A Preliminary Review of Behavioural Biometrics for Health
Monitoring in the Elderly
Jordi Solé-Casals, Mihaela Vancea and Jaume Miquel March
Data and Signal Processing Group, University of Vic-Central University of Catalonia
Sagrada Família 7,08500 Vic, Catalonia, Spain
Keywords: Ageing, Elderly, Health, Behavioural Biometrics, Wearable.
Abstract: This article explores the potential of ICT-based biometrics for monitoring the health status of the elderly
people. It departs from specific ageing and biometric traits to then focus on behavioural biometric traits like
handwriting, speech and gait to finally explore their practical application in health monitoring of elderly.
1 INTRODUCTION
The word biometrics comes from the Greek words
“bios” (life) and “metrikos” (measure). Strictly
speaking, biometrics refers to a science and
technology involving the measuring and statistical
analysis of biological characteristics.
According with Faundez-Zanuy and Chetouani
(2005), biometric recognition offers a promising
approach for security applications, with some
advantages over the classical methods that depend
on something you own (key, card, etc.) or something
you know (password, PIN, etc.). An interesting
feature of biometric traits is that they are based on
something you are or something you do, so you do
not need to remember anything or to hold any token.
But security is not the only field in which biometrics
can be used. For instance, biometric techniques can
be used for the analysis of data in agricultural field
experiments to compare, for example, the yields of
different varieties of wheat; or for the analysis of
biometric characteristics for animal classification.
Among others, one of the most promising fields of
biometrics application is the analysis of data from
human clinical trials to evaluate the possible illness
or malfunction of human body.
Population ageing has become one of the major
challenges for the future of our societies and various
technological innovations have been developed to
promote and enhance an active and healthy ageing.
In Europe, in the last three decades, birth and death
rates have gradually decreased while longevity and
life expectancy rates have significantly increased.
By 2050, the European population segment of over
50 will increase by 35 percent and the one over 85
will triple (Eurostat, 2012). Beside a steady decrease
in fertility and birth rates, strongly linked to the
progressive inclusion of women in the labour
market, scientific developments in different fields
such as medicine, healthcare or hygiene also account
for this process of population ageing.
The use of biometric systems for health
diagnosis/monitoring/screening/ could mark an
important step in dealing with population ageing due
to aspects related to:
The prevalence of diseases, especially those
related with brain deterioration; and
Particular characteristics of biometrics traits.
This article aims to provide an exploratory
review of ICT-based biometrics and how this
technology can be used to improve the quality of life
of elderly people.
This paper is organized as it follows: in section
2, we present various ageing and biometric traits; in
section 3, we focus on behavioural biometrics and
provide three different examples of biometrics traits
for health monitoring; finally, in section 4, we
present the final conclusions and discussion aspects.
2 AGEING AND BIOMETRIC
TRAITS
In order to use biometrics, we first have to decide
which characteristics can be used for biometric
recognition, and which ones can be used for health
365
Solé-Casals J., Vancea M. and Miquel March J..
A Preliminary Review of Behavioural Biometrics for Health Monitoring in the Elderly.
DOI: 10.5220/0005321603650371
In Proceedings of the International Conference on Bio-inspired Systems and Signal Processing (MPBS-2015), pages 365-371
ISBN: 978-989-758-069-7
Copyright
c
2015 SCITEPRESS (Science and Technology Publications, Lda.)
applications. As common sense says, a good
biometric trait must accomplish a set of properties.
Among these properties, we could mention the
following (Clarke, 2014):
Universality of coverage: every relevant person
should have an identifier;
Uniqueness: each relevant person should have
only one identifier, so two persons cannot have
the same identifier;
Permanence: the identifier should not change,
nor be changeable;
Indispensability: the identifier should be one or
more natural characteristics, which each person
has and retains; if artificial, the identifier should
be enforcedly available at all times;
Collectability: the identifier should be collectible
by anyone on any occasion;
Storability: the identifier should be storable in
manual and in automated systems;
Exclusivity: no other form of identification
should be necessary or used;
Precision: every identifier should be sufficiently
different from every other identifier that mistakes
are unlikely;
Simplicity: recording and transmission should be
easy and not error-prone;
Cost: measuring and storing the identifier should
not be unduly costly;
Convenience: measuring and storing the
identifier should not be unduly inconvenient or
time-consuming;
Acceptability: its use should conform to
contemporary social standards.
If we focus on health applications, some of these
characteristics may be very relevant while others
not. In health applications, the focal point is not so
much the subject (like, for example, in security
applications) than the state of health of the subject.
Since health applications are generally designed to
enhance health conditions, the biometric traits used
in this case should accomplish properties related to
the state of health of the individual and not with the
individual per se. In this new scenario, the
exclusivity property, for example, may not be
necessary stricto sensu, but the permanence property
may be really essential.
The aging process may produce changes in
biometric traits, and these changes are of particular
interest as they can tell us a lot of things about the
state of health of an individual. A good example is
the way in which our cognitive functions are related
to the ageing process. Cognitive decline is a natural
part of the ageing process. However, the extent of
decline varies across subjects and body functions.
For instance, handwriting and speech production are
a fine motor control performed by our brain. When
these signals are declining, health problems might be
detected.
Some physical and mental characteristics change
during the lifetime of a human (Sasse and Krol,
2013). In terms of height, physical maturity is
reached around age 20, but different body tissues
mature at different rates. Body height declines from
the age 50 onwards due to bone shrinkage, and the
acuity of sight and vision can start to decline even
earlier. Age-related changes can affect biometric
traits, and make it more difficult to operate the
systems through which users interact with them.
Ageing affects the biometric templates of all
biometric traits in different ways.
Biometric traits can be split into two main
categories:
Physiological Biometrics: based on direct
measurements of a part of the human body;
fingerprint, face, iris, and hand-scan recognition
belong to this group.
Behavioural Biometrics: based on measurements
and data derived from an action performed by the
user, and thus indirectly measuring some
characteristics of the human body; signature,
gait, gesture, and key stroking recognition belong
to this group.
For health applications for elderly, we are
interested in the second category of biometric traits,
behavioural biometrics, in order to characterize the
state of health of the subject. Of course, we have to
keep in mind that, for instance, the speech signal
depends on behavioural traits such as semantics,
diction, pronunciation, idiosyncrasy, etc. (related to
socio-economic status, education, place of birth,
etc.), but it also depends on the speaker’s
physiology, such as the shape of the vocal tract.
Hence, what is really useful (in some cases) in
behavioural biometrics is the evolution of biometric
traits for each subject (intrasubject variability), and
not the global characterization of the state of health
of each subject (intersubject variability).
3 BEHAVIOURAL BIOMETRICS
In this section, we review three of the most
important behavioural biometrics traits for health
monitoring in the elderly, and emphasise some
practical applications.
BIOSIGNALS2015-InternationalConferenceonBio-inspiredSystemsandSignalProcessing
366
3.1 Handwriting
Handwriting refers to a person's writing created with
a writing utensil such as a pen or pencil. This
creation may include not only text but also drawings
or other kind of graphs. Handwriting has been used
extensively for biometric recognition, for example,
as a signature: it is frequently used when signing
credit card receipts, checks, etc.
In the past, the analysis of handwriting had to be
performed in an offline manner. Only the writing
itself (strokes on a paper) were available for
analysis. Nowadays, modern capturing devices, such
as digitizing tablets and pens (with or without ink)
can gather data without losing its temporal
dimension. When spatio-temporal information is
available, its analysis is referred to as online.
Modern digitizing tablets not only gather the x-y
coordinates that describe the movement of the
writing device as it changes its position, but it can
also collect other data, mainly the pressure exerted
by the writing device on the writing surface and also
the azimuth, the angle of the pen in the horizontal
plane, and the altitude, the angle of the pen with
respect to the vertical axis. A very interesting aspect
of the modern online analysis of handwriting is that
it can take into account information collected when
the writing device was not exerting pressure on the
writing surface. Thus, the movements performed by
the hand while writing a text can be split into two
classes:
a) On-surface trajectories (pen-downs),
corresponding to the movements executed while
the writing device is touching the writing
surface; each of these trajectories produces a
visible stroke; and
b) In-air trajectories (pen-ups), corresponding to
the movements performed by the hand while
transitioning from one stroke to the next; during
these movements, the writing device exerts no
pressure on the surface.
Handwriting signals have been used for cognitive
impairment detection. For instance, handwriting skill
degradation and Alzheimer’s disease (AD) appear to
be significantly correlated (Forbes, Shanks and
Venneri, 2004) and some handwriting aspects can be
good indicators for its diagnosis (Neils-Strunjas et
al., 2006) or help differentiate between mild
Alzheimer’s disease and mild cognitive impairment
(Werner et al., 2006). As demonstrated in Faundez-
Zanuy et al. (2014), the visual inspection of the pen
down images suggest a progressive degree of
impairment, where drawing becomes more
disorganized and the three dimensions effect of the
drawing (a house) is only achieved in the mild case.
The visual information provided by the pen up
drawing between AD individuals also indicates a
progressive impairment and disorganization when
the individuals try to plan the drawing. In addition,
the pressure is also different. AD people produce
softer and simpler strokes.
For Parkinson’s disease (PD), handwriting is
thought to be impaired mainly due to hypokinesia
(decreased amplitude of movements) and
bradykinesia, as detailed in Broderick et al. (2009)
and Tucha et al. (2006). As compared to on-surface
movements (Drotár et al., 2014), the in-air
movements elicited during handwriting of a sentence
may involve additional cognitive processes such as
motor planning, programming of the alternating
motor sequences, and movement initiation that may
also have impacted on the kinematic features and
our results. In-air movement possess significant
amount of information relevant to diagnosis of PD
and could be incorporated in decision support
systems that are the important part of the next
generation health-care (Drotár et al., 2014).
Also, the analysis of handwriting has proven
useful to assess the effects of substances such
alcohol (Asıcıoglu et al., 2003; Phillips et al., 2009),
marijuana (Foley and Lamar Miller, 1979) or
caffeine (Tucha et al., 2006). Aided by modern
acquisition devices, the field of psychology has also
benefitted from the analysis of handwriting. For
instance, Rosenblum et al. (2003) link the
proficiency of the writers to the length of the in-air
trajectories of their handwritings.
3.2 Speech
Speech processing is the study of speech signals and
the processing methods of these signals. It is also
closely related to natural language processing
(NLP), and includes several areas of study as for
example Speech recognition, Speaker recognition,
Speech synthesis, Speech enhancement or Voice
analysis for medical purposes, among others. This
last area is the one in which we are interested in our
work, in which biometric techniques can be applied
for different diseases.
Alzheimer’s disease is the most common type of
dementia among the elderly. The cognitive deficits
and behavioural symptoms are severe enough to
limit the ability of an individual to perform everyday
professional, social or family activities. An early and
accurate diagnosis of AD helps patients and their
families to plan for the future and offers the best
opportunity to treat the symptoms of the disease.
APreliminaryReviewofBehaviouralBiometricsforHealthMonitoringintheElderly
367
Various researchers have explored different
solutions to help on the early diagnose of AD using
continuous speech signal. Non-invasive intelligent
diagnosis techniques would be very valuable for this
purpose. Lopez-de-Ipiña et al. (2013) analysed non-
invasive methods based on continuous speech and
used them to design an automatic system that can
offer medical doctors another point of view for the
early diagnosis of AD. By analysing continuous
speech, the system calculates an index whose values
show whether the subject can be classified as being
affected by AD. Emotional Temperature (ET) is
another parameter that, combined with other
traditional speech parameters, can improve and
facilitate the early diagnosis of AD (López-de-Ipiña
et al., 2013). In López-de-Ipiña et al. (2014), the
Fractal Dimension (FD) of the speech signals is
combined with linear parameters in the feature
vector in order to enhance the performance of the
original system while controlling the computational
cost. The advantage of using spontaneous speech
tests for the early diagnosis of AD is that these are
not perceived as stressful. Moreover, the cost of
speech analysis techniques is lower, as they do not
require extensive infrastructure or the availability of
medical equipment.
Another interesting example of speech
processing used in health applications is the
detection of Obstructive Sleep Apnea (OSA) (Solé-
Casals et al., 2014). OSA is a common sleep
disorder that manifests itself by daytime sleepiness
caused by a cease in breathing occurring repeatedly
during sleep, often for a minute or longer and as
many as hundreds of times during a single night.
Diagnosis of the sleep condition is based on the
calculation of the apnea–hypopnea index (AHI)
which measures the frequency of reductions in
airflow associated with upper-airway collapse or
narrowing that occurs with the state change from
wakefulness to sleep (Caples and Gami, 2005). The
gold standard procedure to determine the AHI is
polysomnography, however it is a quite costly
methodology (Kushida et al., 2005). No other
measure has proven to be superior to AHI in
assessing the overall effect of obstructive sleep
apnea. The results presented in Solé-Casals et al.,
(2014), in terms of Correct Classifications Rate,
Sensitivity and Specificity, all above 80% for
several classifiers, point out the good potential of
voice as a discriminating factor between healthy
subjects and severe OSA.
3.3 Gait
Human gait is the pattern of movement of the
extremities during locomotion. Minor variations in
gait style can be used as a biometric identifier to
identify individual people. For example, it’s well
known that stride parameters (stride length and
cadence) are function of body height, weight, and
gender (Abdelkader, Cutler and Davis, 2002). In
health applications we want to detect physiological,
pathological and mental characters of people by their
walk style. The technique of gait recognition, as an
exciting research area of biomedical information
detection, attracts more and more attention.
Several techniques can be used nowadays for
gait analysis: stopwatch and marks on the ground;
march on a pressure mat; range laser sensors
scanning a plane a few centimetres above the floor;
video recordings, etc., but new devices like mobile
phones or other wearable can also be used if they are
equipped with inertial sensors (gyroscopes and
accelerometers). An extensive review about gait
analysis using wearable sensors can be found in
(Tao et al., 2012). The gait analysis is modulated by
many factors, including extrinsic (terrain, clothing,
etc.) and intrinsic (gender, weight, age, etc.) ones,
but also psychological (e.g. emotions) and
pathological (e.g. neurological diseases or
psychiatric disorders) ones. Focussing in these last
factors, gait can be used in elderly people for early
diagnostic of Parkinson’s disease (PD) (Barth et al.,
2011), for the screening of knee osteoarthritis
disease (Turcot et al., 2008), for detecting walking
behaviour abnormalities that may indicate the onset
of adverse health problems, or for the progression of
neurodegenerative diseases (El Sayed et al., 2010).
The presence of gait abnormalities in elderly persons
is often a significant predictor of the risk of the
development of dementia, especially non-
Alzheimer’s dementia (Verghese, et al., 2002). Also,
gait can be used in the mobility and fall risk
assessment of elders (Stone et al., 2014).
One of the most typical applications of gait
analysis is in Parkinson’s disease (PD) because PD
is commonly characterized by motor dysfunctions,
such as resting tremors, slowing of movement, gait
difficulty, and limb rigidity. Hence, gait has been
verified as one of the most reliable diagnostic signs
of this disease. Early diagnosis and effective therapy
monitoring of PD is an important prerequisite to
treat patients and reduce health care costs. In Barth
et al. (2011), a mobile sensor based gait analysis
system was developed to measure gait patterns in
PD and to distinguish mild and severe impairment of
BIOSIGNALS2015-InternationalConferenceonBio-inspiredSystemsandSignalProcessing
368
gait. Examinations of 16 healthy controls, 14 PD
patients in an early stage, and 13 PD patients in an
intermediate stage were included. The system was
able to classify patients and controls (for early
diagnosis) with a sensitivity of 88% and a specificity
of 86%. In addition it was possible to distinguish
mild from severe gait impairment (for therapy
monitoring) with 100% sensitivity and 100%
specificity.
The other common application of gait analysis is
related to fall detection. One third of elderly people
fall each year, and half of them experience recurrent
falls (Mortaza, Osman and Mehdikhani, 2014). A
fall could be defined as a situation in which the
individual comes to rest at a lower level (e.g. floor)
accidentally. Daily activity monitoring and fall
detection is important to healthcare for the elderly
and patients with chronic diseases. Different devices
can be used to measure the spatio-temporal
parameters of gait for fall detection. Video cameras
or infrared cameras are the most common methods,
but also accelerometers are well used these last
years. Typically, measured variables are the
cadence, the stride time, the duration of single and
double support, the walking speed, the stride length
the step length, and the step width. According to
Mortaza, Osman and Mehdikhanim (2014), effect
size analysis showed that time variability, gait speed,
stride length and step length were the spatio-
temporal parameters that were the most different
between elderly fallers and non-fallers. Finally in
Patel et al. (2012), a low-cost, continuous,
environmentally mounted monitoring system based
in a Kinect device is used in order to compute a new
metric, the average in-home gait speed (AIGS).
Results demonstrate that AIGS outperforms
traditional instruments used for mobility and fall risk
assessment of elderly adults.
4 DISCUSSION AND
CONCLUSIONS
This work has aimed to explore behavioural
biometrics for health monitoring in the elderly. In
this scenario, biometrics has been presented as a
technical tool to be considered due to its capacity in
dealing with these tasks. Three different examples of
biometric traits have been presented, covering
applications like AD, PD, OSA or falls.
We can observe that behavioural characteristics
are good traits in order to early detect changes in the
health status. On the other hand, due to the great
variability of these traits (we have to remember that
behavioural traits are also affected by many other
characteristics like physiology, changing
environment, etc.), it is very interesting to develop a
particular biometric system to follow the dynamics
of changes for each individual. For example, we can
be interested in monitoring the speech characteristics
of an elderly subject in order to detect when he/she
starts using shorter sentences and long pauses. This
change in speech may indicate the start of some
cognitive impairment. We can also monitor the
handwriting evolution of an individual during a
certain period of time in order to detect the
beginning of imperceptible tremors in the hands.
Thanks to new wearable devices (a very
interesting review of wearable sensors and systems
with application in rehabilitation can be found in
Patel et al., 2012) and the new capabilities of mobile
phones, new apps can be designed using signal
processing algorithms for health biometrics. This
fact links our field of interest with big data analysis,
in which the main focus and challenges are related to
the acquisition, analysis, sharing, storage, transfer,
visualization or privacy violations of the amount of
personal data collected.
Therefore, a very interesting field of research,
combining behavioural biometrics, health, wearable
and big data may emerge and become a challenge
for the scientists coming from these disciplines.
ACKNOWLEDGEMENTS
This work has been supported by the European
Union trough SEACW project (ICT-PSP-2012).
REFERENCES
Abdelkader, C. B., Ross Cutler, Larry Davis. View-
invariant Estimation of Height and Stride for Gait
Recognition. Biometric Authentication. Lecture Notes
in Computer Science Volume 2359, 2002, pp 155-167
Date: 28 May 2002.
Asıcıoglu, F., N. Turan. Handwriting changes under the
effect of alcohol, Forensic Sci.Int. 132 (2003) 201-
210.
Barth J., Klucken J., Kugler P., Kammerer T., Steidl R.,
Winkler J, Hornegger J, Eskofier B. Biometric and
mobile gait analysis for early diagnosis and therapy
monitoring in Parkinson's disease. Conf Proc IEEE
Eng Med Biol Soc. 2011; 2011:868-71. doi:
10.1109/IEMBS.2011.6090226.
Broderick, M., A. Van Gemmert, H. Shill, G. Stelmach.
Hypometria and bradykinesia during drawing
APreliminaryReviewofBehaviouralBiometricsforHealthMonitoringintheElderly
369
movements in individuals with Parkinson's disease
Exp. Brain Res., 197 (3) (2009), pp. 223–233.
Clarke, R. Human identification in information systems:
Management challenges and public information issues.
Available in http://www.anu.edu.au/people/
Roger.Clarke/ DV/HumanID.html, Dec. 1994 –
Accessed 22/10/2014.
Caples, S. M., A. S. Gami, V. K. Somers Obstructive sleep
apnea. Ann. Intern. Med., 142 (3) (2005), pp. 187–
197.
Drotár, Peter, Jiří Mekyska, Irena Rektorová, Lucia
Masarová, Zdenek Smékal, Marcos Faundez-Zanuy.
Analysis of in-air movement in handwriting: A novel
marker for Parkinson's disease. Comput Methods
Programs Biomed. 2014 Sep 17. pii: S0169-
2607(14)00320-4. doi: 10.1016/j.cmpb.2014.08.007.
Eurostat (2012) Statistics Explained Archive, Vol. 2
Social Statistics (http://epp.eurostat.ec.europa.eu/
cache/ITY_OFFPUB/KS-FM-13-002/EN/KS-FM-13-
002-EN.PDF/) – Accessed 22/10/2014.
El Sayed M., Alsebai A., Salaheldin A., El Gayar N.,
ElHelw M. Ambient and wearable sensing for gait
classification in pervasive healthcare environments.
Proceedings of the 12th IEEE International
Conference on e-Health Networking Applications and
Services (Healthcom); Lyon, France. 1–3 July 2010;
pp. 240–245.
Faundez-Zanuy, M., E. Sesa-Nogueras, J. Roure-Alcobé,
J. Garré-Olmo, K. Lopez-de-Ipiña, J. Solé-Casals,
“Online Drawings for Dementia Diagnose: In-Air and
Pressure Information Analysis”, XIII Mediterranean
Conference on Medical and Biological Engineering
and Computing 2013 IFMBE Proceedings Volume 41,
2014, pp 567-570.
Foley, R. G., A Lamar Miller. The effects of marijuana
and alcohol usage on handwriting, Forensic Sci.Int. 14
(1979) 159-164.
Forbes, K. E., M. F. Shanks, A Venneri. The evolution of
dysgraphia in Alzheimer’s disease, Brain Res.Bull. 63
(2004) 19-24.
Kushida, C. A., M. R. Littner, T. Morgenthaler, C. A.
Alessi, D. Bailey, J. Coleman, L. Friedman, M.
Hirshkowitz, S. Kapen, M. Kramer, T. Lee-Chiong,
D.L. Loube, J. Owens, J.P. Pancer, M. Wise. Practice
parameters for the indications for polysomnography
and related procedures: an update for 2005.
López-de-Ipiña, K., J. B. Alonso, C. M. Travieso, J. Solé-
Casals, H. Egiraun, A. Ezeiza, N.Barroso, M.
Faundez-Zanuy, M. Ecay-Torres, P. Martinez-Lage,
U. Martinez-de-Lizardui, “On the selection of non-
invasive methods based on speech analysis oriented to
automatic Alzheimer Disease diagnosis”, Sensors, vol.
5, pp. 6730-6745, 2013. DOI: 10.3390/s130506730.
López-de-Ipiña, K., J. B. Alonso, J. Solé-Casals, N.
Barroso, P. Henriquez, M. Faundez-Zanuy, C. M.
Travieso, M. Ecay-Torres, P. Martínez-Lage, H.
Eguiraun, “On Automatic Diagnosis of Alzheimer’s
Disease Based on Spontaneous Speech Analysis and
Emotional Temperature”, Cognitive Computation,
2013. 10.1007/s12559-013-9229-9.
López-de-Ipiña, K., J. Solé-Casals, , H. Eguiraun, J.B.
Alonso, C.M. Travieso, A. Ezeiza, N. Barroso, M.
Ecay-Torres, P. Martinez-Lage, B. Beitia. Feature
selection for spontaneous speech analysis to aid in
Alzheimer’s disease diagnosis: A fractal dimension
approach. Computer Speech & Language Available
online 27 August 2014. DOI: 10.1016/
j.csl.2014.08.002.
Mortaza N1, Abu Osman NA, Mehdikhani M. Are the
spatio-temporal parameters of gait capable of
distinguishing a faller from a non-faller elderly? Eur J
Phys Rehabil Med. 2014 Apr 24.
Neils-Strunjas, J., K Groves-Wright, P Mashima, S
Harnish. Dysgraphia in Alzheimer's disease: a review
for clinical and research purposes, J.Speech
Lang.Hear.Res. 49 (2006) 1313-1330.
Patel, Shyamal, Hyung Park, Paolo Bonato, Leighton
Chan and Mary Rodgers. A review of wearable
sensors and systems with application in rehabilitation
Journal of NeuroEngineering and Rehabilitation 2012,
9:21 doi:10.1186/1743-0003-9-21.
Phillips, J. G., RP Ogeil, F Müller. Alcohol consumption
and handwriting: A kinematic analysis, Human
Movement Science. 28 (2009) 619-632.
Rosenblum, S., S Parush, P Weiss. The in Air
Phenomenon: Temporal and Spatial Correlates of the
Handwriting Process, Perceptual & Motor Skills. 96
(2003) 933.
Sasse, M. Angela, and Kat Krol. "Usable biometrics for an
ageing population." (2013): 303-320.
Solé-Casals, J., C. Munteanu, O.Capdevila Martín, F.
Barbé, C. Queipo, J.Amilibia, J. Durán-Cantolla.
Detection of severe obstructive sleep apnea through
voice analysis. Applied Soft Computing, Volume 23,
October 2014, Pages 346–354 DOI:
10.1016/j.asoc.2014.06.017.
Stone E., Skubic M., Rantz M., Abbott C., Miller S.
Average in-home gait speed: Investigation of a new
metric for mobility and fall risk assessment of elders.
Gait & Posture. 2014 Sep 6. DOI: 10.1016/
j.gaitpost.2014.08.019.
Tao, W., Tao Liu, Rencheng Zheng,Hutian Feng. Gait
Analysis Using Wearable Sensors. Sensors (Basel).
2012; 12(2): 2255–2283. doi: 10.3390/s120202255.
Tucha, O., S Walitza, L Mecklinger, D Stasik, T Sontag,
KW Lange. The effect of caffeine on handwriting
movements in skilled writers, Human Movement
Science. 25 (2006) 523-535.
Tucha, O., L. Mecklinger, J. Thome, A. Reiter, G.L.
Alders, H. Sartor, M. Naumann, K.W. Lange
Kinematic analysis of dopaminergic effects on skilled
handwriting movements in Parkinson's disease J.
Neural Transm., 113 (5) (2006), pp. 609–623
http://dx.doi.org/10.1007/s00702-005-0346-9.
Turcot K, Aissaoui R, Boivin K, Pelletier M, Hagemeister
N, de Guise J. A. New accelerometric method to
discriminate between asymptomatic subjects and
patients with medial knee osteoarthritis during 3-d
gait. IEEE Trans Biomed Eng. 2008 Apr; 55(4):1415-
22.
BIOSIGNALS2015-InternationalConferenceonBio-inspiredSystemsandSignalProcessing
370
Verghese J., Lipton R. B., Hall C. B., Kuslansky G., Katz
M. J., Buschke H. Abnormality of gait as a predictor
of non-Alzheimer's dementia. N Engl J Med. 2002
Nov 28; 347(22):1761-8.
Werner, P., S. Rosenblum, G Bar-On, J. Heinik, A.
Korczyn. Handwriting process variables
discriminating mild Alzheimer's disease and mild
cognitive impairment, J.Gerontol.B
Psychol.Sci.Soc.Sci. 61 (2006) p228-36.
Zanuy, M. F. and M. Chetouani, “Nonlinear predictive
models: overview and possibilities in speaker
recognition,” Workshop on Nonlinear Speech
Processing, Springer, 2005, pp. 170-189.
APreliminaryReviewofBehaviouralBiometricsforHealthMonitoringintheElderly
371
