Instrumentation & Measurement Magazine 24-6 - 19

None of the articles surveyed tracked participants long
enough to observe individual trajectories of cognitive
change. The ability to do so is one of the purported benefits
of a home-based system, but it has not yet been achieved
in reality, perhaps due to unaddressed privacy concerns.
While it makes sense to first design and validate a prototype
system over short time spans, it is our hope to see more
long-term, longitudinal investigations of privacy-preserving
speech technology in the home. Such a study design
would also answer the question of repeatability, or test-retest
reliability. Related to this issue, many of the one-shot, crosssectional
research studies on speech analysis actually use
elicited speech protocols (e.g., describe a picture, or recall
a story). These protocols induce a highly-specific structure
and content in the narrative speech. In contrast, most smart
home studies measure conversational speech, which is more
natural and presumably more repeatable. However, it is not
clear whether the same amount or kind of information is
available in conversation, as compared to the more targeted
elicitation tasks.
Most studies consider speech as an independent data
stream [22]. Again, this is reasonable first step to assess the
advantages and disadvantages of speech as a measurement
instrument for cognition; however, the true value in a smart
home environment will likely lie in multimodal data streams
coming together to form a complete assessment of health and
well-being. For example, combining speech analysis with
other smart home measures of cognition [37] and physiological
health [38], in addition to measures of functional ability
such as activities of daily living [39] and gait analysis [40], will
almost certainly give a more accurate and robust measurement
of physical, mental, and cognitive health.
Related to this, most studies based on speech have looked
into a single health condition; a holistic approach that simultaneously
assesses multiple health conditions may help justify
privacy concerns and facilitates integration of such a system
within a smart home environment.
Finally, there is a pressing need for validation with more diverse
user populations. In additional to clinical diversity (i.e.,
different types of dementia and related pathologies, brain injuries,
comorbid mental health issues, etc.), it is important
that these technologies are validated with users who represent
different cultural and educational backgrounds, varieties
of English, and varied living situations. This will ensure the
validity and reproducibility of the instrument on a heterogeneous
user base.
Conclusion
Smart home technologies offer the potential to change the way
we live throughout our lives. Speech, as a natural and ubiquitous
human behavior, will undoubtedly be incorporated into
such environments-and indeed, already is, through our use
of voice commands and virtual assistants. As a measurement
instrument of cognitive status, speech presents many opportunities
but also many open questions to be answered in future
research.
September 2021
It will be crucial to bring together multi-disciplinary teams
to fully understand and serve the needs of older adults as they
face the challenges of cognitive decline. Nelson et al. [31] recommend
recruiting expertise from disciplines as varied as
" behavioral science, affective computing, mobile sensing,
mental health, signal processing, artificial intelligence, biomedical
engineering, data mining, computer networks,
machine learning, bioethics, and technology development,
along with industry partners. " In this article, we have emphasized
the need for the thoughtful application of the principles
of measurement to the assessment of such automated systems,
to protect and promote the health and well-being of our aging
population.
References
[1] M. Anderson and A. Perrin, " Technology use among seniors, "
Pew Research Center for Internet & Technology, 2017.
[2] B. E. Lyons, D. Austin, A. Seelye, J. Petersen, J. Yeargers, T. Riley,
N. Sharma, N. Mattek, K. Wild, H. Dodge et al., " Pervasive
computing technologies to continuously assess Alzheimer's
disease progression and intervention efficacy, " Frontiers in Aging
Neuroscience, vol. 7, p. 102, 2015.
[3] K. C. Fraser, J. A. Meltzer, and F. Rudzicz, " Linguistic features
identify Alzheimer's disease in narrative speech, " J. of Alzheimer's
Disease, vol. 49, no. 2, pp. 407-422, 2016.
[4] B. Mirheidari, D. Blackburn, M. Reuber, T. Walker, and H.
Christensen, " Diagnosing people with dementia using automatic
conversation analysis, " in Proc. of Interspeech (ISCA), pp.
1220-1224, 2016.
[5] B. Roark, M. Mitchell, J.-P. Hosom, K. Hollingshead, and J. Kaye,
" Spoken language derived measures for detecting mild cognitive
impairment, " IEEE Trans. Audio, Speech, and Language Process., vol.
19, no. 7, pp. 2081-2090, 2011.
[6] D. Beltrami, G. Gagliardi, R. Rossini Favretti, E. Ghidoni, F.
Tamburini, and L. Calza', " Speech analysis by natural language
processing techniques: a possible tool for very early detection of
cognitive decline? " Frontiers in Aging Neuroscience, vol. 10, p. 369,
2018.
[7] G. Gosztolya, V. Vincze, L. Tóth, M. Pákáski, J. Kálmán, and I.
Hoffmann, " Identifying mild cognitive impairment and mild
Alzheimer's disease based on spontaneous speech using ASR
and linguistic features, " Computer Speech Language, vol. 53, pp.
181-197, 2019.
[8] M. Komeili, C. Pou-Prom, D. Liaqat, K. C. Fraser, M. Yancheva,
and F. Rudzicz, " Talk2me: automated linguistic data collection for
personal assessment, " Plos one, vol. 14, no. 3, p. e0212342, 2019.
[9] M. Parvis and A. Vallan, " Medical measurements and
uncertainties, " IEEE Instrum. Meas. Mag., vol. 5, no. 2, pp. 12-17,
2002.
[10] M. Crocco, M. Cristani, A. Trucco, and V. Murino, " Audio
surveillance: a systematic review, " ACM Computing Surveys, vol.
48, no. 4, Feb. 2016.
[11] A. Ferrero and S. Salicone, " Measurement uncertainty, " IEEE
Instrum. Meas. Mag., vol. 9, no. 3, pp. 44-51, 2006.
[12] V. Berisha, S. Wang, A. LaCross, J. Liss, and P. Garcia-Filion,
" Longitudinal changes in linguistic complexity among
IEEE Instrumentation & Measurement Magazine
19
Instrumentation & Measurement Magazine 24-6

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