SPLENDID II; Development of a system for monitoring eating and physical activity behavior

The overall goal of SPLENDID, an EU-funded project, is to develop a system of
sensors that can be used for the prevention of obesity and eating disorders.
Initially it will be used for mapping behavioral patterns regarding eating and
physical activity of the individuals that are wearing the system. Consecutively
it will provide them real-time feedback on how to improve these patterns. The
system will act like a personal coach.

This system includes, among others, sensors that will map the eating- and
physical activity behavior of its wearers. These sensors and/or the
corresponding algorithms are still under development and include:
- an eating detection sensor (i.e. a sensor, that is worn as an hearing aid,
that registers if and when someone is eating),
- the Mandometer (i.e. a scale, that is placed underneath the plate while
eating, that registers the intake of food throughout the meal)
- en een accelerometer (i.e. a sensor, that is worn at the hip, that registers
acceleration of the body)

The corresponding algorithms are step-by-step procedures for calculations that
enable automatic detection of behaviors (e.g. if someone is eating or the level
of physical activity a person is engaging in) using the singals produced by the
sensors.

The main objective of the current study is to see how well the current versions
of the sensors (mainly the eating detection sensor) function/ how useful the
produced signals are. The sensors have already shown to produce valuable
signals in controlled settings and we would like to see if in a semi-controlled
setting the produced signals are still informative.

Furthermore, the produced signals will be used to 'feed' the algorithms. If
more/different signals are fed to the algorithms they will learn to better
understand the signals and will become better at detecting behaviors from those
signals.

Additionally we want to see how wearing/using the sensors is experienced by the
participants, mainly with regards to wearer-comfort and user-friendliness.

The study is a semi-controlled experiment. Participants will participate in two
testing days at the university. While wearing/using the sensors of the
SPLENDID-system they will perform predetermined *daily life-activities*; i.e.
have lunch, spend some free-time, have snack(s), do moderate exercise and have
diner. However, the participants will have quite some freedom in choosing the
'content' of these activities. They will be able to compose their own lunch,
choose what snacks to consume and choose what (physical) activities to perform.

At predetermined moments participants are expected eat a food product of choice
or perform an (physical) activity of choice (see 'study design').

The participants of 'SPLENDID II' will visit the university 3 times; once for
an information meeting (±60 min) and twice for a testing day (±7 hours). During
the information meeting the participants will fill out some questionnaires and
their weight and height will be measured. Furthermore, in the two hours prior
to the testing days the participants are requested not to eat or drink anything
(except for water). During the actual testing days the participants will fill
out some short questionnaires, but the main focus will be on performing *daily
life-activities* which will be monitored non-invasively with the SPLENDID
sensors. Examples of such activities are having lunch, snacking, relaxing,
taking the stairs and cycling on a hometrainer. This is not expected to provide
too much burden for the participants.

Furthermore, participation in the study is not expected to put an individual at
risk. Previous experiences with the separate SPLENDID-sensors have not shown
any adverse effects. Additionally, the sensors are non-invasive, which reduces
the potential damage they can cause. Finally, there are always risks associated
with the consumption of foods and moderate excercise, but also this is expected
to be limited.