Continuous Body Monitoring - Ambient Intelligence for Scientific Discovery - page 219

not in collaborative work. To successfully work as a team it is helpful to be more
aware of individual and team dynamics so participants can act progressively
toward goals.
Sandia Collaborative Working [5], used BodyMedia armbands while it “ex-
plored the potential for real-time signal analysis to provide information that
enables emergent and desirable group behavior and improved task performance”
by mapping audio, video, physiological, and somatic data of test subjects play-
ing a collaborative game. The group used an array of sensors, processors, and
software to develop a tool that can identify individual and group trends, identify
their states, and give warnings and help in decision-making [19]. This pointed
to individual and group stress and eciency trends useful for studying future
group work situations.
Nasoz et al [16] looked at the ability to predict emotions from physiological
information taken from SenseWear PRO 2 armbands while watching emotionally
scored movie clips. “The affective intelligent user interfaces we plan to create
will adapt to user affect dynamically in the current context, thus providing en-
hanced social presence.” The team predicts that machines that identify and
modify output based on users' emotions can optimize teaching, improve emo-
tional communication between remote users, and suggest modifications to unsafe
emotions in particular contexts.
A multi-university team used the BodyMedia armbands to detect whether a
user is busy or not busy in order to determine interruptibility. “Our motivation
is derived from the observation that context does not require a descriptive label
to be used for adaptivity and contextually sensitive response. This makes an
approach towards completely unsupervised learning feasible [13].” This investi-
gation was a preliminary step towards making devices subservient to their users
instead of the typical, opposite reality.
7 Application, Capability, and Acceptance
BodyMedia armbands and software are useful devices for comfortably collecting
body measurements, which can lead to better understanding and detection of
the user's physiological and emotional states, improved progress checking and
training, and data collection in situations requiring high mobility. When coupled
with environmental information or data from user groups, information can lead
to better design of physical environments.
Contextual aware devices can calculate situation appropriate responses that
may greatly change future context-specific human computer interaction and in-
formation feedback loops, leading to beneficial changes in education, health,
work and leisure.
The SenseWear continuous body monitor is presently mainly deployed in
health and fitness domains. Currently available are consumer products, profes-
sionally supervised products and scientific and social research products. The
data from SenseWear remains a largely untapped diamond mine. At the 2004
International Conference on Machine Learning eight workshop papers discussed