Instrumentation & Measurement Magazine 24-6 - 84

Table 2 - Summary of sensor types and applications in measuring activity
Measured ADL
House-based ADLs
Specific ADLs
* Cooking
* Cleaning
* Person Recognition
* Falls
Driving
* Travel/Destination patterns / complexity
* Route planning
* Specific driving actions
Night-time Wandering
* (re)Direction
* (re)Orientation
* Lapping
* Random
* Pacing
Games
* Performance of game challenge
* Frequency of play
movement and activity examined in this section. Where the
first two sections of this paper discussed the sensor technologies
used to detect movement and action, this section discusses
the information gleaned from longitudinal measurements and
details a general system for efficient, comprehensive and cohesive
longitudinal monitoring.
In examining the role of ambient sensing of human behavior
for cognitive and physical well-being, an exemplary case study
illustrating the need for ambient setting is that of an individual
with dementia. Dementia is characterized by multiple cognitive
and behavioral impairments, which affects an individual's
ability to carry out ADLs. This condition progresses over time
at individually variable rates and accelerations [56]. By the
time an individual with dementia sees a doctor, cognitive decline
has significantly progressed [76]. Despite noted cognitive
decline, diagnosis is challenging and requires a combination
of assessments including
a history, a clinical assessment,
a neurological and
physical exam, blood work
and medical imaging [76].
Considering the individual
nature of cognitive decline,
and the ability of ambient
sensors to detect and
analyze movement and
action, it seems an intuitive
conclusion to derive
that ambient sensing is a
valuable tool in identifying
behavioral changes related
to cognitive decline on an
individual level.
84
Ambient Sensor Types
* Force sensors
* Infrared cameras
* Microphones
* WiFi
* Accelerometers
* GPS
* Engine computer sensors
* Vehicle speed sensors
* Pressure sensors
* Motion sensors
* Intervention technologies
* Fine motor movements
including mouse
movement, typing and
deletions.
Relevant
References
[57]-[63]
[64]-[68]
[69]-[71]
[72]-[74]
Much effort has been exerted to improve the sensitivity
of measurements, enabling the extraction of more information
from a given subject, movement or action. While there are
many sensitive, accurate and automated components of the
clinical assessment for dementia, there is a need for systems or
frameworks that pull the components together in a clinically
meaningful manner. This is not a simple task; among other
challenges, the number and type of sensors in a given system
is often numerous, the data can therefore be innumerable and
require processing simply to maintain compatibility and the
progression of cognitive and physical well-being over time
is generally non-linear [77]. A useful technological solution
therefore requires sophisticated algorithms including machine
learning techniques to extract patterns in behavior indicative
of an individual's cognitive and physical well-being from an
environment equipped with ambient sensors [77]-[79].
Fig. 2. Mobility related cognition measurements over time.
IEEE Instrumentation & Measurement Magazine
September 2021
Instrumentation & Measurement Magazine 24-6

Table of Contents for the Digital Edition of Instrumentation & Measurement Magazine 24-6
