2.2 Clinical Focus
PULSE project focuses on the link between air pollution and the respiratory disease of
Asthma, and between physical inactivity and the metabolic disease of Type 2 Diabetes.
The risk assessment for this two pathologies comprises the evaluation respectively of:
– for type 2 diabetes: behavioural risks associated (i.e. reduced exercise/physical
activity at home or in public places). This is associated with higher risk of T2D
onset in a dose-response relationship. The assessment use unobtrusive sensing/data
collection and volunteered data to collect baseline measures of health and wellbeing, and tracking and model mobility at home and across the city (including time,
frequency and route of mode of transit and/or movement).
– for asthma: Environmental/exposure risks (i.e. exposure to air pollution, especially
with regard to near roadway air pollution). Poor air quality is associated with higher
risk of Asthma onset and exacerbation.
Risks of diseases onset are evaluated thorough risk assessment models, that in
PULSE are biometric simulation models that predict the risk of the onset of the ashtma
and diabetes in relationship to air quality.
The models has been developed by chosen ones from a literature review of the
prediction models of type 2 diabetes (T2D) onset and asthma adult-onse. Some of them
were selected to be implemented and recalibrated on the datasets available on PULSE
repository and adding new variables [12].
2.3 Data Architecture
PULSE architecture is composed by 5 main structures [15]: PULSEAir, App Server,
AIR Quality distributed sensor system, GisDB, WebGIS and Personal DB.
– PULSE App: is the personal App provided to the participants in charge of collecting
sensors data and interacting with the users to propose interventions and gamification. PulsAIR is available both for iOS and Android and can be connected to FitBit,
Garmin and Asus health tracker devices.
– AIR Quality distributed sensor system: the PULSE air quality sensor’s system is
composed of multiple type of sensors and sensor’s datasets: it combines mobile
sensors and mobile network of sensors in order monitor the variable trends in
emission within urban areas with an high resolution and to appropriately address the
temporal and spatial scales where usually pollutants are spread. Two types of
sensors has been used across pilots that are the AQ10x of DunavNet (20+, deployed
in all pilots) and PurpleAir PA-II sensor.
– App Server: This structure internally connects PULSE components.
– Personal DB: This repository contains personal detail, connectivity, activity logs,
pilot sites structured data, etc.…
– GISDB: This repository is in charge of collecting non-personal sources of data:
satellite, open repositories and fixed sensors.
– WebGIS: This data engine is in charge of aggregating and exploiting all the collected data to build front-end visualizations through maps.
426
D. Vito et al.
PULSE project focuses on the link between air pollution and the respiratory disease of
Asthma, and between physical inactivity and the metabolic disease of Type 2 Diabetes.
The risk assessment for this two pathologies comprises the evaluation respectively of:
– for type 2 diabetes: behavioural risks associated (i.e. reduced exercise/physical
activity at home or in public places). This is associated with higher risk of T2D
onset in a dose-response relationship. The assessment use unobtrusive sensing/data
collection and volunteered data to collect baseline measures of health and wellbeing, and tracking and model mobility at home and across the city (including time,
frequency and route of mode of transit and/or movement).
– for asthma: Environmental/exposure risks (i.e. exposure to air pollution, especially
with regard to near roadway air pollution). Poor air quality is associated with higher
risk of Asthma onset and exacerbation.
Risks of diseases onset are evaluated thorough risk assessment models, that in
PULSE are biometric simulation models that predict the risk of the onset of the ashtma
and diabetes in relationship to air quality.
The models has been developed by chosen ones from a literature review of the
prediction models of type 2 diabetes (T2D) onset and asthma adult-onse. Some of them
were selected to be implemented and recalibrated on the datasets available on PULSE
repository and adding new variables [12].
2.3 Data Architecture
PULSE architecture is composed by 5 main structures [15]: PULSEAir, App Server,
AIR Quality distributed sensor system, GisDB, WebGIS and Personal DB.
– PULSE App: is the personal App provided to the participants in charge of collecting
sensors data and interacting with the users to propose interventions and gamification. PulsAIR is available both for iOS and Android and can be connected to FitBit,
Garmin and Asus health tracker devices.
– AIR Quality distributed sensor system: the PULSE air quality sensor’s system is
composed of multiple type of sensors and sensor’s datasets: it combines mobile
sensors and mobile network of sensors in order monitor the variable trends in
emission within urban areas with an high resolution and to appropriately address the
temporal and spatial scales where usually pollutants are spread. Two types of
sensors has been used across pilots that are the AQ10x of DunavNet (20+, deployed
in all pilots) and PurpleAir PA-II sensor.
– App Server: This structure internally connects PULSE components.
– Personal DB: This repository contains personal detail, connectivity, activity logs,
pilot sites structured data, etc.…
– GISDB: This repository is in charge of collecting non-personal sources of data:
satellite, open repositories and fixed sensors.
– WebGIS: This data engine is in charge of aggregating and exploiting all the collected data to build front-end visualizations through maps.
426
D. Vito et al.