Search

International audience

International audience; Poster PresentationsP-20 | Omics approaches in toxicology

Technology innovations create possibilities to capture exposure-related data at a great depth and breadth. Considering, though, the substantial hurdles involved in collecting individual data for whole populations, this study introduces a first approach of simulating human movement and interaction behaviour, using Agent Based Modelling (ABM). A city scale ABM was developed for urban Thessaloniki, Greece that feeds into population-based exposure assessment without imposing prior bias, basing its estimations onto emerging properties of the behaviour of the computerised autonomous decision makers (agents) that compose the city-system. Population statistics, road and buildings networks data were transformed into human, road and building agents, respectively. Survey outputs with time-use patterns were associated with human agent rules, aiming to model representative to real-world behaviours. Moreover, time-geography of exposure data, derived from a local sensors campaign, was used to inform and enhance the model. As a prevalence of an agent-specific decision-making, virtual individuals of different sociodemographic backgrounds express different spatiotemporal behaviours and their trajectories are coupled with spatially resolved pollution levels. Personal exposure was evaluated by assigning PM concentrations to human agents based on coordinates, type of location and intensity of encountered activities. Study results indicated that PM2.5 inhalation adjusted exposure between housemates can differ by 56.5% whereas exposure between two neighbours can vary by as much as 87%, due to the prevalence of different behaviours. This study provides details of a new methodology that permits the cost-effective construction of refined timeactivity diaries and daily exposure profiles, taking into account different microenvironments and sociodemographic characteristics. The proposed method leads to a refined exposure assessment model, addressing effectively vulnerable subgroups of population. It can be used for evaluating the probable impacts of different public health policies prior to implementation reducing, therefore, the time and expense required to identify efficient measures.

A major advantage of human biomonitoring (HBM) data is that they provide an integrated overview of the body burden to xenobiotics that an individual is exposed to. However, quantification of exposure based on HBM data poses significant challenges that are worth facing, given the opportunities that HBM provides in terms of informing and effectively supporting risk assessment. Based on the above, the aim of this within the HBM4EU project was to derive EU-wide external exposure estimates starting from HBM data and to derive and risk characterization ratio (RCR) by comparing these estimates with existing regulatory thresholds. For the reconstruction of exposure the INTEGRA computational platform was properly parameterised for the compounds of interest, namely bisphenol-A (BPA), phthalates (DEHP, DiNP and DnBP) and DINCH, emerging flame retardants (TCEP) and Perfluorinated compounds (PFOA and PFOS). The results indicated that for the majority of the examined compounds, daily intake levels are below the existing regulatory thresholds. For BPA, mean daily intake is almost 2 orders of magnitude below the respective threshold proposed by EFSA. For phthalates, daily intake estimates are usually one or two orders of magnitude below the respective TDI, with the exception of BBzP, for which intake estimates of the upper part of the exposure distribution is close to the threshold of 10 μg/kg_bw/d. Regarding TCEP, which is a typical emerging flame retardant, the mean daily intake estimate is below 0.1 μg/kg_bw/d, which is far below the calculated ‘provisional’ TDI of 13 μg/kg_bw/d, however, at the moment very few HBM data were available and these exposure levels are rather indicative than representative of the European countries. Finally, regarding the estimated intakes of PFCs, intake levels of PFOS are very close to TDI (0.15 μg/ kg_bw/d proposed by the CONTAM Panel), while the calculated levels for PFOA are one order of magnitude below the respective TDI of 1.5 μg/kg_bw/d.

Nowadays, the advancement of mobile technology in conjunction with the introduction of the concept of exposome has provided new dynamics to the exposure studies. Since the addressing of health outcomes related to environmental stressors is crucial, the improvement of exposure assessment methodology is of paramount importance. Towards this aim, a pilot study was carried out in the two major cities of Greece (Athens, Thessaloniki), investigating the applicability of commercially available fitness monitors and the Moves App for tracking people’s location and activities, as well as for predicting the type of the encountered location, using advanced modeling techniques. Within the frame of the study, 21 individuals were using the Fitbit Flex activity tracker, a temperature logger, and the application Moves App on their smartphones. For the validation of the above equipment, participants were also carrying an Actigraph (activity sensor) and a GPS device. The data collected from Fitbit Flex, the temperature logger, and the GPS (speed) were used as input parameters in an Artificial Neural Network (ANN) model for predicting the type of location. Analysis of the data showed that the Moves App tends to underestimate the daily steps counts in comparison with Fitbit Flex and Actigraph, respectively, while Moves App predicted the movement trajectory of an individual with reasonable accuracy, compared to a dedicated GPS. Finally, the encountered location was successfully predicted by the ANN in most of the cases. © 2018, Springer International Publishing AG, part of Springer Nature.

In the present work, two types of artificial neural network (NN) models using the multilayer perceptron (MLP) and the radial basis function (RBF) techniques, as well as a model based on principal component regression analysis (PCRA), are employed to forecast hourly PM(10) concentrations in four urban areas (Larnaca, Limassol, Nicosia and Paphos) in Cyprus. The model development is based on a variety of meteorological and pollutant parameters corresponding to the 2-year period between July 2006 and June 2008, and the model evaluation is achieved through the use of a series of well-established evaluation instruments and methodologies. The evaluation reveals that the MLP NN models display the best forecasting performance with R (2) values ranging between 0.65 and 0.76, whereas the RBF NNs and the PCRA models reveal a rather weak performance with R (2) values between 0.37-0.43 and 0.33-0.38, respectively. The derived MLP models are also used to forecast Saharan dust episodes with remarkable success (probability of detection ranging between 0.68 and 0.71). On the whole, the analysis shows that the models introduced here could provide local authorities with reliable and precise predictions and alarms about air quality if used on an operational basis. Environmental Science and Pollution Research

The objective of the present study is the exploitation of active sampling personal exposure data in assessing the factors that affect exposure to benzene in combination with the widely accepted scheme of passive sampling-time microenvironment-activity diaries (TMAD). The campaign included personal exposure measurements with both passive and active sampling in several microenvironments, evaluation of TMAD kept by the volunteers, and a variety of environmental data (ambient air benzene determination, traffic and meteorological observations). Due to the relatively elevated benzene traffic emissions, average personal exposure was determined to be equal to 8.9 mu g/m(3), ranging between 5 and 20 mu g/m(3), which is a value highly related to the average urban concentration (9.2 mu g/m(3)). The information gained from TMAD was embedded (in terms of spatial and temporal distribution) into three zones respectively, in order to draw statistically significant conclusions about the exposure levels and the activity patterns. The contribution of the activities to the overall amount of exposure was further quantified and refined by active sampling measurements. These data revealed that driving in a traffic-congested road was the main activity leading to elevated exposure levels (up to 70 mu g/m(3)), followed by walking on the roadside of a congested road (up to 35 mu g/m(3)). Indoor exposure to benzene was in general lower than outdoor (indicating that traffic is the dominant source of benzene emissions in the wider area), and it was significantly affected by the presence of environmental tobacco smoke. The higher significance of the regression coefficients obtained by statistical analysis of the active sampling data was fundamental for the development of a regression-based prediction exposure model. The model was evaluated through comparison with the passive sampling data, which were considered as an unknown but realistic data exposure pattern. The model performed very well in terms of expressing the variance of the exposure data with an average score of R (2) equal to 0.935. All of the above indicate that active sampling is a necessary albeit more laborious tool that needs to be used as a complement to passive sampling for precise quantification of the factors determining personal exposure patterns. Environmental Modeling & Assessment

The objective of the current study was the development of a reliable modeling platform to calculate in real time the personal exposure and the associated health risk for filling station employees evaluating current environmental parameters (traffic, meteorological and amount of fuel traded) determined by the appropriate sensor network. A set of Artificial Neural Networks (ANNs) was developed to predict benzene exposure pattern for the filling station employees. Furthermore, a Physiology Based Pharmaco-Kinetic (PBPK) risk assessment model was developed in order to calculate the lifetime probability distribution of leukemia to the employees, fed by data obtained by the ANN model. Bayesian algorithm was involved in crucial points of both model sub compartments. The application was evaluated in two filling stations (one urban and one rural). Among several algorithms available for the development of the ANN exposure model, Bayesian regularization provided the best results and seemed to be a promising technique for prediction of the exposure pattern of that occupational population group. On assessing the estimated leukemia risk under the scope of providing a distribution curve based on the exposure levels and the different susceptibility of the population, the Bayesian algorithm was a prerequisite of the Monte Carlo approach, which is integrated in the PBPK-based risk model. In conclusion, the modeling system described herein is capable of exploiting the information collected by the environmental sensors in order to estimate in real time the personal exposure and the resulting health risk for employees of gasoline filling stations. Sensors

In the present study, the results of a measurement campaign aiming to assess cancer risk among two special groups of population: policemen and laboratory technicians exposed to the toxic substances, benzene and formaldehyde are presented. The exposure is compared to general population risk. The results show that policemen working outdoor (traffic regulation, patrol on foot or in vehicles, etc.) are exposed at a significantly higher benzene concentration (3-5 times) than the general population, while the exposure to carbonyls is in general lower. The laboratory technicians appear to be highly exposed to formaldehyde while no significant variation of benzene exposure in comparison to the general population is recorded. The assessment revealed that laboratory technicians and policemen run a 20% and 1% higher cancer risk respectively compared to the general population. Indoor working place air quality is more significant in assessing cancer risk in these two categories of professionals, due to the higher Inhalation Unit Risk (IUR) of formaldehyde compared to benzene. Since the origin of the danger to laboratory technicians is clear (use of chemicals necessary for the experiments), in policemen the presence of carbonyls in indoor air concentrations due to smoking or used materials constitute a danger equal to the exposure to traffic originated air pollutants. Environ Monit Assess

Epidemiology

In the present study, the exposure to benzene of employees working in two filling stations (one urban and one rural) was estimated, through the method of passive sampling. Additional data (30' measurements of benzene exposure through active sampling to employees dealing with different activities, meteorological and traffic data) were collected. The measurements campaign was performed in both summer and wintertime to determine the seasonal variation of the exposure pattern. In addition, a set of artificial neural networks (ANNs) was developed to predict benzene exposure pattern for the filling station employees based on active sampling data and the parameters related to the employees' exposure. The quantification of the contribution of each parameter to the overall exposure pattern was also attempted. The results showed that although vapour recovery technologies are installed in the refuelling systems and benzene emissions are significantly reduced compared to the past, filling station employees are still highly exposed to benzene (52-15 mu g m(-3)). Benzene exposure is strongly correlated to car refuelling (exposure levels up to 85 mu g m(-3)), while activities like car washing or working in cash machine inside an office contribute to lower exposure levels (up to 44 and 24 mu g m(-3) respectively). In rural filling station, exposure levels were in general lower compared to the urban ones, due to the smaller amount of gasoline that was traded and the absence of any significant traffic effect or urban background concentration. The developed ANN seemed to be a promising technique in the prediction of the exposure pattern giving very good results, and the quantification of the parameters affirmed the importance of the refueling procedure to the exposure levels. (C) 2007 Elsevier Ltd. All rights reserved. Atmospheric Environment

This work examines the contribution of petrol stations to the ambient benzene concentrations and attempts to estimate the possible health risks for the people living in the vicinity of such installations. Three monitoring sites (urban, suburban and rural) were used as reference points and the benzene concentrations were recorded at several distances along their perimeter. In order to evaluate the net contribution of the petrol station to the ambient benzene concentrations, the urban background concentration, measured by passive samplers and the contribution of the roads, estimated with both the COPERT and the linear source model CALINE 4, were deduced. Validation and optimization of the modeling system COPERT and CALINE4 was done in advance to ensure the reliability of the results. It seems that petrol stations have a significant contribution to ambient benzene concentrations in their vicinity. Finally, a risk assessment evaluation was attempted in terms of increased cancer risk due to the presence of the petrol stations in an area. The results show remarkable increase of the population risks in the vicinity, ranging from 3% to 21% in comparison to the population in the rest of the town. (c) 2006 Elsevier Ltd. All rights reserved. Atmospheric Environment

The work deals with the comparison of two models: (i) an artificial neural network (ANN) and (ii) a semi empirical deterministic model (DET), used to simulate benzene concentrations in a street canyon. Furthermore, the response of models to 'what if scenarios' was also examined. The ANN was based on a training procedure using measurements collected in a specific street canyon (benzene concentrations, traffic density, vehicle's type distribution). The DET model was based on road traffic emission rate, wind speed and direction, and the geometrical characteristics of the road. Although both model, produced very good results, given the limited amount of data available, the ANN succeeded slightly better than DET in predicting benzene concentrations. On the other hand, the ANN is less able to reproduce the effect of significant changes in traffic flow patterns on benzene concentrations. The results from the simulations indicate that the ANN is a promising technique for benzene modeling in an urban environment and in can be used for environmental management purposes. (C) 2005 Elsevier B.V. All rights reserved. Ecological Modelling

Nowadays, the prediction of atmospheric pollutant concentrations in street canyons' environment is of great importance. To achieve this, many kinds of modeling techniques were proposed. One of the most promising techniques is Artificial Neural Networks (ANNs). In this study, an ANN was developed to predict benzene concentrations in a heavily trafficted street canyon. It also evaluates the importance of the variables determining these concentrations. The training procedure was developed based on data collected by an annual measurement's campaign, performed in a specific street canyon. The data include benzene concentration, traffic flow and speed, vehicle's type distribution, wind speed and direction. The results from the simulations indicate that ANN is a promising technique for predicting benzene in an urban environment, and can be used for environmental management purposes. Fresenius Environmental Bulletin

In this work an approach is presented that describes the changes of benzene concentration in the air, in relation with the special traffic characteristics of a road. The dominant input parameters of the model are traffic density and the vehicle's type distribution according to seven main categories characterized by different emission factors. The dispersion approach used is a semi-empirical relationship that apart from emission rates requires also wind speed and the direction, as well as the geometrical characteristics of the road. The methodology was validated for Ioannina, a Greek medium sized town with special traffic and geographic characteristics presenting high atmospheric pollution values. It is found that the benzene concentrations estimated by the methodology are in a very good agreement with the measurements. (c) 2005 Elsevier B.V. All rights reserved. Science of the Total Environment

The BTX levels are significantly high compared to the EU directive for benzene in European cities with population around or higher one million. Since there are hundreds of towns in Europe with smaller population, it is important to know the levels of aromatics in these areas. This work presents the results of a benzene, toluene and xylene (BTX) measurement campaign that took place in Ioannina, a medium-sized Greek city. As a result of traffic situation and the local meteorological conditions, pollution levels in Ioannina are unusually high, at least for a city of that size. BTX levels were measured using passive samplers placed at several points around the city, as well as across a selected street canyon using both passive and active samplers, combined with simultaneous measurements of traffic flow and wind speed. The measurement procedure was repeated in an exact manner for all four seasons and the results suggest that benzene levels, at all sampling points, exceed the limit set by EU Directive 2000/69. Benzene levels appear correlated to traffic density, while benzene/toluene/xylene ratios present a seasonal variation linked to meteorological conditions. (c) 2005 Elsevier Ltd. All rights reserved. Atmospheric Environment

A semi-empirical approach by applying existing techniques to model benzene concentrations in a street canyon is presented here. The dominant input parameters of the model are traffic density and the type of vehicle distribution, which are necessary to calculate with preciseness the road's emission rate. The dispersion aspect of the model is a semi-empirical relationship based on the road emission rate, wind speed and direction, and also the geometrical characteristics of the road. The model produces very good results (RMSE 10.45 and 10.11 mug m(-3); RRMSE 1.56 and 1.54, for spring and autumn, respectively) given its simplicity and the limited amount of data available with which to optimize the model. (C) 2004 Elsevier Ltd. All rights reserved. Atmospheric Environment