEAR > 2020 AND PUBYEAR < 2026 AND (LIMIT-TO (DOCTYPE, “ar”) OR LIMIT-TO (DOCTYPE, “cp”) OR LIMIT-TO (DOCTYPE, “re”)) AND ( LIMIT-TO (LANGUAGE, “English”))

Appendix A.2. Summary of Related Literature on ML for Air Pollution Risk Assessment

Table A1. Related Literature on ML for Air Pollution Risk Assessment.
Table A1. Related Literature on ML for Air Pollution Risk Assessment.
AuthorsAir PollutantFocusMachine Learning Models Risks Respiratory Health Outcome Findings Best Model
Famiyeh et al. [76]PAHs pollution Employed a component-based potency factor approach to estimate LECR (lung excess cancer risk) in Ningbo.
Potency factors are BaP unit risk values from both the WHO and the Environmental Protection Agency (EPA).
ML algorithms: RF, extremely randomized trees (ERT), and extreme gradient boosting (XGBoost), to enhance the accuracy of source-specific LECR assessments. Lung excess cancer risk LECR estimation can effectively mitigate PAH pollution and reduce lung cancer risks. ERT identified the primary contributors to elevated LECR in Ningbo from sources such as industrial emissions, coal combustion, and gasoline engine exhaust.
A moderate PAH exposure risk level > 1.0 × 10−6
ERT
Abbafati et al. [77]PM2.5, PM10 Investigate relationship between type 2 diabetes (T2DM) and exposure to PM2.5 and PM10. Random effects models and non-parametric methods to assess association between air pollution and T2DM. Type-2 diabetes (T2DM) -Statistically significant relationship between T2DM and PM2.5.
T2DM incidence rates were significantly negatively associated with time (coefficient = −0.07961, p  <  0.01), indicating a decreasing trend over time.
Increases in the ratio of PM2.5 to PM10 (pwratio) were significantly positively associated with increases in T2DM incidence rates (coefficient  =  0.52304, p  <  0.01).
-
Weichenthal et al. [78]ultrafine particles (UFPs, <0.1 μm) and black carbon (BC)Examine long-term health consequences of traffic pollutants (ultrafine particles) and BC in urban areas.
Estimate associations between long-term exposures to outdoor UFPs and BC and non-accidental and cause-specific mortality.
exposure assessment models:
(1) land use regression (LUR) models;
(2) CNN trained with mobile monitoring data and aerial images;
(3) combined LUR and CNN.
Cox proportional hazard model estimated association between long-term exposures to outdoor UFPs and BC, and potential confounding factors (that is, socio-demographic factors and co-pollutants identified) were adjusted.
Mortality -Outdoor UFP were positively associated with mortality, independent of other outdoor air pollutants like PM2.5 mass concentrations and oxidant gases (e.g., nitrogen dioxide (NO2) and O3). Combined LUR and CNN performed slightly better than LUR model.
Wang et al. [79]Volatile Organic Compounds (VOCs) Examine health risks from exposure to volatile organic compounds (VOCs) in e-waste recycling areas (ERA) and predict the presence of e-waste pollution Non-machine learning model (ultrahigh performance liquid chromatography) and Bayesian kernel machine regression model--Significantly higher levels of VOC exposure and oxidative damage biomarkers (ODBs) among e-waste workers-
Wang et al. [80]PM2.5 bound mercury (PBM2.5) Explore the driving factors, spatiotemporal pollution distribution and associated health risks of PBM2.5RF applied to predict PBM2.5 concentration.
Health risk assessment model
Non-carcinogenic risk of PBM2.5 was negligible in the year 2020, with HQ value mostly < 0.02 during winter. Population density and PM2.5 from power generation stations contributed mainly to PBM2.5 concentration.RF
Wang et al. [81]Airborne trace element (TE) in PM2.5quantify variations in pollution characteristics and health risks of TEs Weather normalization and health risk assessment models.Threat to human health and ecosystems. TEs show dual effects (i.e., a tendency towards higher TE concentrations due to meteorological factors and its health-related risks during polluted period).
Selenium (Se), manganese (Mn), and lead (Pb) are the most meteorologically influenced TEs, whereas chromium (Cr) and manganese (Mn) are the dominant TEs that pose health risks
-
Chen et al. [82]Fourteen common air pollutants Examine habitual cooking, indoor air pollutants and the risk of lung cancer. Gas chromatography–mass spectrometry.risk of lung adenocarcinomalung cancerFrequent cooking and indoor incense burning increase the risk of lung adenocarcinoma. -
Yang et al. [83]PM1, PM2.5, PM10, and NO2Examine air pollution and visual impairmentMachine learning methods.visual impairment Exposure to air pollution were positively associated with the odds of visual impairment.
However, changes in a child’s age, gender, and area of residence, parent level of education and smoking of cigarette may change the association
Yang et al. [84]PM1 and PM2.5Examine long-term exposure to PM1, PM2.5 and children’s lung function. -Lung functionLung functionPM1 may be very hazardous to children’s respiratory health than PM2.5 exposure.
Al Noaimi et al. [85]PM2.5, NO2, SO2 Maternal exposure of air pollutants-Higher birth defect (BD) risk, neural tube defects (NTD), genitourinary defects (GUD) risk Exposure to PM2.5 during the first trimester is significantly associated with a higher overall.
Maternal exposure to NO2 during BD Gestational Time Window of Risk (GTWR) showed a significant protective effect for neural tube defects (NTD).
Maternal exposure to SO2 during GTWR showed a significant association with a higher genitourinary defects (GUD) risk
Li et al. [6]Air toxics (that is, hazardous air pollutants) Examine multi-air toxic combinations and its association with disease like asthma. Machine learning Asthma outcomesmulti-air toxic significantly associated with asthma outcomes
Wang et al. [86]Prenatal ambient fine particulate matter PM2.5Prenatal ambient fine particulate matter PM2.5 exposure on early childhood neurodevelopment General linear mixed model with binomially distributed errorsRisk of suspected developmental delay (SDD) in children, specifically in problem-solving context for girls. Prenatal PM2.5 exposure affected early childhood neurodevelopment.
Xu et al. [87]Chemical constituents (that is, organic matter OM, black carbon (BC), sulfate (SO42−), nitrate (NO3), ammonium (NH4+), and soil dust (Dust-PM2.5)). investigate association between prenatal exposure to PM2.5 and neurodevelopment in infants (1 year)Machine learning model estimates daily concentration.
Geospatial-statistical model evaluates average concentration of the chemical constituents.
Risk of a child’s non-optimal gross motor development. -Prenatal exposure to PM2.5, and with high SO42− concentration, were related to children’s non-optimal gross motor development. However, short- and long-term effects of perinatal PM2.5 exposure on children’s neurodevelopment merit further investigation.
Zhang et al. [88]PM2.5Estimate spatiotemporal heterogeneity in fine particle concentration and its health risks exposure and inhalation of PM2.5. Combine three-dimensional landscape pattern index (TDLPI) and extreme gradient boosting (XGBOOST) to improve LUR model (LTX) The health risks of human exposure to fine particles were moderately high in winter in the study area.LTX (RMSE of 8.73 μg/m3
Huang et al. [20]PM2.5, PM10, sulfur dioxide (SO2) and nitrogen dioxide (NO2) Examine association between long-term exposure to multiple air pollutants and cardiopulmonary mortality.
Identify air pollutant contributing to mortality risk
Satellite-derived machine learning model.
Time-varying Cox proportional hazards model evaluated individual association between air pollutants and mortality from non-accidental causes, cardiovascular diseases (CVDs), non-malignant respiratory diseases (RDs) and lung cancer, accounting for demographic and socioeconomic factors.
Mortality risks linked to air pollutant mixturelung cancer PM2.5 regularly contributed the most to high mortality risks associated with air pollutant mixture, followed by SO2 or PM10.
There was strong positive association of long-term individual and joint exposure to PM10, PM2.5, SO2, and NO2.
PM2.5 is potentially the main contributor to mortalities from non-accidental causes, CVDs, non-malignant RDs and lung cancer in high-exposure settings.
Sun et al. [65]PM2.5, SO4, NO3, ammonium (NH4+), and chloride (Cl) investigate the individual and joint mortality risks related to PM2.5 inorganic chemical compositions, and identify primary contributorsSatellite-based machine learning model calculated the chemical compositions.
Time-varying Cox proportional hazards model analyzed associations between the chemical compositions
Cardiopulmonary mortality.Risk of cardiorespiratory mortality.Higher incomes earners with lower level of education were more vulnerable to inorganic chemical exposure.
Long-term exposure to higher levels of PM2.5 inorganic compositions was connected to significantly increase cardiopulmonary mortality, with SO42− as the potential primary contributor.
PM2.5 sources impact health.
Joint exposure model shows that simultaneous rise in one IQR in all four compositions, increased the risk of cardiorespiratory mortality by at least 36.3%, with long-term exposure to SO42− contributing the most to non-accidental and cardiopulmonary deaths.
Gao et al. [89]Chemical components (sulfate(SO2−4), ammonium (NH4+)), PM2.5Examine Chemical components and the expiratory airflow limitation (EAL) in adult. Land use regression model predicts the exposure to six air pollutants.
PM2.5 was determined using a validated machine-learning algorithm.
Logistic regression model employed to estimate effect sizes.
Pulmonary function evaluated with medical-grade pulmonary function analyzer.
Air pollution score (APS) was associated with a 25% higher risk of EAL.-PM2.5 exposure had the sturdiest link with the risk of EAL.
Combined effects of air pollution increased the risk of EAL in youth, where SO2−4 and NH4+ predominantly contributed to chemical components

Appendix A.3. xAI Technique Integrated in ML for Air Pollution Risk Assessment in the Literature

Table A2. xAI Technique in ML for Air Pollution Risk Assessment in the Literature.
Table A2. xAI Technique in ML for Air Pollution Risk Assessment in the Literature.
AuthorAir PollutantFocusRespiratory Health OutcomeML ModelxAI ModelAccuracy MeasureFindings
Abdillah et al. [90]UFPs, PM0.1 Assessed UFP number concentrations (UFPs PNC) exposure dose for healthy adults and children.Inhaled UFPs were mainly deposited in tracheobronchial (TB) respiratory fraction for adults (67.7%) and in alveoli (ALV) fraction for children (67.5%)Multiple linear regression (MLR), XGBoost and RF.SHAP analysis with the XGBoost model XGBoost has highest estimation performance with RMSE (0.79)Spatial variability was successfully pointed out for each roadside.
Xu et al. [91]PM2.5, organic matter (OM), nitrate (NO3) and ammonium (NH4+),black carbon (BC) and sulfate (SO42−) Assessed the concentration thresholds at which components of PM2.5 affect mortality. -LightGBM model based on Bayesian. SHAP algorithms to identify the concentration thresholds at which the components of PM2.5 affect mortality-The mortality rates influenced by five PM2.5 components suggesting a consistent downward trend.
Relative importance OM, NO3 and NH4+ in influencing mortality increased by 6.3%, 17.4% and 4%, respectively.
Relative importance of BC and SO42− in influencing mortality decreased rapidly to approx. 2%.
Jing et al. [92]nature park visits and environmental factors (PM2.5)Investigate the associations between nature park visits and adult asthma risk in urbanized areas. Exposure–response relationship shows that increasing park visits reduces asthma risk and enhance respiratory health in urban settings, but the protective effect plateaus when visits exceed 51.94 per year.XGBoost, NDVI (Normalized Difference Vegetation Index) SHAP -High urban heat island index (UHI) and high PM2.5 levels increased asthma risk.
High PM2.5 levels (AP = 0.24, 95% CI: 0.15 to 0.32) increased asthma risk.
Wang et al. [64]PM2.5, PM10, Effects of airborne particulate matter on climate and human health -LightGBM, XGBoost SHAP used to separate meteorological contributions because of the strong influence on PM concentration.lightGBM model trains 45 times faster than the XGBoost The SHAP technique had good agreement with meteorological normalization approach to separate meteorological contributions (R2 > 0.5).

References

  1. Jitkajornwanich, K.; Vijaranakul, N.; Jaiyen, S.; Srestasathiern, P.; Lawawirojwong, S. Enhancing risk communication and environmental crisis management through satellite imagery and AI for air quality index estimation. MethodsX 2024, 12, 102611. [Google Scholar] [CrossRef]
  2. Hang, Y.; Meng, X.; Li, T.; Wang, T.; Cao, J.; Fu, Q.; Dey, S.; Li, S.; Huang, K.; Liang, F.; et al. Assessment of long-term particulate nitrate air pollution and its health risk in China. iScience 2022, 25, 104899. [Google Scholar] [CrossRef]
  3. Pajot, A.; Yapo, M.; Coulibaly, S.; Doumbia, M.; Gnamien, S.; Kouao, K.; Ahoua, S.; Adjoua Dje, S.; Liousse, C.; Moh, R.; et al. Air pollution exposure, respiratory consequences, and perceptions among urban African children living in poor conditions—A case study in Abidjan, Côte d’Ivoire. PLoS Glob. Public Health 2025, 5, e0003703. [Google Scholar] [CrossRef]
  4. Sankar, L.; Akrasu, K. Efficient multi-station air quality prediction in Delhi with wavelet and optimization-based model. PLoS ONE 2025, 20, e0330465. [Google Scholar] [CrossRef]
  5. Zaben, S.O.; Zainon, W.M.N.W.; Sabry, A.H. Machine learning-based methods for detecting respiratory abnormalities using audio and visual analysis: A review. Results Eng. 2025, 26, 104744. [Google Scholar] [CrossRef]
  6. Li, Y.C.; Hsu, H.L.; Chun, Y.; Chiu, P.H.; Arditi, Z.; Claudio, L.; Pandey, G.; Bunyavanich, S. Machine learning-driven identification of early-life air toxic combinations associated with childhood asthma outcomes. J. Clin. Investig. 2021, 131, e152088. [Google Scholar] [CrossRef]
  7. Neo, E.X.; Hasikin, K.; Mokhtar, M.I.; Lai, K.W.; Azizan, M.M.; Razak, S.A.; Hizaddin, H.F. Towards Integrated Air Pollution Monitoring and Health Impact Assessment Using Federated Learning: A Systematic Review. Front. Public Health 2022, 10, 851553. [Google Scholar] [CrossRef]
  8. Ye, Y.; Aizezi, N.; Feng, J.; Han, B.; Li, X.; Su, Z.; Li, L.; Liu, Y. Advanced Characterization of Industrial Smoke: Particle Composition and Size Analysis with Single Particle Aerosol Mass Spectrometry and Optimized Machine Learning. Anal. Chem. 2025, 97, 5554–5562. [Google Scholar] [CrossRef]
  9. Zylowski, T. Study on Criteria for Explainable AI for Laypeople. In CEUR Workshop Proceedings, Proceedings of the Joint of 45th German Conference on Artificial Intelligence Workshops, Tutorials and Doctoral Consortium, KI-JP 2022, Trier, Germany, September 19–23 2022; CEUR-WS: Aachen, Germany, 2022. [Google Scholar]
  10. Leivaditis, V.; Maniatopoulos, A.A.; Lausberg, H.; Mulita, F.; Papatriantafyllou, A.; Liolis, E.; Beltsios, E.; Adamou, A.; Kontodimopoulos, N.; Dahm, M. Artificial Intelligence in Thoracic Surgery: A Review Bridging Innovation and Clinical Practice for the Next Generation of Surgical Care. J. Clin. Med. 2025, 14, 2729. [Google Scholar] [CrossRef] [PubMed]
  11. Lin, Z.; Lin, S.; Neamtiu, I.A.; Ye, B.; Csobod, E.; Fazakas, E.; Gurzau, E. Predicting environmental risk factors in relation to health outcomes among school children from Romania using random forest model—An analysis of data from the SINPHONIE project. Sci. Total Environ. 2021, 784, 147145. [Google Scholar] [CrossRef] [PubMed]
  12. Chadalavada, S.; Faust, O.; Salvi, M.; Seoni, S.; Raj, N.; Raghavendra, U.; Gudigar, A.; Barua, P.D.; Molinari, F.; Acharya, R. Application of artificial intelligence in air pollution monitoring and forecasting: A systematic review. Environ. Model. Softw. 2025, 185, 106312. [Google Scholar] [CrossRef]
  13. Masood, A.; Ahmad, K. A review on emerging artificial intelligence (AI) techniques for air pollution forecasting: Fundamentals, application and performance. J. Clean. Prod. 2021, 322, 129072. [Google Scholar] [CrossRef]
  14. Brereton, P.; Kitchenham, B.A.; Budgen, D.; Turner, M.; Khalil, M. Lessons from applying the systematic literature review process within the software engineering domain. J. Syst. Softw. 2007, 80, 571–583. [Google Scholar] [CrossRef]
  15. Subramaniam, S.; Raju, N.; Ganesan, A.; Rajavel, N.; Chenniappan, M.; Prakash, C.; Pramanik, A.; Basak, A.K.; Dixit, S. Artificial Intelligence Technologies for Forecasting Air Pollution and Human Health: A Narrative Review. Sustainability 2022, 14, 9951. [Google Scholar] [CrossRef]
  16. Liu, D.; Cheng, K.; Huang, K.; Ding, H.; Xu, T.; Chen, Z.; Sun, Y. Visualization and Analysis of Air Pollution and Human Health Based on Cluster Analysis: A Bibliometric Review from 2001 to 2021. Int. J. Environ. Res. Public Health 2022, 19, 12723. [Google Scholar] [CrossRef]
  17. Ye, T.; Yu, P.; Wen, B.; Yang, Z.; Huang, W.; Guo, Y.; Abramson, M.J.; Li, S. Greenspace and health outcomes in children and adolescents: A systematic review. Environ. Pollut. 2022, 314, 120193. [Google Scholar] [CrossRef]
  18. Vachon, J.; Kerckhoffs, J.; Buteau, S.; Smargiassi, A. Do machine learning methods improve prediction of ambient air pollutants with high spatial contrast? A systematic review. Environ. Res. 2024, 262 Pt 2, 119751. [Google Scholar] [CrossRef] [PubMed]
  19. Kibria, H.B.; Hossain, M.A.; Rehman, S.; Alahakoon, D.; Rahman, M.A. An explainable lightweight parallel depth-wise separable model for lung infection detection from chest X-rays. Neural Comput. Appl. 2025, 37, 4545–4566. [Google Scholar] [CrossRef]
  20. Huang, W.; Zhou, Y.; Chen, X.; Zeng, X.; Knibbs, L.D.; Zhang, Y.; Jalaludin, B.; Dharmage, S.C.; Morawska, L.; Guo, Y.; et al. Individual and joint associations of long-term exposure to air pollutants and cardiopulmonary mortality: A 22-year cohort study in Northern China. Lancet Reg. Health West. Pac. 2023, 36, 100776. [Google Scholar] [PubMed]
  21. Liu, K.; Li, S.; Qian, Z.M.; Dharmage, S.C.; Bloom, M.S.; Heinrich, J.; Jalaludin, B.; Markevych, I.; Morawska, L.; Knibbs, L.D.; et al. Benefits of influenza vaccination on the associations between ambient air pollution and allergic respiratory diseases in children and adolescents: New insights from the Seven Northeastern Cities study in China. Environ. Pollut. 2020, 256, 113434. [Google Scholar] [CrossRef]
  22. Atzeni, M.; Cossu, L.; Gaiotti, S.; Cappon, G.; Tine, M.; Previtero, D.; Padrin, Y.; Baraldo, S.; Semenzato, U.; Vettoretti, M. AirPredict: An eHealth platform for asthma management leveraging wearable sensors, digital diaries, and air quality monitoring to optimize patient outcomes. Front. Digit. Health 2025, 7, 1573342. [Google Scholar] [CrossRef]
  23. Rowland, S.N.; Green, C.G.; Halliwill, J.R.; Singanayagam, A.; Heaney, L.M. Gut feelings on short-chain fatty acids to regulate respiratory health. Trends Endocrinol. Metab. 2025. [Google Scholar] [CrossRef]
  24. Renouf, B.; Sutanto, E.N.; Kidd, C.; Lim, J.; Amin, M.; Berry, L.; Hoyne, G.F.; D’Vaz, N.; Kicic-Starcevich, E.; Stick, S.M.; et al. Profiling epithelial viral receptor expression in amniotic membrane and nasal epithelial cells at birth. Placenta 2025, 160, 82–88. [Google Scholar] [CrossRef]
  25. Xu, J.; Yang, W.; Bai, Z.; Zhang, R.; Zheng, J.; Wang, M.; Zhu, T. Modeling spatial variation of gaseous air pollutants and particulate matters in a Metropolitan area using mobile monitoring data. Environ. Res. 2022, 210, 112858. [Google Scholar] [CrossRef]
  26. Li, Z.; Chen, Y.; Tao, Y.; Zhao, X.; Wang, D.; Wei, T.; Hou, Y.; Xu, X. Mapping the personal PM2.5 exposure of China’s population using random forest. Sci. Total Environ. 2023, 871, 162090. [Google Scholar] [CrossRef]
  27. Maio, S.; Fasola, S.; Marcon, A.; Angino, A.; Baldacci, S.; Bilò, M.B.; Bono, R.; La Grutta, S.; Marchetti, P.; Sarno, G.; et al. Relationship of long-term air pollution exposure with asthma and rhinitis in Italy: An innovative multipollutant approach. Environ. Res. 2023, 224, 115455. [Google Scholar] [CrossRef]
  28. Tang, C.X.; Dong, Y.; Yuan, X.Y.; Wang, R.; Wu, C.C.; Bao, L.J.; Zeng, E.Y. Effects of organic carbon/elemental carbon and particle size on inhalation bioaccessibility of particle-bound PAHs. Sci. Total Environ. 2023, 889, 164225. [Google Scholar] [CrossRef] [PubMed]
  29. Yang, Q.; Yuan, Q.; Gao, M.; Li, T. A new perspective to satellite-based retrieval of ground-level air pollution: Simultaneous estimation of multiple pollutants based on physics-informed multi-task learning. Sci. Total Environ. 2023, 857 Pt 2, 159542. [Google Scholar] [CrossRef] [PubMed]
  30. Zhang, B.; Zhang, Y.; Zhang, K.; Zhang, Y.; Ji, Y.; Zhu, B.; Liang, Z.; Wang, H.; Ge, X. Machine learning assesses drivers of PM2.5 air pollution trend in the Tibetan Plateau from 2015 to 2022. Sci. Total Environ. 2023, 878, 163189. [Google Scholar] [CrossRef] [PubMed]
  31. Zhou, S.; Guo, Y.; Su, T.; Chen, G.; Liu, H.; Li, Q.; Bao, H.; Ji, Y.; Luo, S.; Liu, Z.; et al. Individual and joint effect of indoor air pollution index and ambient particulate matter on fetal growth: A prospective cohort study. Int. J. Epidemiol. 2023, 52, 690–702. [Google Scholar] [CrossRef]
  32. Ahmadian, F.; Rajabi, S.; Maleky, S.; Baghapour, M.A. Spatiotemporal analysis of airborne pollutants and health risks in Mashhad metropolis: Enhanced insights through sensitivity analysis and machine learning. Environ. Geochem. Health 2025, 47, 34. [Google Scholar] [CrossRef] [PubMed]
  33. Chen, J.; Zhu, S.; Wang, P.; Zheng, Z.; Shi, S.; Li, X.; Xu, C.; Yu, K.; Chen, R.; Kan, H.; et al. Predicting particulate matter, nitrogen dioxide, and ozone across Great Britain with high spatiotemporal resolution based on random forest models. Sci. Total Environ. 2024, 926, 171831. [Google Scholar] [CrossRef]
  34. Guo, C.; Yang, J.; Ma, J.; Chen, J.; Chen, S.; Zheng, Y.; Huang, B.; Yu, J.; Li, T.; He, S. Ambient fine particulate matter and its constituents may exacerbate the acceleration of aging in adults. Environ. Int. 2024, 192, 109019. [Google Scholar] [CrossRef]
  35. Jiang, P.; Gao, C.; Zhao, J.; Li, F.; Ou, C.; Zhang, T.; Huang, S. An exploration of urban air health navigation system based on dynamic exposure risk forecast of ambient PM2.5. Environ. Int. 2024, 190, 108793. [Google Scholar] [CrossRef]
  36. Lv, L.; Wei, P.; Hu, J.; Chu, Y.; Liu, X. High-spatiotemporal-resolution mapping of PM2.5 traffic source impacts integrating machine learning and source-specific multipollutant indicator. Environ. Int. 2024, 183, 108421. [Google Scholar] [CrossRef]
  37. Masseran, N.; Safari, M.A.M.; Tajuddin, R.R.M. Probabilistic classification of the severity classes of unhealthy air pollution events. Environ. Monit. Assess. 2024, 196, 523. [Google Scholar] [CrossRef]
  38. Chen, L.; Yousaf, M.; Xu, J.; Ma, X. Ultrafine particles: Sources, toxicity, and deposition dynamics in the human respiratory tract —experimental and computational approaches. J. Environ. Manag. 2025, 376, 124458. [Google Scholar] [CrossRef] [PubMed]
  39. Amini, H.; Bergmann, M.L.; Taghavi Shahri, S.M.; Tayebi, S.; Cole-Hunter, T.; Kerckhoffs, J.; Khan, J.; Meliefste, K.; Lim, Y.H.; Mortensen, L.H.; et al. Harnessing AI to unmask Copenhagen’s invisible air pollutants: A study on three ultrafine particle metrics. Environ. Pollut. 2024, 346, 123664. [Google Scholar] [CrossRef] [PubMed]
  40. Cao, F.; Zhao, X.; Fu, X.; Jin, Y. Computational insights into exploring the potential effects of environmental contaminants on human health. Sci. Rep. 2025, 15, 11779. [Google Scholar] [CrossRef]
  41. Chen, Z.Y.; Achebak, H.; Petetin, H.; Méndez Turrubiates, R.F.; Guo, Y.; Pérez García-Pando, C.; Ballester, J. Trends in population exposure to compound extreme-risk temperature and air pollution across 35 European countries: A modelling study. Lancet Planet. Health 2025, 9, e384–e396. [Google Scholar] [CrossRef]
  42. Gao, K.; Hua, K.; Wang, S.; Chen, X.; Zhu, T. Exploring the reproductive exposure risks of phthalates and organophosphates in atmospheric particulate matter based on quantitative structure-activity relationships and network toxicology models. J. Hazard. Mater. 2025, 488, 137395. [Google Scholar] [CrossRef]
  43. Gao, K.; Hua, K.; Chen, X.; Zheng, C.; Li, X.; Wu, Q.; Ji, L.; Wang, L.; Wei, W.; Lu, L. Occurrence, Characteristics, and Mixed Reproductive Exposure Risk Assessment of Traditional Phthalates and Their Novel Alternatives in Campus Indoor Dust. Environ. Sci. Technol. 2025, 59, 6708–6718. [Google Scholar] [CrossRef]
  44. Holloway, T.; Bratburd, J.R.; Fiore, A.M.; Kerr, G.H.; Mao, J. Satellite data to support air quality assessment and management. J. Air Waste Manag. Assoc. 2025, 75, 429–463. [Google Scholar] [CrossRef] [PubMed]
  45. Kuo, C.P.; Fu, J.S.; Liu, Y. Perspective improvement of regional air pollution burden of disease estimation by machine intelligence. Front. Public Health 2025, 13, 1436838. [Google Scholar] [CrossRef]
  46. Lee, S.J.; Cho, I.G.; Lee, H.Y.; Ju, J.T.; Shin, H.J.; Choi, S.D. Development of a comprehensive air risk index and its application to high spatial-temporal health risk assessment in a large industrial city. Environ. Pollut. 2025, 367, 125545. [Google Scholar] [CrossRef] [PubMed]
  47. Liu, Y.; Jin, B.; Zhang, X.; Liu, X.; Wang, T.; Thuy Dinh, V.N.; Jaffrezo, J.L.; Uzu, G.; Dominutti, P.; Darfeuil, S.; et al. Source apportionment of PM(10) particles in the urban atmosphere using PMF and LPO-XGBoost. Environ. Res. 2025, 278, 121659. [Google Scholar] [CrossRef] [PubMed]
  48. Wenjie, W.; Rui, L.; Pengpeng, Z.; Chao, D.; Donglin, Z. Integrated network toxicology, machine learning and molecular docking reveal the mechanism of benzopyrene-induced periodontitis. BMC Pharmacol. Toxicol. 2025, 26, 118. [Google Scholar] [CrossRef]
  49. Yang, H.M.; Wang, J.Y.; Li, C.; Zhang, Y.Q.; Wang, R.; Yang, Q.; Yao, Y.; Wang, Z.; Xu, S.L.; Huang, H.H.; et al. Is there an association between eye-level greenness and childhood hypertension using street view? Findings from the Seven Northeastern Cities study in China. Environ. Res. 2025, 268, 120768. [Google Scholar] [CrossRef]
  50. Zhao, P.; Zhao, P.; Zhan, Z.; Dai, Q.; Casuccio, G.S.; Gao, J.; Li, J.; He, Y.; Qian, H.; Bi, X.; et al. Advancing Source Apportionment of Atmospheric Particles: Integrating Morphology, Size, and Chemistry Using Electron Microscopy Technology and Machine Learning. Environ. Sci. Technol. 2025, 59, 3645–3655. [Google Scholar] [CrossRef]
  51. Paul, N.; Yao, J.; McLean, K.E.; Stieb, D.M.; Henderson, S.B. The Canadian Optimized Statistical Smoke Exposure Model (CanOSSEM): A machine learning approach to estimate national daily fine particulate matter PM2.5 exposure. Sci. Total Environ. 2022, 850, 157956. [Google Scholar] [CrossRef]
  52. Brooks, N.; Biswas, D.; Hossin, R.; Yu, A.; Saha, S.; Saha, S.; Saha, S.K.; Luby, S.P. Health consequences of small-scale industrial pollution: Evidence from the brick sector in Bangladesh. World Dev. 2023, 170, 106318. [Google Scholar] [CrossRef]
  53. Hu, C.Y.; Gutierrez-Avila, I.; He, M.Z.; Lavigne, É.; Alcala, C.S.; Yitshak-Sade, M.; Lamadrid-Figueroa, H.; Tamayo-Ortiz, M.; Mercado-Garcia, A.; Just, A.C.; et al. Windows of susceptibility and joint effects of prenatal and postnatal ambient air pollution and temperature exposure on asthma and wheeze in Mexican children. Environ. Int. 2024, 193, 109122. [Google Scholar] [CrossRef]
  54. Okello, G.; Nantanda, R.; Tatah, L.; Sserunjogi, R.; Johnson, O.; Awokola, B.; Okure, D.; Thondoo, M.; Green, P.; Babajide, O.; et al. Association between ambient air pollution and respiratory health in Kampala, Uganda: Implications for policy and practice. Urban Clim. 2024, 58, 102128. [Google Scholar] [CrossRef]
  55. Bouma, F.; Hoek, G.; Koppelman, G.H.; Vonk, J.M.; Janssen, N.A.H.; van Ratingen, S.; Hendricx, W.; Wesseling, J.; Kerckhoffs, J.; Vermeulen, R.; et al. Comparison of air pollution exposure assessment methods and the association with children’s respiratory health. Environ. Int. 2025, 198, 109407. [Google Scholar] [CrossRef]
  56. Leão, M.L.P.; Zhang, L.; da Silva Júnior, F.M.R. Effect of particulate matter PM2.5 and PM10 on health indicators: Climate change scenarios in a Brazilian metropolis. Environ. Geochem. Health 2023, 45, 2229–2240. [Google Scholar] [CrossRef] [PubMed]
  57. Shams, S.R.; Choi, Y.; Singh, D.; Kayastha, S.; Park, J. Deep learning-based forecasting of daily maximum ozone levels and assessment of socioeconomic and health impacts in South Korea. Sci. Total Environ. 2025, 983, 179684. [Google Scholar] [CrossRef]
  58. Chen, F.; Zhang, W.; Mfarrej, M.F.B.; Saleem, M.H.; Khan, K.A.; Ma, J.; Raposo, A.; Han, H. Breathing in danger: Understanding the multifaceted impact of air pollution on health impacts. Ecotoxicol. Environ. Saf. 2024, 280, 116532. [Google Scholar] [CrossRef] [PubMed]
  59. Teyton, A.; Baer, R.J.; Benmarhnia, T.; Bandoli, G. Exposure to Air Pollution and Emergency Department Visits during the First Year of Life among Preterm and Full-term Infants. JAMA Netw. Open 2023, 6, E230262. [Google Scholar] [CrossRef] [PubMed]
  60. Danesh-Yazdi, M.; Wang, Y.; Di, Q.; Wei, Y.; Requia, W.J.; Shi, L.; Sabath, M.B.; Dominici, F.; Coull, B.A.; Evans, J.S. Long-Term Association of Air Pollution and Hospital Admissions among Medicare Participants Using a Doubly Robust Additive Model. Circulation 2021, 143, 1584–1596. [Google Scholar] [CrossRef]
  61. Ma, Y.; Zhou, C.; Li, M.; Huang, Q. High-resolution monthly assessment of population exposure to PM2.5 and its relationship with socioeconomic activities using multisource geospatial data. Environ. Monit. Assess. 2025, 197, 342. [Google Scholar] [CrossRef]
  62. Pei, S.; Yang, L.; Gao, H.; Liu, Y.; Dai, E.; Feng, F.; Lu, J. Pollutants-mediated viral hepatitis in different types: Assessment of different algorithms and time series models. Sci. Rep. 2024, 14, 21141. [Google Scholar] [CrossRef] [PubMed]
  63. Shen, Q.; Liu, Y.; Li, G.; An, T. A review of disrupted biological response associated with volatile organic compound exposure: Insight into identification of biomarkers. Sci. Total Environ. 2024, 948, 174924. [Google Scholar] [CrossRef]
  64. Wang, S.; Wang, P.; Zhang, R.; Meng, X.; Kan, H.; Zhang, H. Estimating particulate matter concentrations and meteorological contributions in China during 2000–2020. Chemosphere 2023, 330, 138742. [Google Scholar] [CrossRef]
  65. Sun, H.; Chen, X.; Huang, W.; Wei, J.; Yang, X.; Shan, A.; Zhang, L.; Zhang, H.; He, J.; Pan, C.; et al. Association between Long-Term Exposure to PM2.5 Inorganic Chemical Compositions and Cardiopulmonary Mortality: A 22-Year Cohort Study in Northern China. Environ. Health 2024, 2, 530–540. [Google Scholar] [CrossRef]
  66. Zhang, H.; Yi, H.; Hao, Y.; Zhao, L.; Pan, W.; Xue, Q.; Liu, X.; Fu, J.; Zhang, A. Deciphering exogenous chemical carcinogenicity through interpretable deep learning: A novel approach for evaluating atmospheric pollutant hazards. J. Hazard. Mater. 2024, 465, 133092. [Google Scholar] [CrossRef]
  67. Rajesh, M.; Babu, R.G.; Moorthy, U.; Easwaramoorthy, S.V. Machine learning driven framework for realtime air quality assessment and predictive environmental health risk mapping. Sci. Rep. 2025, 15, 28801. [Google Scholar] [CrossRef]
  68. Razavi-Termeh, S.V.; Sadeghi-Niaraki, A.; Ali, F.; Naqvi, R.A.; Choi, S.-M. Spatio-temporal modeling of asthma-prone areas: Exploring the influence of urban climate factors with explainable artificial intelligence (XAI). Sustain. Cities Soc. 2024, 116, 105889. [Google Scholar] [CrossRef]
  69. Agbehadji, I.E.; Obagbuwa, I.C. Integration of Explainable Artificial Intelligence into Hybrid Long Short-Term Memory and Adaptive Kalman Filter for Sulfur Dioxide (SO2) Prediction in Kimberley, South Africa. Atmosphere 2025, 16, 523. [Google Scholar] [CrossRef]
  70. Agbehadji, I.E.; Obagbuwa, I.C. Spatiotemporal Graph Convolutional Network-Based Long Short-Term Memory Model with A* Search Path Naviga-tion and Explainable Artificial Intelligence for Carbon Mon-oxide Prediction in Northern Cape Province, South Africa. Atmosphere 2025, 16, 1107. [Google Scholar] [CrossRef]
  71. Sunder, M.S.S.; Tikkiwal, V.A.; Kumar, A.; Tyagi, B. Unveiling the Transparency of Prediction Models for Spatial PM2.5 over Singapore: Comparison of Different Machine Learning Approaches with eXplainable Artificial Intelligence. AI 2023, 4, 787–811. [Google Scholar] [CrossRef]
  72. Chang, T.H.; Liu, Y.C.; Lin, S.R.; Chiu, P.H.; Chou, C.C.; Chang, L.Y.; Lai, F.P. Clinical characteristics of hospitalized children with community-acquired pneumonia and respiratory infections: Using machine learning approaches to support pathogen prediction at admission. J. Microbiol. Immunol. Infect. 2023, 56, 772–781. [Google Scholar] [CrossRef]
  73. Chi, C.Y.; Moghadas-Dastjerdi, H.; Winkler, A.; Ao, S.; Chen, Y.P.; Wang, L.W.; Su, P.I.; Lin, W.S.; Tsai, M.S.; Huang, C.H. Clinical Validation of Explainable Deep Learning Model for Predicting the Mortality of In-Hospital Cardiac Arrest Using Diagnosis Codes of Electronic Health Records. Rev. Cardiovasc. Med. 2023, 24, 265. [Google Scholar] [CrossRef]
  74. Li, C.; Xia, Y.; Wang, L. Household unclean fuel use, indoor pollution and self-rated health: Risk assessment of environmental pollution caused by energy poverty from a public health perspective. Environ. Sci. Pollut. Res. Int. 2024, 31, 18030–18053. [Google Scholar] [CrossRef] [PubMed]
  75. Commodore, S.; Christopher, S.; Wolf, B.; Svendsen, E. Assessment of trace elements directly from archived total suspended particulate filters by laser ablation ICP-MS: A case study of South Carolina. J. Trace Elem. Miner. 2023, 3, 100041. [Google Scholar]
  76. Famiyeh, L.; Chen, K.; Tesema, F.B.; Kelly, C.; Ji, D.; Xiao, H.; Tong, L.; Wang, Z.; He, J. Refining source-specific lung cancer risk assessment from PM2.5-bound PAHs: Integrating component-based potency factors and machine learning in Ningbo, China. Ecotoxicol. Environ. Saf. 2025, 297, 118174. [Google Scholar] [CrossRef] [PubMed]
  77. Abbafati, C.; Nieddu, L.; Cattani, G.; Reatini, M.A.; Ponzani, P. Association between air pollution and type II diabetes in Italy from clinical data and population-weighted exposure at the municipality level. Sci. Rep. 2025, 15, 28326. [Google Scholar] [CrossRef] [PubMed]
  78. Weichenthal, S.; Lloyd, M.; Ganji, A.; Simon, L.; Xu, J.; Venuta, A.; Schmidt, A.; Apte, J.; Chen, H.; Lavigne, E.; et al. Long-Term Exposure to Outdoor Ultrafine Particles and Black Carbon and Effects on Mortality in Montreal and Toronto, Canada. Res. Rep. Health Eff. Inst. 2024, 2024, 217. [Google Scholar]
  79. Wang, J.R.; Kuang, H.X.; Liu, Y.; Li, X.Y.; Chen, T.H.; Zhu, X.H.; Fan, R.F.; Xiang, M.D.; Yu, Y.J. Associations between volatile organic compounds exposure and multiple oxidative damage biomarkers: Method development, human exposure, and application for e-waste pollution prediction. Sci. Total Environ. 2024, 956, 177402. [Google Scholar]
  80. Wang, H.; Li, T.; Wang, G.; Peng, Y.; Zhang, Q.; Wang, X.; Ren, Y.; Liu, R.; Yan, S.; Meng, Q.; et al. Significant spatiotemporal changes in atmospheric particulate mercury pollution in China: Insights from meta-analysis and machine-learning. Sci. Total Environ. 2024, 955, 177184. [Google Scholar]
  81. Wang, H.; Guan, X.; Li, J.; Peng, Y.; Wang, G.; Zhang, Q.; Li, T.; Wang, X.; Meng, Q.; Chen, J.; et al. Quantifying the pollution changes and meteorological dependence of airborne trace elements coupling source apportionment and machine learning. Sci. Total Environ. 2024, 948, 174452. [Google Scholar]
  82. Chen, K.C.; Tsai, S.W.; Shie, R.H.; Zeng, C.; Yang, H.Y. Indoor Air Pollution Increases the Risk of Lung Cancer. Int. J. Environ. Res. Public Health 2022, 19, 1164. [Google Scholar] [CrossRef]
  83. Yang, B.Y.; Guo, Y.; Zou, Z.; Gui, Z.; Bao, W.W.; Hu, L.W.; Chen, G.; Jing, J.; Ma, J.; Li, S.; et al. Exposure to ambient air pollution and visual impairment in children: A nationwide cross-sectional study in China. J. Hazard. Mater. 2021, 407, 124750. [Google Scholar] [CrossRef]
  84. Yang, M.; Guo, Y.M.; Bloom, M.S.; Dharmagee, S.C.; Morawska, L.; Heinrich, J.; Jalaludin, B.; Markevychd, I.; Knibbsf, L.D.; Lin, S.; et al. Is PM1 similar to PM2.5? A new insight into the association of PM1 and PM2.5 with children’s lung function. Environ. Int. 2020, 145, 106092. [Google Scholar] [CrossRef]
  85. Al Noaimi, G.; Yunis, K.; El Asmar, K.; Abu Salem, F.K.; Afif, C.; Ghandour, L.A.; Hamandi, A.; Dhaini, H.R. Prenatal exposure to criteria air pollutants and associations with congenital anomalies: A Lebanese national study. Environ. Pollut. 2021, 281, 117022. [Google Scholar] [CrossRef]
  86. Wang, P.; Zhao, Y.; Li, J.; Zhou, Y.; Luo, R.; Meng, X.; Zhang, Y. Prenatal exposure to ambient fine particulate matter and early childhood neurodevelopment: A population-based birth cohort study. Sci. Total Environ. 2021, 785, 147334. [Google Scholar] [CrossRef] [PubMed]
  87. Xu, X.; Tao, S.; Huang, L.; Du, J.; Liu, C.; Jiang, Y.; Jiang, T.; Lv, H.; Lu, Q.; Meng, Q.; et al. Maternal PM2.5 exposure during gestation and offspring neurodevelopment: Findings from a prospective birth cohort study. Sci. Total Environ. 2022, 842, 156778. [Google Scholar] [CrossRef] [PubMed]
  88. Zhang, P.; Yang, L.; Ma, W.; Wang, N.; Wen, F.; Liu, Q. Spatiotemporal estimation of the PM2.5 concentration and human health risks combining the three-dimensional landscape pattern index and machine learning methods to optimize land use regression modeling in Shaanxi, China. Environ. Res. 2022, 208, 112759. [Google Scholar] [CrossRef]
  89. Gao, J.; Zhang, X.; Ding, H.; Bao, Y.; Zhang, C.; Chi, B.; Xia, Y.; Zhao, Y.; Zhang, H. Air pollution exposure, chemical compositions, and risk of expiratory airflow limitation in youth in Northeast China. Ecotoxicol. Environ. Saf. 2024, 285, 117055. [Google Scholar] [CrossRef]
  90. Abdillah, S.F.I.; You, S.J.; Wang, Y.F. Characterizing sector-oriented roadside exposure to ultrafine particles (PM0.1) via machine learning models: Implications of covariates influences on sectors variability. Environ. Pollut. 2024, 359, 124595. [Google Scholar] [CrossRef] [PubMed]
  91. Xu, H.; Luo, W.; Dai, C.; Zhao, S.; Qian, R.; Dong, B.; Li, Z.; Ma, J. Spatiotemporal evolution and risk thresholds of PM2.5 components in China from the human health perspective. Environ. Pollut. 2025, 373, 126194. [Google Scholar] [CrossRef]
  92. Jing, F.; Ma, W.; Li, Z.; Zhou, S.; Ruan, Y.; He, G.; Hu, J.; Liu, T. Nationwide analysis of the association between nature park visits and adult asthma risk in urbanized neighborhoods. Cities 2025, 166, 106301. [Google Scholar] [CrossRef]
Figure 1. PRISMA flow diagram.
Figure 1. PRISMA flow diagram.
Atmosphere 16 01154 g001
Table 1. Methodological frameworks and research focus.
Table 1. Methodological frameworks and research focus.
Author(s)Research FocusMethodology Data RepositoriesNumber of Articles SelectedYear Range
Chadalavada et al. [12]Monitoring and prediction. Systematic review: PRISMA guidelines.PubMed (43), Scopus (224) and Web of Science (145). 642019 to 2023
Masood et al. [13]Application, fundamentals and performance. Systematic review based on the three-phased approach suggested by [14]: plan, conduct review and report the review.Total of 300 articles from Scopus, Web of Science, Ei Compendex, PubMed and ScienceDirect, Harvard University library catalog (HOLLIS).902003 to 2021
Subramaniam et al. [15]Forecasting and human health.Classic narrative review that follows the systematic review technique.Google Scholar, Springer, PubMed, ScienceDirect, IEEE explore, Scopus and Web of Science (WoS). January 2015 through June 2022
Liu et al. [16] Air pollution and health. Bibliometric techniques. -14,955 2001 to 2021
Leivaditis et al. [10] Applications, benefits, limitations and future direction of AI focusing on thoracic surgery. PRISMA and risk of bias assessment were conducted using the Cochrane Risk of Bias Tool and ROBINS-I for non-randomized studies. PubMed, Scopus, WoS and Cochrane Library.36 Studies published up to January 2025
Ye et al. [17]Measurement methods of greenspace, their health outcomes and potential mechanisms.Systematic review.MEDLINE and EMBASE.140 Up to 11 April 2022
Vachon et al., 2024 [18]Compared statistical and machine learning models to predict NO2, ultrafine particles (UFPs) and black carbon (BC).Systematic review.WoS and Scopus. 38 Up to 13 June 2024
Table 2. Model categories and related algorithms.
Table 2. Model categories and related algorithms.
Model TypeType of AlgorithmTypical Datasets UsedValidation ApproachComparability
Tree-based modelsRandom Forest (RF), XGBoost, LightGBMAir quality monitoring station data; meteorological variables.K-fold cross-validation; train/test splitWidely used; strong performance, but feature importance varies across studies.
Deep learning modelsCNN, LSTM, DNN [57]Spatiotemporal pollution data; satellite-based remote sensing; health outcomes.Hold-out validation; temporal splitHandle complex patterns; often limited transparency; high data demands.
Hybrid modelsLUR + ML, SARIMA + MLLand use regression data; time series pollution data.Rolling-origin or temporal CVImprove accuracy but add complexity; results are difficult to generalize.
Statistical baselinesSARIMA, regression modelsTime series pollutant concentration.Train/test split; rolling validationServe as benchmark; often outperforms ML in small datasets.
Table 3. Methodological heterogeneity of xAI-integrated ML studies for air pollution and respiratory health outcomes.
Table 3. Methodological heterogeneity of xAI-integrated ML studies for air pollution and respiratory health outcomes.
AuthorModel(s)Dataset/ApplicationxAI Technique(s)Outcome FocusKey Findings
Zhang et al. [66]GCN (CarcGC)Chemical carcinogenicity (molecular structural graphs)Model-intrinsic interpretabilityAir pollution and lung cancerGCN-based interpretability identified structural predictors of carcinogenicity.
Rajesh et al. [67]Gradient Boosting, RF, XGBoost, LSTMReal-time multi-location AQ and health riskSHAPAir pollutant prediction + AQ classificationAccurate AQ prediction with interpretable pollutant contributions.
Razavi-Termeh et al. [68]XGBoost + Bat algorithmAsthma-prone locations (spatiotemporal)SHAPAsthma risk + seasonal pollutant drivers.Identified seasonal pollutant importance (e.g., PM2.5 in spring, CO in summer).
Agbehadji et al. [69] LSTM + Adaptive Kalman FilterSO2 concentration predictionSHAPAir pollution (no direct health outcomes)Improved interpretability of SO2 estimates; highlighted noise impact.
Agbehadji et al. [70](GeoxAI-GCN-LSTM-A*)CO prediction.LIMEAir pollutionOptimal navigation route with RMSE (0.4059 for 8 h and 0.4124 for 16 h).
Kibria et al. [19] LW-PDS-CovidNet (CNN)Lung infection detection (CT, X-ray, RT-PCR)SHAP, Grad-CAMRespiratory infections (COVID-19, SARS, MERS, etc.)High accuracy (>90%); highlighted key imaging features.
Sunder et al. [71]RF, Gradient Boosting, TP AutoML.PM2.5 temporal predictionxAI (feature importance, SHAP-like methods)Air pollutionMeta-heuristic TP AutoML achieved highest GPI; interpretability aided temporal feature analysis.
Chang et al. [72] ML classifiersPediatric ARI predictionSHAPRespiratory pathogens (adenovirus, influenza, etc.)Identified biomarkers (CRP most important for adenovirus).
Chi et al. [73] D-SHAP (Deep SHAP)Clinical diagnosis codesDeep SHAPRespiratory failure, sepsis, pneumonia, etc.Explanations aligned with physician judgment, improving trust in predictions.
Table 4. Reported performance metrics (selected examples) across ML/xAI studies.
Table 4. Reported performance metrics (selected examples) across ML/xAI studies.
Model(s)Dataset/TaskReported Metric(s)PerformanceFindings
RF, XGBoost, LSTM [68]Real-time AQ and health risk predictionR2 (0.82–0.89), RMSE (between 4.2 and 6.1 µg/m3)High temporal accuracy.SHAP improved interpretability of pollutant drivers.
XGBoost + Bat Alg. [69]Asthma-prone locationsAUC (between 0.973 and 0.984)Seasonal variations noted.PM2.5, CO, O3, PM10 dominant factors.
LSTM + Adaptive Kalman Filter [70]SO2 predictionRMSE (0.79)Reduced noise error.No direct health outcomes.
CNN (LW-PDS-CovidNet) [19]Lung infection detectionAccuracy between 92–98%.High diagnostic accuracy.xAI (SHAP, Grad-CAM) explained imaging features
Pediatric ARI ML [72]Clinical infections prediction.AUC between 0.86 and 0.92.Event patterns predictive.SHAP highlighted biomarkers.
D-SHAP model [73]Diagnosis codes and mortality risk.AUC > 0.90Clinician-aligned explanations.Improved clinical trust.
Table 5. Comparison of xAI techniques for air pollution and respiratory health applications.
Table 5. Comparison of xAI techniques for air pollution and respiratory health applications.
xAI TechniqueStrengthsLimitationsBest-Use Scenarios
SHAP (Shapley Additive Explanations)Provides consistent, game-theoretic feature attributions; both global and local interpretability; widely adopted in environmental/epidemiological studies.Computationally expensive; sensitive to correlated features; explanations may be complex for non-technical users.Tabular spatiotemporal data; epidemiological studies; policy translation requiring robust feature attribution.
LIME (Local Interpretable Model-Agnostic Explanations).Fast; flexible; model-agnostic; intuitive local approximations.Unstable across runs; explanations vary with sampling; less reliable for high-stakes contexts.Quick interpretability checks; exploratory analysis; low-stakes decision support.
Grad-CAM (Gradient-weighted Class Activation Mapping).Visual explanations overlayed on images; intuitive for clinicians; highlights regions of interest.Limited to CNNs; qualitative rather than quantitative; can miss non-visual factors.Medical imaging (X-rays, CT scans); radiology applications.
Integrated Gradients.Provides attribution scores along a path from baseline to input; less noisy than saliency maps; theoretically grounded.Requires careful baseline selection; computationally intensive.Deep learning models with structured or sequential data (e.g., respiratory signals, spatiotemporal patterns).
Counterfactual Explanations.Intuitive “what-if” reasoning; directly actionable; useful for clinical/policy interventions.Computationally complex; less standardized; limited adoption in air pollution.-
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Agbehadji, I.E.; Obagbuwa, I.C. Explainable Artificial Intelligence and Machine Learning for Air Pollution Risk Assessment and Respiratory Health Outcomes: A Systematic Review. Atmosphere 2025, 16, 1154. https://doi.org/10.3390/atmos16101154

AMA Style

Agbehadji IE, Obagbuwa IC. Explainable Artificial Intelligence and Machine Learning for Air Pollution Risk Assessment and Respiratory Health Outcomes: A Systematic Review. Atmosphere. 2025; 16(10):1154. https://doi.org/10.3390/atmos16101154

Chicago/Turabian Style

Agbehadji, Israel Edem, and Ibidun Christiana Obagbuwa. 2025. "Explainable Artificial Intelligence and Machine Learning for Air Pollution Risk Assessment and Respiratory Health Outcomes: A Systematic Review" Atmosphere 16, no. 10: 1154. https://doi.org/10.3390/atmos16101154

APA Style

Agbehadji, I. E., & Obagbuwa, I. C. (2025). Explainable Artificial Intelligence and Machine Learning for Air Pollution Risk Assessment and Respiratory Health Outcomes: A Systematic Review. Atmosphere, 16(10), 1154. https://doi.org/10.3390/atmos16101154

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop