Database Reference
In-Depth Information
4.3 Reality Mining
Reality mining [72] has recently been identified as one of 10 emerging
technologies that could change the world, as it allows us to build com-
prehensive pictures of our lives, with the potential of transforming our
understanding of ourselves, our organizations, and our society. Reality
mining pulls together digital trace data that we generate as part of our
daily activities with data mining and machine learning techniques to en-
able new non-intrusive applications in diagnosis, patient and treatment
monitoring, health services use, surveillance of disease and risk factors,
and public health investigation and disease control.
One of the key sensors employed by reality mining techniques is the
mobile phone - that has become a central part of our lives. Mobile
phones currently capture a lot of contextual information about users,
including location (communication between the device and towers or
GPS sensors) as well as data about their social connections (call and
duration information). In addition, newer smart phones, e.g. the iPhone,
include special sensors such as the microphone or the accelerometers
that allow the capture of important diagnostic and health related data.
These devices now also have the processing power of low-end desktop
computers, allowing the deployment of several local analytics in support
of healthcare applications.
Reality mining of these behavior signals may be correlated to the func-
tion of some major brain systems. It has been shown that arousal of the
autonomic nervous system produces changes in activity levels. Hence,
recent pilot projects have shown that it may be possible to diagnose de-
pression from the way a person talks - depressed people tend to speak
more slowly, a change that speech analysis software on a phone might
recognize more readily than friends or family do [111]. Similarly, mon-
itoring a phones motion sensors can also reveal small changes in gait,
which could be an early indicator of ailments such as Parkinsons disease.
The phone sensors may be used to measure time-coupling between
speech and movement of people, to capture indications of attention and
screen for language development problems. The sensors can potentially
capture the unconscious mimicry between people (e.g., reciprocated head
nods, posture changes, etc.) as reliable predictors of trust and empathy,
and improve compliance [112]. Similarly, the sensors can also be used
to measure consistency or fluidity of movement or speech production
to capture cognitive load. These different types of measurements of
brain function have been shown to be predictive measures of human
behavior [113], and play an important role in human social interactions
