Instrumentation & Measurement Magazine 24-6 - 85

Fig. 3. Human behavior and relevant processing.
and monitor movement,
activity and ADLs [61].
Parallel to sensor development,
machine learning,
neural networks and deep
learning techniques have
become sophisticated.
These algorithms allow
for the extraction of useful
information from large
volumes of complex sensor
data, fusion of many
sensor data streams and
mining for patterns in behavior
otherwise unknown
[77], [81], [82]. Each area
of study has matured to
a stage at which ambient
sensing of ADLs for the
continuous assessment
of cognitive and physical
health at home is possible.
This could support aging
An episodic cognition assessment framework, has been
proposed for this purpose [76], [80]. The framework consists
of three major blocks; sensor sub-systems, fusion and data analytics
and reporting and alerts [76]. This framework focuses
on the extraction of episodic data from subjects over time, the
fusion and data analytics to distill the clinically meaningful information
before providing a meaningful call to action via an
alert [76]. A synonymous model, in the context of long-term
monitoring of movement, activity and human behavior is
illustrated in Fig. 3. Movement detection and identification requires
the use of sensors, signal processing, feature extraction
and classification. The measurement of activity requires capturing
a series of movements as action and multiple actions as
ADLs, requiring sensor systems of multiple sensor types and
often in multiple environments using machine learning methods
to condense and extract valuable information. Monitoring
behavior involves long term monitoring of ADLs to establish
regular and abnormal behaviors. This requires sensor fusion
techniques to amalgamate information from extensive sensor
systems, network solutions to account for increased amounts
of data and reporting to provide necessary alerts for calls to
action.
This type of system therefore would passively gather
long-term information on an individual, characterize the individual's
normal activities and could subsequently identify
a meaningful change in individual behavior. This would allow
each system to be highly individualized, accounting for
the variable nature of disease presentation and may provide
the opportunity to begin early treatment for a condition such
as dementia.
In ambient sensing, there are countless sensor types capable
of exceptionally fine measurements at extremely high
frequencies. These have been shown to accurately identify
September 2021
individuals with independent living, remaining in contact
with family, caregivers and medical staff and consequently
maintaining a high quality of life. The possible impacts are significant;
hospitals currently experience high volumes of older
adult patients. If these rates were reduced, there would be
more time and money available to help these patients in more
meaningful ways. To realize these possibilities, efforts must be
focused on new challenges, primarily the real-time data analysis
requirements of pervasive computing coupled with the
large volumes of data and the resulting need for processing infrastructure
to be close to the data source.
References
[1] A. McNeill, P. Briggs, J. Pywell, and L. Coventry, " Functional
privacy concerns of older adults about pervasive healthmonitoring
systems, " in Proc. 10th Int. Conf. Pervasive
Technologies Related to Assistive Environments, pp. 96-102, Jun.
2017.
[2] N. Thompson and W. Livingston, " Promoting well-being, " Illn.
Crisis Loss, vol. 26, no. 2, pp. 98-110, Apr. 2018.
[3] J. Olszewska, " IEEE Recommended Practice for Assessing the
Impact of Autonomous and Intelligent Systems on Human WellBeing:
IEEE Standard 7010-2020, " 2020.
[4] P. Moorhouse and K. Rockwood, " Frailty and its quantitative
clinical evaluation, " J. R. Coll. Physicians Edinb., vol. 42, no. 4, pp.
333-340, 2012.
[5] A. H. Snijders, B. P. van de Warrenburg, N. Giladi, and B. R.
Bloem, " Neurological gait disorders in elderly people: clinical
approach and classification, " Lancet Neurol., vol. 6, no. 1, pp.
63-74, Jan. 2007.
[6] A. Clegg, J. Young, S. Iliffe, M. O. Rikkert, and K. Rockwood,
" Frailty in elderly people, " The Lancet, vol. 381, no. 9868, pp.
752-762, Mar. 2013.
IEEE Instrumentation & Measurement Magazine
85
Instrumentation & Measurement Magazine 24-6

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