![](bg2.png)
check for the occurrence of any problem in
thecoronary or peripheral arteries.
Direct blood pressure measurement is almost
impossible during exercise, due to body artefacts
that might generate noises and disturbances to the
extent that the measurement becomes inaccurate or
unacceptable. It should also be noted that cuff-based
measurement during exercises can be painful, due to
the increase of blood pressure. The Pulse Arrival
Time (PAT) measurement is capable of generating
different information regarding a cardiovascular
system (Poon, 2005). Exercise affects the properties
of cardiovascular and blood, so the viscosity of BP
(Naka, 2003
), diameters of arterial and vessels
(Kingwell, 1997
) and the flexibility of vessels
increases (Zhang, 2007). Indirect BP estimation
using the PAT-approach is cuffless; disadvantages
of the auscultatory and oscillometric methods will be
virtually nonexistent. Furthermore, these
techniquescause a lot of discomfort, pain, and
restrict the mobility of the patients.
PAT is the time interval between the R-peak of
an electrocardiogram (ECG), and a reference point
in a pulse pressure signal in the same cardiac cycle.
The R-peak is used as a reference to demonstrate the
ventricular depolarization. Generally,the pressure
pulse is detected by an optoelectronic set. Photo
Pletysmogram (PPG) or the Pulse Oximetery are the
two common names of devices used for the purpose
of blood pressure pulse signal recording.
PAT is made up of two main components: the
pre-ejection period (PEP) and the vascular transit
time (TT). PEP is defined as the time interval from
theinitial contractions in left ventricular until the
blood is ejected from the heart. It is also classified
asan electro-mechanical delay,while TT is the
duration for blood pulse pressure to propagate via a
segment of arteries. It has been tested and confirmed
that PAT has higher correlations with blood
pressure, rather than only TT, during and after
exercises (Wong, 2011
).
Artificial intelligence such as fuzzy systems and
neural network (NN) is capable of providing a
solution for indirect blood pressure measurement.
An advantage of this method is that they perceive
the system as a black box, and do not require a
mathematical model for estimation. Non-linear in-
out mapping, adaptivity and flexibility (Forouzanfar,
2011
). (Jia-Jung, 2002) proposed a developed model
of Fuzzy logic controller in a non-invasive and
continuous BP in radial arteries. Classification of BP
into different groups such as high, normal and low
has been done in (Colak, 2003
). Using a hybrid
neuro-fuzzy technique, a novel method has been
proposedfor blood pressure estimation by
oscillometric (Forouzanfar, 2011
).
The main goal of this study is to investigate the
cufflessblood pressure estimations before,
during,and after a medical stress test.The correlation
between BP and parameters such as the Heart rate
and PAT will be carried out. Our previous study
indicated that Systolic BP (SBP) and Diastolic BP
(DBP) estimation during the five stages of stress test
is acceptable, based on least-squares regression on
the data derived from 55 subjects (Colak, 2003
).
In this paper, BP estimation, utilizing LS
regression, is retested for 87 healthy subjects. Then,
by using more intelligent methods, we are going to
demonstrate the fact that the accuracy and
correlation of the estimation significantly increases.
This technique will greatly augment our ability to
monitor BP during the medical stress test, and
prevent sudden deathsduring the test.
2 METHODOLOGY
Many research groups conducted research on
indirect and cuffless blood pressure measurements.
The Moens-Kortwege model, experimental
procedures, and dynamics of blood pressure during
the exercise stress test are discussed in this section.
2.1 Corrected Moens-Kortwege Model
The Moens-Kortwege equation describes the
relationship between blood pressure and Pulse Wave
Velocity (PWV). A corrected version of the Moens-
Kortwege equation is presented. When the heart
contracts, the blood pressure wave speed is given by:
1
/2
(1)
where E is elasticity modulus of vessel wall, h is
wall thickness, is density of blood and r is the
vessels radius and is known as Poisson’s ratio,
which is the ratio of transverse to longitudinal strain
(Shahsavari, 2011). Parameters in the equations are
subject-dependent, which means that self-calibration
is necessary.
The linear relationship between PAT and PAT-
HR during the stress test has been investigated, and
for the purpose of calibration, a new method has
been proposed (Colak, 2003
).
2.2 Experimental Procedure
This study was performed on 87 subjects, (52 male),
BIOSIGNALS2014-InternationalConferenceonBio-inspiredSystemsandSignalProcessing
258