Indoor air最新文献

{"title":"Rapid In-Air Inactivation of Influenza A Virus Using an In-Duct Cold Atmospheric Plasma Panel","authors":"Luutzen F. A. Wymenga,&nbsp;Suzanne M. A. Mijnhardt,&nbsp;Mohamad Ghaffarian Niasar,&nbsp;Willem D. van Driel,&nbsp;GuoQi Zhang,&nbsp;Pieter L. A. Fraaij,&nbsp;Ron A. M. Fouchier,&nbsp;Sander Herfst","doi":"10.1155/ina/7268405","DOIUrl":"https://doi.org/10.1155/ina/7268405","url":null,"abstract":"<p>Respiratory virus infections pose a major public health burden and pandemic risk, especially when they are transmissible via the air. Consequently, there is an urgent need for effective control measures to mitigate airborne transmission. Cold atmospheric plasma (CAP) offers a promising solution for in-air virus inactivation by exposing viruses to a mixture of electrons, ions, photons, and reactive oxygen and nitrogen species (RONS). Here, a grating dielectric barrier discharge (DBD) panel consisting of 88 glass-covered electrodes was developed. Plasma formation using a high-voltage neon transformer was confirmed across all electrodes, with a power deposition of approximately 5 W. In an open acrylic tube, representative of an air duct, ozone levels increased to a maximum of 90 ppm and returned to zero within 30 s after the DBD panel was turned off. The obstructing area of the electrodes was 13%, resulting in a lower pressure drop than the current standard HEPA filters. The DBD panel was further tested in an aerosol chamber where the viability of artificially aerosolized influenza A virus (IAV) after a single passage through the DBD panel was assessed. Plaque assays showed that virus inactivation increased with exposure time, reaching 96.9% inactivation after 15 s of exposure and up to 99.9% inactivation when aerosolization followed prior RONS accumulation, whereas RNA levels and air temperature remained largely unaffected. These results demonstrate that in-duct DBD panels can effectively inactivate IAV in aerosols without leaving persistent toxic residues. This proof-of-concept study might give future directions for implementation of CAP technology in high-risk settings such as hospitals, livestock facilities, and transportation hubs to reduce the spread of infectious virus in the environment.</p>","PeriodicalId":13529,"journal":{"name":"Indoor air","volume":"2026 1","pages":""},"PeriodicalIF":2.7,"publicationDate":"2026-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1155/ina/7268405","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149205992","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Spatial Variation′s Impact on Pathogen Concentration and Occupant Presence for Exposure Risk Estimation","authors":"Tsz Wun Tsang,&nbsp;Kwok Wai Mui,&nbsp;Kwun Hei Cheung,&nbsp;Ling Tim Wong","doi":"10.1155/ina/8887595","DOIUrl":"https://doi.org/10.1155/ina/8887595","url":null,"abstract":"<p>Recent hospital ward designs focus on fewer beds, more spacing, and the use of curtains or partitions for better privacy and improved infection control. Traditional infection risk models assume well-mixed air and consider only total exposure time. Such models may not be suitable for these new layouts. This study presents a targeted framework for assessing airborne pathogen exposure. It uses spatial and temporal pathogen concentration data to offer precise exposure risk estimates. The approach integrates literature-based computational fluid dynamics (CFD) datasets for different ward layouts and ventilation scenarios. It also uses a distance-based probabilistic model of nurse presence to quantify the limitations of previous assumptions. The framework identifies mismatches between exposure concentrations and actual risk based on likely nurse locations in a ward. It shows that spaces with high pathogen concentrations may have a low risk if nurse presence is unlikely. In contrast, low-concentration corridors may still pose a risk due to frequent occupancy. Compared with traditional models, the proposed framework provides risk estimates that are up to 11.2% lower or 6.1% higher. On average, estimations are 3.1% lower, leading to clearer and more actionable infection control strategies and management practices.</p>","PeriodicalId":13529,"journal":{"name":"Indoor air","volume":"2026 1","pages":""},"PeriodicalIF":2.7,"publicationDate":"2026-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1155/ina/8887595","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149205598","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Analysis of Factors Associated With Household Dust Loading Measurements","authors":"Cristina Fayad-Martinez,&nbsp;Annabella Cordovez,&nbsp;Olusola Olabisi Ogunseye,&nbsp;Maribeth Gidley,&nbsp;Foluke Adelabu,&nbsp;Jenna K. Honan,&nbsp;Jacqueline L. Larson,&nbsp;Paloma I. Beamer,&nbsp;Helena Solo-Gabriele,&nbsp;Alesia Ferguson","doi":"10.1155/ina/8054412","DOIUrl":"https://doi.org/10.1155/ina/8054412","url":null,"abstract":"<p>Dust loading is a critical parameter in exposure assessments. Current literature documenting indoor dust loadings, however, is limited, with most studies reporting results from vacuum samples that measure surface dust plus dust that can be extracted through suction. Wipes, on the other hand, have the ability to sample multiple high-contact surfaces and measure dust from the immediate surface, which is more applicable to exposures through contact scenarios. To address the need to expand datasets on dust loading measurements, the objectives of this study were to: (1) measure and report indoor dust loadings through wipe and vacuum samples, (2) use a regression model to evaluate factors that predict dust loading measurements by integrating survey responses and field data, and (3) compare loadings between sample types (wipe and vacuum). Dust loading samples were taken from nine different places (eight wipes and one vacuum sample) within 101 households across three states in the United States (North Carolina, Florida, and Arizona). Overall, for wipe samples, dust loading ranged from &lt; 0.2 to 2,492 <i>μ</i>g/cm², with a median of 8.01 <i>μ</i>g/cm². For the vacuum samples, dust loadings were statistically higher (<i>p</i> &lt; 0.001), ranging from 1.15 to 2,654 <i>μ</i>g/cm² with a median of 87.3 <i>μ</i>g/cm². For wipe loadings, the primary significant factor was surface elevation, with surface material (e.g., tile and cloth) or texture of surface (e.g., hard soft) as additional significant factors. The age of the home accounted for a smaller portion of the explained variance. Similarly, vacuum loadings were associated primarily with surface material or texture and secondarily with the age of the home. The data reported in this study increase the repository of reported dust levels on surfaces within homes and can be used as data inputs for exposure models.</p>","PeriodicalId":13529,"journal":{"name":"Indoor air","volume":"2026 1","pages":""},"PeriodicalIF":2.7,"publicationDate":"2026-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1155/ina/8054412","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149205600","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Impact of Urban Land Use on Indoor Air Quality: Quantifying Ambient Particle Infiltration and Asthma Risk in Naturally Ventilated School Buildings","authors":"Murnira Othman,&nbsp;Mohd Talib Latif,&nbsp;Elia Syarafina Abdul Shakur,&nbsp;Shih-Chun Candice Lung","doi":"10.1155/ina/5050009","DOIUrl":"https://doi.org/10.1155/ina/5050009","url":null,"abstract":"<p>Children spend a substantial portion of their daytime in classrooms, yet the contribution of outdoor pollution to indoor particulate exposure in naturally ventilated schools remains insufficiently characterised in rapidly urbanising regions of Southeast Asia. This study presents a comprehensive assessment of indoor–outdoor PM_2.5 and PM₁ dynamics and associated personal exposure across 10 primary schools in the urban Klang Valley, Malaysia, representing high-traffic; industrial; residential; mixed industrial and high-traffic; and mixed industrial and port. Significant spatial variability was observed between school location categories (<i>p</i> &lt; 0.05). The highest 24 h average indoor PM_2.5 and PM₁ concentrations were recorded in mixed industrial and port schools (35.33 ± 9.25 and 23.79 ± 5.87 <i>μ</i>g/m³, respectively), while the highest 24 h average outdoor concentrations were observed at high-traffic schools (34.50 ± 20.64 and 23.78 ± 13.09 <i>μ</i>g/m³, respectively). The average infiltration factor (<i>F</i>_inf &gt; 0.6) indicates substantial outdoor contributions to indoor particle levels in naturally ventilated classrooms. Personal monitoring further showed that classroom occupancy accounted for approximately 53% of school-hour exposure. Translating PM_2.5 concentrations into relative risk estimates suggested a 10%–20% increase in asthma risk during school hours across location categories. These findings highlight the importance of outdoor pollution infiltration and activity-related indoor sources in shaping schoolchildren’s exposure to particulate matter in naturally ventilated school environments.</p>","PeriodicalId":13529,"journal":{"name":"Indoor air","volume":"2026 1","pages":""},"PeriodicalIF":2.7,"publicationDate":"2026-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1155/ina/5050009","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149145905","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Indoor Air Quality and Exposure Characterization in Operational Waterpipe Cafés: Field Measurements of Particulate and Volatile Pollutants in Al-Madinah Al-Munawwarah","authors":"Seham N. Aljabri,&nbsp;Samah Ali,&nbsp;Amr Mohamed,&nbsp;Osama Albarbarawi,&nbsp;Sultan K. Alharbi,&nbsp;Moayad M. Khashoqji,&nbsp;Eman A. Assirey","doi":"10.1155/ina/8494753","DOIUrl":"https://doi.org/10.1155/ina/8494753","url":null,"abstract":"<p>Waterpipe cafés represent combustion-dominated indoor environments where tobacco heating and charcoal burning generate complex mixtures of airborne pollutants. Despite the widespread presence of such venues in Saudi Arabia and globally, quantitative field measurements describing real-world exposure conditions remain limited, particularly in the Gulf region. This study characterizes indoor air quality in five operational waterpipe cafés in Al-Madinah Al-Munawwarah through direct measurements of particulate matter (PM₂.₅ and PM₁₀), total volatile organic compounds (TVOCs), and formaldehyde (HCHO) during peak occupancy periods on weekdays and weekends. Substantial spatial and temporal variability in pollutant concentrations was observed across venues, largely reflecting differences in smoking intensity and ventilation effectiveness. Mean weekday PM₂.₅ and PM₁₀ concentrations reached 285.5 ± 149 <i>μ</i>g/m³ and 301.6 ± 156 <i>μ</i>g/m³, respectively, while weekend TVOC concentrations averaged 0.50 mg/m³. In multiple instances, measured particulate and volatile pollutant levels exceeded international health-based guideline values, indicating elevated exposure potential under high-use conditions. Formaldehyde concentrations remained below short-term reference thresholds. The findings identify waterpipe cafés as significant indoor sources of fine particulate matter and volatile pollutants. This study provides region-specific evidence supporting the need for stricter indoor air quality regulations in waterpipe café environments. The results contribute to advancing national priorities in public health protection, environmental sustainability, and urban environmental management within Saudi Arabia.</p>","PeriodicalId":13529,"journal":{"name":"Indoor air","volume":"2026 1","pages":""},"PeriodicalIF":2.7,"publicationDate":"2026-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1155/ina/8494753","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149145652","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Integrated Assessment of Indoor Air Pollution and Associated Health Risks in a Hemodialysis Unit","authors":"Diana Rodríguez,&nbsp;Rafael Quiles-Zafra,&nbsp;Marta Cuerva,&nbsp;Elena Jiménez","doi":"10.1155/ina/4283416","DOIUrl":"https://doi.org/10.1155/ina/4283416","url":null,"abstract":"<p>Indoor air quality (IAQ) is critical for protecting occupant health, particularly in healthcare settings such as hemodialysis (HD) units, where patients spend prolonged periods indoors. This study assessed IAQ in an HD treatment room over 1 year, identified pollutant sources using indoor/outdoor (I/O) ratios, and evaluated associated health risks for patients and healthcare staff. IAQ was assessed using approximately 1400 time-resolved measurements of CO₂, O₃, and particulate matter (PM_0.3, PM_2.5, and PM₁₀), together with 12 monthly sampling campaigns for volatile organic compounds (VOCs). Simultaneous outdoor measurements were used to calculate I/O ratios, whereas indoor temperature (<i>T</i>) and relative humidity (<i>RH</i>) were monitored to assess thermal comfort. <i>T</i> (23.2°C–25.5°C) and <i>RH</i> (36.5%–63.9%) generally remained within acceptable comfort ranges. Indoor CO₂ concentrations frequently exceeded the threshold established by Spanish regulations, with indoor-outdoor differences above 350 ppm and peak values exceeding 1100 ppm, indicating insufficient air renewal. Indoor O₃, PM_2.5, and PM₁₀ generally presented I/O ratios below 1, suggesting a predominantly outdoor origin, whereas PM_0.3 concentrations remained low. Although average O₃ and PM concentrations were generally below World Health Organization guideline values, occasional concentration peaks were recorded. Following patient disconnection and cleaning activities, indoor O₃ concentrations decreased, likely due to reactions with VOCs emitted by cleaning products. Most VOCs exhibited I/O ratios above 1, indicating substantial indoor sources. Health risk assessment showed that carcinogenic risks associated with benzene and ethylbenzene were below the benchmark value of 10⁻⁶ for both patients and healthcare staff. Hazard indices for both groups were below 1, indicating no significant noncarcinogenic risk. These findings highlight ventilation deficiencies and substantial indoor VOC emissions, emphasizing the need to improve mechanical ventilation performance and implement continuous IAQ monitoring in HD units.</p>","PeriodicalId":13529,"journal":{"name":"Indoor air","volume":"2026 1","pages":""},"PeriodicalIF":2.7,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1155/ina/4283416","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149145666","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evaluation of Two Cyclonic Bioaerosol Samplers for Hospital Air Monitoring in a Calibrated Test Chamber Integrating Aerodynamics and Biological Endpoints","authors":"Jessica Morel,&nbsp;Jean-Michel Lenoir,&nbsp;Camille Ménard,&nbsp;Samuel Le Hir,&nbsp;Clara Bourgeay,&nbsp;Tiago Zambujo,&nbsp;Emma Ensenat,&nbsp;Ibtissam Barhoumi,&nbsp;Mariana Gaboriau,&nbsp;Thomas Julien,&nbsp;Anaïs Proust,&nbsp;Manuel Rosa-Calatrava","doi":"10.1155/ina/2773156","DOIUrl":"https://doi.org/10.1155/ina/2773156","url":null,"abstract":"<p>Indoor air quality (IAQ) monitoring in hospital settings is critical to limit airborne transmission of pathogens to vulnerable patients and healthcare workers. However, bioaerosol sampling remains challenging because of the diversity of devices and methodologies and the very low airborne concentration of pathogens in real-world settings, especially when infectivity-based measurements, are required for assessing transmission risk. In the absence of harmonized evaluation procedures, it remains difficult to select sampling strategies that are both reliable and fit for purpose. In this study, we developed a controlled and reproducible experimental framework for the evaluation of bioaerosol samplers under conditions relevant to hospital air monitoring. Two widely used cyclonic bioaerosol samplers, Coriolis+ and Coriolis Compact, were assessed in a calibrated 20 m³ test chamber generating stable atmospheres containing infectious SARS-CoV-2, <i>Staphylococcus aureus</i>, or <i>Aspergillus fumigatus</i> at different concentration levels. Samplers′ performance was evaluated using complementary molecular and viability-based endpoints and was combined with computational fluid dynamics (CFD) and experimental airflow measurements to characterize airflow dynamics. Our results demonstrated that samplers′ performance strongly depended on the biological endpoint and microorganism considered. For SARS-CoV-2, the two devices showed comparable recovery of infectious virus, at intermediate and high concentrations, but neither detected infectious particles at the lowest level. In contrast, Coriolis+ achieved near-complete recovery of viral genomes, whereas Coriolis Compact recovered only a fraction of the expected signal. Regarding <i>S. aureus</i> and <i>A. fumigatus</i>, Coriolis+ achieved superior recovery of viable microorganisms, whereas Coriolis Compact generally yielded higher genomic recovery at moderate and high concentrations. CFD analyses further showed that local airflow conditions and sampler-specific flow patterns, particularly for the higher flow Coriolis+, are critical for interpreting biological recovery and for defining appropriate placement and operating modes. Overall, this study demonstrates that bioaerosol sampler evaluation must jointly consider molecular, microbiological, and aerodynamic dimensions. The proposed framework offers a basis for more standardized and reproducible performance assessment of bioaerosol samplers and supports evidence-based selection of air monitoring strategies in healthcare environments.</p>","PeriodicalId":13529,"journal":{"name":"Indoor air","volume":"2026 1","pages":""},"PeriodicalIF":2.7,"publicationDate":"2026-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1155/ina/2773156","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149147017","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Desk-Level CO2 Sensing for Monitoring Occupancy and Indoor Environmental Quality","authors":"Christina J. Higgins,&nbsp;Tamás Mezei,&nbsp;Marco Placidi,&nbsp;Susan J. Hughes,&nbsp;Matteo Carpentieri","doi":"10.1155/ina/2251842","DOIUrl":"https://doi.org/10.1155/ina/2251842","url":null,"abstract":"<p>Occupants of buildings with poor indoor environmental quality (IEQ) suffer from a lack of comfort, poor work performance, health effects and effects on wellbeing. IEQ can vary within a space, and building occupants both influence and are influenced by this, by generating carbon dioxide (CO₂), heat and moisture. Monitoring both IEQ and occupancy enhances understanding of the indoor conditions and provides information on space utilisation. In this study, a novel method of utilising low-cost, distributed CO₂ sensing for monitoring both occupancy and IEQ was designed and validated. Spatiotemporal variation in IEQ was captured via the distributed system, and ground truth occupancy data were obtained via a thermal imaging camera and computer vision–based person detection model. Linking person detection and IEQ data, statistical analysis of the influence of occupant presence on IEQ was conducted and found that while office-average CO₂, temperature and relative humidity were within proposed comfort criteria, desk-level data demonstrated strong spatial heterogeneity, suggesting the presence of microenvironments. CO₂ and temperature differed significantly between desks, and occupancy had a significant effect on office CO₂ and temperature, with relative humidity comparatively uniform and uninfluenced. Rate of change and roughness metrics, derived from the CO₂ data and defined as the mean absolute first difference and the standard deviation of the first difference of the CO₂ time signal sampled at 1-min intervals, respectively, demonstrated moderate correlations with occupancy (<i>r</i> = 0.38–0.40). Threshold-based classification at frequently occupied desks achieved good discriminative performance for both dynamic features, with the area under the curve (AUC) = 0.85–0.86, compared with moderate performance across all desks (AUC = 0.75). These results demonstrate that desk-level CO₂ sensing represents a viable low-cost alternative to dedicated occupancy sensors at regularly used workstations, with simple feature-based classifiers achieving good discriminative performance.</p>","PeriodicalId":13529,"journal":{"name":"Indoor air","volume":"2026 1","pages":""},"PeriodicalIF":2.7,"publicationDate":"2026-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1155/ina/2251842","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149147018","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Indirect Machine Learning Prediction of a Composite Indoor Air Quality Index in an Airport Passenger Transit Area","authors":"Cosmina-Mihaela Rosca,&nbsp;Adrian Stancu","doi":"10.1155/ina/3119205","DOIUrl":"https://doi.org/10.1155/ina/3119205","url":null,"abstract":"<p>Indoor air quality (IAQ) is identified as a health protection measure, especially in heavily trafficked public spaces, where pollutant variations are influenced by human activities, building characteristics, ventilation, infiltration, and external atmospheric conditions. The study proposes a methodology for estimating IAQ through an indicator called the Proposed Indoor Air Quality Index (PIAQI) in a passenger transit area at Brindisi Airport, Italy. It is calculated based on indoor concentrations of CO₂, PM_2.5, and PM₁₀. The main objective is to evaluate the possibility of predicting PIAQI without using the variables directly involved in its formula as predictors. For this purpose, a dataset was constructed consisting of indoor measurements collected within an airport, historical weather data, and outdoor air pollution data provided through public platforms. The records were imported using the C# language into a SQL Server database and temporally aligned through an SQL view. The analytical dataset combined high-frequency indoor sensor measurements, historical meteorological data, and outdoor air-pollution data through hourly temporal alignment. Subsequently, a feature importance analysis was performed on a reduced set of observations using the LightGBM regression model implemented in ML.NET through the drop column importance method. The variables with a positive contribution to the <i>R</i>² coefficient loss were retained for the subsequent stage. A sample of 1 million observations was extracted from the SQL View and used in Azure Machine Learning to train seven regression algorithms. They were calculated, and from the comparison, it was found that the XGBoost regressor model had the best performance. The study showed that PIAQI estimates the combination of alternative indoor aerosol measurements with meteorological variables and outdoor pollutant concentrations, even after removing indoor CO₂, PM_2.5, and PM₁₀ from the predictor set. For internal validation, the data were randomly divided into 70% for training and 30% for holdout evaluation. XGBoost regressor achieved the best internal performance, with <i>R</i>² = 0.9675, MAE = 2.32, and RMSE = 4.04. A complementary chronological evaluation, in which the first 70% of the observations were used for training and the final 30% for testing, produced <i>R</i>² = −1.3905, MAE = 33.50, and RMSE = 41.90. This performance decrease indicates that random splitting yields optimistic estimates when temporally adjacent observations occur in both subsets. The methodology serves as a foundation for the development of intelligent monitoring, warning, ventilation control, and filtration control systems in crowded public spaces, as well as for taking appropriate preventive measures.</p>","PeriodicalId":13529,"journal":{"name":"Indoor air","volume":"2026 1","pages":""},"PeriodicalIF":2.7,"publicationDate":"2026-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1155/ina/3119205","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149146903","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Indoor Air Quality Forecasting With Machine Learning: A Systematic Review","authors":"Lewis Hall Rudford,&nbsp;Sara Sharifzadeh,&nbsp;Nicholas Micallef,&nbsp;Paul Lewis,&nbsp;Philip Webb,&nbsp;Daniele Doneddu","doi":"10.1155/ina/8524332","DOIUrl":"https://doi.org/10.1155/ina/8524332","url":null,"abstract":"<p>Indoor air pollution poses substantial health risks. People spend over 90% of their time indoors, making the indoor environment a major contributor to overall pollution exposure. Therefore, indoor air quality (IAQ) prediction is critical for proactive mitigation strategies. Despite advances in machine learning (ML), research on IAQ forecasting remains fragmented, with limited integration of occupant behaviour, inconsistent evaluation practices and persistent challenges in predicting high pollution events. This systematic review synthesises evidence from 99 peer-reviewed studies published between 2010 and 2026, examining data sources, modelling approaches, forecasting horizons and performance metrics used for IAQ prediction. The analysis reveals rapid growth in publications after 2020, with deep learning time-series architectures such as long short-term memory (LSTM) dominating because of their ability to capture temporal dependencies, whereas transformer-based models show emerging potential for longer-term forecasting. Only 13% of reviewed studies incorporate behavioural or occupancy-related variables, despite their demonstrated influence on indoor pollution and model accuracy. Model evaluation primarily relies on conventional metrics such as RMSE and R2, with limited assessment of health-relevant thresholds or forecast horizon-specific performance. Geographic disparities are also evident, with developing regions such as Africa and South America under-represented. The review identifies critical research gaps and provides actionable recommendations for integrating behavioural data, adopting context-aware evaluation metrics and addressing data imbalance to improve the robustness and applicability of IAQ forecasting models in real-world buildings.</p>","PeriodicalId":13529,"journal":{"name":"Indoor air","volume":"2026 1","pages":""},"PeriodicalIF":2.7,"publicationDate":"2026-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1155/ina/8524332","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149146904","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}