Database Reference
In-Depth Information
provide great improvements in health education efforts and behavioral
interventions.
Other attempts to model large-scale population health include Google
Flu Trends [116] to detect influenza outbreaks indirectly by tracking the
frequency of World Wide Web searches for terms related to influenza-
like illnesses. For geographic areas as small as states in the U.S., Google
researchers have demonstrated that such search frequencies correlate
strongly with estimated influenza incidence based on conventional surveil-
lance of cases detected in a Centers for Disease Control and Prevention
(CDC) network of sentinel laboratories and physicians. Similarly, the
Automated Epidemiologic Geotemporal Integrated Surveillance System
(AEGIS), developed by Children Hospital Boston, involves Internet-
based data collection, management, and analysis systems to produce
timely estimates of incidence. Almost 30,000 residents of Belgium, the
Netherlands, and Portugal voluntarily report on their influenza symp-
toms on a weekly basis at the Gripenet web sites [117].
Reality mining can also significantly impact epidemiologic investiga-
tions that capture impact of exposure to different types of environments
and pathogens on population health
7
. For instance, traditional inves-
tigations into links between individual exposures to airborne pollutants
(particulate matter, carbon monoxide, and nitric oxide) and health con-
ditions have relied on comparisons of aggregates of persons, or static
measures and snapshots of exposure. This has impacted the effectiveness
of such studies, and the associated costs. As opposed to these aggregate
or static approaches, reality mining can be used to capture dynamic
measures of time-activity patterns in relation to exposures. The cell
phone location data can be combined with existing air quality moni-
toring stations or inferred from vehicle trac patterns and locations of
industrial facilities to yield spatially precise measures of exposure suit-
able for studying large samples of individuals.
While the discussion on reality mining in this chapter has been dom-
inated by information captured from individual mobile phones, several
aspects of our cities are getting instrumented. This includes our trans-
portation infrastructures, security infrastructures, energy and utility sys-
tems, food production and distribution etc. Combining all of this infor-
mation at scale, overcoming the associated data ownership, privacy, and
connectivity challenges, and analyzing it can provide significant benefits
7
The Spatio-Temporal Epidemiological Modeler (STEM) [118] activity tool has recently
been proposed as an open source application designed to help scientists and public health
o
cials create and use models of emerging infectious diseases.
