Comparing Airborne Particulate Matter Intake Dose Assessment Models Using Low-Cost Portable Sensor Data

Fine particulate matter (PM2.5) exposure assessment among active daily commuters to induce behaviour change to reduce air pollution

Fine particulate matter (PM2.5), a detrimental urban air pollutant primarily emitted by traffic and biomass burning, poses disproportionately significant health risks at relatively limited exposure during commuting. Previous studies have mainly focused on fixed locations when assessing PM2.5 exposure, while neglecting pedestrians and cyclists, who often experience higher pollution levels. In response, this research aimed to independently validate the effectiveness of bicycle-mounted low-cost sensors (LCS) adopted by citizens, evaluate temporal and spatial PM2.5 exposure, and assess associated health risks in Ljubljana, Slovenia. The LCS quality assurance results, verified by co-location field tests by air quality monitoring stations (AQMS), showed comparable outcomes with an average percentage difference of 21.29 %, attributed to humidity-induced nucleation effects. The colder months exhibited the highest air pollution levels (μ = 32.31 μg/m3) due to frequent thermal inversions and weak wind circulation, hindering vertical air mixing and the adequate dispersion of pollutants. Additionally, PM2.5 levels in all sampling periods were lowest in the afternoon (μ = 12.09 μg/m3) and highest during the night (μ = 61.00 μg/m3) when the planetary boundary layer thins, leading to the trapping of pollutants near the surface, thus significantly affecting diurnal and seasonal patterns. Analysis of exposure factors revealed that cyclists were approximately three times more exposed than pedestrians. However, the toxicological risk assessment indicated a minimal potential risk of PM2.5 exposure. The collaborative integration of data from official AQMS and LCS can enhance evidence-based policy-making processes and facilitates the realignment of effective regulatory frameworks to reduce urban air pollution.

Empowering Participatory Research in Urban Health: Wearable Biometric and Environmental Sensors for Activity Recognition

Participatory exposure research, which tracks behaviour and assesses exposure to stressors like air pollution, traditionally relies on time-activity diaries. This study introduces a novel approach, employing machine learning (ML) to empower laypersons in human activity recognition (HAR), aiming to reduce dependence on manual recording by leveraging data from wearable sensors. Recognising complex activities such as smoking and cooking presents unique challenges due to specific environmental conditions. In this research, we combined wearable environment/ambient and wrist-worn activity/biometric sensors for complex activity recognition in an urban stressor exposure study, measuring parameters like particulate matter concentrations, temperature, and humidity. Two groups, Group H (88 individuals) and Group M (18 individuals), wore the devices and manually logged their activities hourly and minutely, respectively. Prioritising accessibility and inclusivity, we selected three classification algorithms: k-nearest neighbours (IBk), decision trees (J48), and random forests (RF), based on: (1) proven efficacy in existing literature, (2) understandability and transparency for laypersons, (3) availability on user-friendly platforms like WEKA, and (4) efficiency on basic devices such as office laptops or smartphones. Accuracy improved with finer temporal resolution and detailed activity categories. However, when compared to other published human activity recognition research, our accuracy rates, particularly for less complex activities, were not as competitive. Misclassifications were higher for vague activities (resting, playing), while well-defined activities (smoking, cooking, running) had few errors. Including environmental sensor data increased accuracy for all activities, especially playing, smoking, and running. Future work should consider exploring other explainable algorithms available on diverse tools and platforms. Our findings underscore ML's potential in exposure studies, emphasising its adaptability and significance for laypersons while also highlighting areas for improvement.

Simulating the impact of particulate matter exposure on health-related behaviour: A comparative study of stochastic modelling and personal monitoring data

Epidemiological and exposure studies concerning particulate matter (PM) often rely on data from sparse governmental stations. While low-cost personal monitors have some drawbacks, recent developments have shown that they can provide fairly accurate and fit-for-purpose data. Comparing a stochastic, i.e., agent-based model (ABM), with environmental, biometric and activity data, collected with personal monitors, could provide insight into how the two approaches assess PM exposure and dose. An ABM was constructed, simulating a PM exposure/dose assessment of 100 agents. Their actions were governed by inherent probabilities of performing an activity, based on population data. Each activity was associated with an intensity level, and a PM pollution level. The ABM results were compared with real-world results. Both approaches had comparable results, showing similar trends and a mean dose. Discrepancies were seen in the activities with the highest mean dose values. A stochastic model, based on population data, does not capture well some specifics of a local population. Combined, personal sensors could provide input for calibration, and an ABM approach can help offset a low number of participants. Implementing a function of agents influencing others transport choice, increased the importance of cycling/walking in the overall dose estimate. Activists, agents with an increased transport influence, did not play an important role at low PM levels. As concentrations rose, higher shares of activists (and their influence) caused the dose to increase. Simulating a person's PM exposure/dose in different scenarios and activities in a virtual environment provides researchers and policymakers with a valuable tool.

Assessment of Individual-Level Exposure to Airborne Particulate Matter during Periods of Atmospheric Thermal Inversion

Air pollution exposure is harmful to human health and reducing it at the level of an individual requires measurements and assessments that capture the spatiotemporal variability of different microenvironments and the influence of specific activities. In this paper, activity-specific and general indoor and outdoor exposure during and after a period of high concentrations of particulate matter (PM), e.g., an atmospheric thermal inversion (ATI) in the Ljubljana subalpine basin, Slovenia, was assessed. To this end, personal particulate matter monitors (PPM) were used, worn by participants of the H2020 ICARUS sampling campaigns in spring 2019 who also recorded their hourly activities. ATI period(s) were determined based on data collected from two meteorological stations managed by the Slovenian Environmental Agency (SEA). Results showed that indoor and outdoor exposure to PM was significantly higher during the ATI period, and that the difference between mean indoor and outdoor exposure to PM was much higher during the ATI period (23.0 µg/m3) than after (6.5 µg/m3). Indoor activities generally were associated with smaller differences, with cooking and cleaning even having higher values in the post-ATI period. On the other hand, all outdoor activities had higher PM values during the ATI than after, with larger differences, mostly >30.0 µg/m3. Overall, this work demonstrated that an individual-level approach can provide better spatiotemporal resolution and evaluate the relative importance of specific high-exposure events, and in this way provide an ancillary tool for exposure assessments.

User-Centred Design of a Final Results Report for Participants in Multi-Sensor Personal Air Pollution Exposure Monitoring Campaigns

Using low-cost portable air quality (AQ) monitoring devices is a growing trend in personal exposure studies, enabling a higher spatio-temporal resolution and identifying acute exposure to high concentrations. Comprehension of the results by participants is not guaranteed in exposure studies. However, information on personal exposure is multiplex, which calls for participant involvement in information design to maximise communication output and comprehension. This study describes and proposes a model of a user-centred design (UCD) approach for preparing a final report for participants involved in a multi-sensor personal exposure monitoring study performed in seven cities within the EU Horizon 2020 ICARUS project. Using a combination of human-centred design (HCD), human-information interaction (HII) and design thinking approaches, we iteratively included participants in the framing and design of the final report. User needs were mapped using a survey (n = 82), and feedback on the draft report was obtained from a focus group (n = 5). User requirements were assessed and validated using a post-campaign survey (n = 31). The UCD research was conducted amongst participants in Ljubljana, Slovenia, and the results report was distributed among the participating cities across Europe. The feedback made it clear that the final report was well-received and helped participants better understand the influence of individual behaviours on personal exposure to air pollution.

Assessing the Current Integration of Multiple Personalised Wearable Sensors for Environment and Health Monitoring

The ever-growing development of sensor technology brings new opportunities to investigate impacts of the outdoor environment on human health at the individual level. However, there is limited literature on the use of multiple personalized sensors in urban environments. This review paper focuses on examining how multiple personalized sensors have been integrated to enhance the monitoring of co-exposures and health effects in the city. Following PRISMA guidelines, two reviewers screened 4898 studies from Scopus, Web of Science, ProQuest, Embase, and PubMed databases published from January 2010 to April 2021. In this case, 39 articles met the eligibility criteria. The review begins by examining the characteristics of the reviewed papers to assess the current situation of integrating multiple sensors for health and environment monitoring. Two main challenges were identified from the quality assessment: choosing sensors and integrating data. Lastly, we propose a checklist with feasible measures to improve the integration of multiple sensors for future studies.

Harmonization and Visualization of Data from a Transnational Multi-Sensor Personal Exposure Campaign

Use of a multi-sensor approach can provide citizens with holistic insights into the air quality of their immediate surroundings and their personal exposure to urban stressors. Our work, as part of the ICARUS H2020 project, which included over 600 participants from seven European cities, discusses the data fusion and harmonization of a diverse set of multi-sensor data streams to provide a comprehensive and understandable report for participants. Harmonizing the data streams identified issues with the sensor devices and protocols, such as non-uniform timestamps, data gaps, difficult data retrieval from commercial devices, and coarse activity data logging. Our process of data fusion and harmonization allowed us to automate visualizations and reports, and consequently provide each participant with a detailed individualized report. Results showed that a key solution was to streamline the code and speed up the process, which necessitated certain compromises in visualizing the data. A thought-out process of data fusion and harmonization of a diverse set of multi-sensor data streams considerably improved the quality and quantity of distilled data that a research participant received. Though automation considerably accelerated the production of the reports, manual and structured double checks are strongly recommended.