Advanced Biomedical Engineering DOI:10.14326/abe.4.

1
4: 16, 2015. Original Paper

Evaluation of Baroreflex Function Using Green Light Photoplethysmogram

in Consideration of Resistance to Artifacts

Makoto ABE,*, # Makoto YOSHIZAWA,** Kazuma OBARA,* Norihiro SUGITA,* Noriyasu HOMMA,*** Tomoyuki YAMBE

AbstractThe maximum cross-correlation coefficient (max) between blood pressure (BP) and heart rate (HR) variability for
frequency components limited to the Mayer wave-related band is useful for the evaluation of baroreflex function. However, con-
tinuous BP measurement with an expensive and bulky measuring device is required to calculate max. This study proposes a sim-
pler method to obtain max using a green light photoplethysmogram (PPG). A green PPG sensor is less affected by motion arti-
facts than a near-infrared PPG sensor. In this study, an electrocardiogram, continuous BP, green PPG, and near-infrared PPG
were obtained from the subjects. HR, mean BP, and pulse transit time were estimated from the signals, and max was subsequent-
ly calculated. Compared to the max obtained from the near-infrared PPG signal, the max obtained from the green PPG signal is
closer in value to the max obtained from mean BP. These results show that the green PPG sensor can be used to estimate baro-
reflex function instead of using continuous BP measurement.

Keywords: baroreflex function, green light emitting diode, photoplethysmogram.

Adv Biomed Eng. 4: pp. 16, 2015.

baroreflex function based on autonomic nervous activity. Howev-

[Link]
er, max requires continuous BP measurement using an expensive
In Japan, the growth in medical expenditure has become a serious and bulky device, and is not practical or convenient for measure-
problem. In order to reduce costs for medical care, prevention and ment at home.
early recognition of lifestyle-related diseases are important. An We have established an easier method to obtain max using
abnormal baroreflex system is linked to circulatory diseases such near-infrared photoplethysmogram (PPG) signals. This method
as hypertension [1, 2]. The baroreflex is a mechanism that con- involves analysis of the pulse transit time (PTT) obtained from an
trols heart activity based on information from blood pressure con- R-wave on the electrocardiogram (ECG) signal and a feature
stantly sensed by the carotid sinus baroreceptor and the aortic point on the PPG signal, or independent component analysis to
arch baroreceptors. Therefore, we hypothesize that monitoring extract the BP-related parameters [9, 10]. However, PPG sensors
baroreflex function may lead to prevention of lifestyle-related dis- with near-infrared LEDs are highly sensitive to motion arti-
eases. facts [11, 12], an issue that has to be resolved for monitoring max
To evaluate baroreflex function based on autonomic nervous at home.
activity, methods related to heart rate variability, such as coeffi- In this paper, we propose a method to obtain max with a PPG
cient of variation of RR intervals (CVRR), RR at 50% of QTmax sensor using green LED. Previous studies indicate that green PPG
(RR50), and low frequency/high frequency ratio (LF/HF), have sensors are more resistant to motion artifacts than near-infrared
been used [36]. However, these indices have problems with time PPG sensors [1315]. Therefore, a green PPG sensor is often
resolution and reproducibility. We previously proposed a physio- used to estimate the state of blood flow and to measure heart rate.
logical index, max, representing the maximum cross-correlation However, few researchers have estimated baroreflex function
coefficient between blood pressure (BP) and heart rate (HR) vari- based on autonomic nervous activity using an index obtained
ability for frequency components limited to the Mayer wave-re- from green PPG signals.
lated band [79]. We confirmed that max is effective to estimate Therefore, we verified whether a green PPG signal can be
used to estimate baroreflex function and replace the conventional
continuous BP measurement device.
This study was presented at the Symposium on Biomedical Engi- [Link]
neering 2014, Tokyo, September, 2014.
Received on August 1, 2014; revised on October 6, 2014; accept- 2.1Photoplethysmogram
ed on November 11, 2014. A photoplethysmogram is obtained optically from volumetric
*
Graduate School of Engineering, Tohoku University, Sendai, Ja- measurement of the arteries in the fingers or ears. A commonly
pan. used PPG sensor consists of a near-infrared LED and a photodi-
**
Cyberscience Center, Tohoku University, Sendai, Japan. ode to detect reflection at the surface of the fingers or ears. PPG
***
Graduate School of Medicine, Tohoku University, Sendai, Japan. signals can be measured noninvasively, easily, and inexpensively.

Institute of Development, Aging and Cancer, Tohoku University, Recently, a PPG sensor consisting of a green LED has been
Sendai, Japan. studied. The main difference between the near-infrared PPG and
#
63 Aoba, Aramaki, Aoba-ku, Sendai 9808578, Japan. the green PPG is the depth of light penetration from the skin [16,
E-mail: abe@[Link] 17]. The green PPG signal reflects the blood flow in the arteries
 (2) Advanced Biomedical Engineering. Vol. 4, 2015.

Fig. 1Definition of the pulse transit time (PTT).

near the skin surface.

The PTT has been used to obtain a BP-related index from a
PPG [18], and it is defined as the time interval from the R-wave Fig. 2Experimental setup.
on the ECG to a feature point on the PPG (Fig. 1). The PTT cor-
relates inversely with BP because PTT reflects the compliance of
the arteries. The PTT can be calculated using a near-infrared PPG 2.3Experiment
signal or a green PPG signal. In this study, we used the PTT cal- Nine healthy subjects (8 males and 1 female, aged 23.3 1.6
culated from PPG signals to obtain max in the same manner as BP years) participated in this study. The experiment setup is shown
variability. schematically in Fig. 2. The subject rested in a sitting posture in a
chair for 5 min. The subject s ECG, continuous BP, near-infrared
2.2Maximum cross-correlation coefficient PPG, and green PPG were recorded during the experiment. The
The maximum cross-correlation coefficient, max, was calculated ECG was measured by limb leads, and continuous BP was mea-
as described previously [79]. sured from the left middle finger using a Portapres (Finapres
First, let u(t) and v(t) [t=it s, (i=0, 1, 2, ...)] denote the Medical Systems). The reference signal of the near-infrared PPG
time series data (BP and HR variability, respectively) sampled (Nellcor manufactured by EnviteC) was recorded from the left
every t=0.5 s. These data are filtered through a band-pass digi- index finger, as in previous study [10]. The near-infrared PPG
tal filter with a bandwidth between 0.08 and 0.12 Hz to limit the sensor and the green PPG sensor both consisted of an LED and a
frequency components to the Mayer wave-related band. At a cer- phototransistor called a photoreflector. Four pairs of near-infrared
tain time, a Hamming window with an interval between t 60 s and green PPG sensors were applied to the left ring finger, medial
and t+ 60 s is applied to u(t) and v(t), where the length of the side of the wrist, lateral side of the forearm, and lateral side of the
Hamming window was decided experimentally based on the re- upper arm. These PPG sensors were made from commercially
sponse time of autonomic nervous activity. The cross-correlation available photoreflectors. All signals were stored in a personal
coefficient, uv(), for a lag of =jt s where j=..., 1, 0, 1, ..., computer through an amplifier and 16-bit analog-to-digital con-
was calculated as follows: verter (MP150 manufactured by BIOPAC systems Inc.) at a sam-
uv () pling frequency of 1 kHz.
uv () = , (1)
uu (0) vv (0) The experimental protocol was approved by the Tohoku Uni-
where uv() is the cross-correlation function between u(t) and versity Internal Review Board, and informed consent was ob-
v(t), and uu() and vv() are the autocorrelation functions for tained from all subjects before commencing experiments.
u(t) and v(t), respectively. The maximum cross-correlation coeffi-
[Link]
cient, max, and its delay, max, are defined as
Figure 3(a) and (b) show examples of the PTT data obtained
max = max uv () (2)
0s10s from the near-infrared and green PPG, respectively. IRPTT rep-
resents the PTT obtained from the near-infrared PPG sensor, and
max = arg max uv () (3)
0s10s GPTT represents the PTT obtained from the green PPG sensor.
In this study, max was successively calculated every second IRPTTi and GPTTi are the measurements from four sites on the
between t=60 s (start time for calculating max) and t=T60 s body, where i=1 denotes the ring finger, i=2 denotes the wrist,
(end time for calculating max), where T is the end time for the i=3 denotes the forearm, and i=4 denotes the upper arm. In all
experimental data. subjects, GPTTi had smaller standard deviation than IRPTTi, ex-
The max is an index representing the linear correlativity be- cept for the ring finger measurement (i=1). These results indicate
tween BP and HR variability in the time domain. In this study, that the green PPG is superior to the near-infrared PPG for mea-
max was calculated using PTT obtained from the green PPG sig- surement of multiple sites.
nals and HR variability, and also using PTT obtained from the Figure 4(a) shows the comparison between the max calcu-
near-infrared PPG signals and HR variability. lated from the near-infrared PPG [max(IRPTTi)] and the max cal-
culated from continuous BP [max(BP)]. Figure 4(b) shows the
 Makoto ABE, et al: Evaluation of Baroreflex Function Using Green Light PPG (3)

Table 1RMSE between max(BP) and the other max s (n=9).

i=1 i=2 i=3 i=4
(finger) (wrist) (forearm) (upper arm)
near-infrared PPG 0.068 0.268 0.202 0.264
green PPG 0.089 0.080 0.080 0.031

Fig. 5Signal-to-noise ratio for each PTT (n=9).

comparison between the max calculated from the green PPG

[max(GPTTi)] and max(BP). Each max reported was the average
of the values obtained every 30 s for a single subject, and then the
Fig. 3Example of PTT data obtained from the (a) near-infrared PPG
max was averaged for all subjects, as in previous studies [9, 10].
and (b) green PPG.
In addition, Table 1 shows the root mean square errors (RMSEs)
between max(BP) and the other max. The results of Fig. 4 and
Table 1 show that compared to max(IRPTTi), max(GPTTi) is
closer in value to max(BP) in all measurements except that at the
ring finger (i=1). Furthermore, max(IRPTTi) (excluding the ring
finger) is low even though the subjects were in a resting condition.
Therefore, max(IRPTTi) has lower reliability.
The signal-to-noise ratio (SNR) for each PTT is shown in
Fig. 5. In this figure, the error bar represents standard deviation of
each SNR. A desirable signal is defined as one having frequency
components less than 0.3 Hz, based on the frequency band related
to autonomic nervous activity. Two-way analysis of variance was
used to test for differences between near-infrared PPG and green
PPG and differences between four measurement sites. From the
results of this analysis, there was a significant difference (p <
0.05) between near-infrared PPG and green PPG but no signifi-
cant difference between four measurements sites. These results
indicate that GPTTi is more resistant to noise, such as motion ar-
tifacts, than IRPTTi. In particular, the SNR for GPTT4 was higher
than those for other PTT. This result implies that the feature point
on the green PPG used to calculate PTT was determined more
accurately when measured at the upper arm because it is least af-
fected by motion artifacts.
[Link]
The results of Figs. 3, 4, 5 and Table 1 show that GPTT is more
stable than IRPTT.
In Fig. 3(a), IRPTTi (excluding IRPTT1) vary by approxi-
mately 100 ms. In Fig. 4(a), max(IRPTTi) [excluding
Fig. 4The max obtained from (a) near-infrared PPG and (b) green max(IRPTT1)] have low values for the resting condition. These
PPG (n=9). results are not physiologically rational. IRPTT changed widely
 (4) Advanced Biomedical Engineering. Vol. 4, 2015.

mental results showed that the max obtained from a green light
PPG is more stable than that from a near-infrared PPG.
The main limitation of this study was the small number of
subjects. Further experiments with more subjects under various
conditions should be conducted.
In addition, we have to verify that the max obtained from the
green light PPG reflects the change in autonomic nervous activity.
We are planning an experiment to change autonomic nervous ac-
tivity by varying BP intentionally.
Conflict of Interest
We have no conflicts of interest relationship with any companies
or commercial organizations based on the definition of Japanese
Society of Medical and Biological Engineering.
Acknowledgement
This work was supported by JSPS KAKENHI Grant Numbers
24650415 and 25870040.
References
Fig. 6Raw signals from (a) near-infrared PPG and (b) green PPG
placed on the upper arm. 1. Parati G, Di Rienzo M, Bertinieri G, Pomidossi G, Casadei R,
Groppelli A, Pedotti A, Zanchetti A, Mancia G: Evaluation of the
baroreceptor-heart rate reflex by 24-hour intra-arterial blood
because of inaccurate detection of the feature point on PPG pressure monitoring in humans, hemodynamics and sym-
patho-adrenomedullary function. Hypertension. 12(2), pp. 214
(Fig. 1). Figure 6 shows the raw signals obtained from a near-in-
222, 1988.
frared PPG and a green PPG placed on the upper arm of a subject.
2. Grassi G, Cattaneo BM, Seravalle G, Lanfranchi A, Mancia G:
As shown in this figure, the feature points of the near-infrared
Baroreflex control of sympathetic nerve activity in essential and
PPG signal is obscure compare to that of the green PPG signal.
secondary hypertension. Hypertension. 31(1), pp. 6872, 1998.
Therefore, Fig. 5 indicates that the IRPTT cannot be calculated 3. Shimazaki M, Kikuchi K, Kobayakawa H, Yamamoto M, Kudoh
accurately because of its susceptibility to noise or motion arti- C, Wada A, Sakamoto T, Sawai N, Mukai H, Iimura O: The coef-
facts, despite the subject being in a resting condition. This advan- ficient of variation of RR intervals (CVRR) on electrocardiogram
tage of the green PPG is consistent with the conclusions from in patients with essential hypertension with reference to aging,
previous studies [1315]. We speculate that this is due to the fact hemodynamics and sympatho-adrenomedullary function. Nihon
that the light path length of green LED is shorter than that of Ronen Igakkai Zasshi (Jpn J Geriatr). 28(5), pp. 640645, 1991.
near-infrared LED. 4. Ewing DJ, Neilson JM, Travis P: New method for assessing car-
An important finding is that the green PPG sensors provide diac parasympathetic activity using 24 hour electrocardiograms.
reliable PTT data from multiple sites (Fig. 5). Therefore, we may Br Heart J. 52, pp. 396402, 1984.
be able to calculate PTT at two different sites using the green PPG 5. Bigger JT, Fleiss JL, Steinman RC, Rolnitzky LM, Kleiger RE,
signals, without the need for ECG signal. This indicates that the Rottman JN: Frequency domain measures of heart period vari-
green PPG sensor is more suitable as a wearable device for mon- ability and mortality after myocardial infarction. Circulation.
85(1), pp. 164171, 1992.
itoring baroreflex function at home than the near-infrared PPG
6. Sevre K, Lefrandt JD, Nordby G, Os I, Mulder M, Gans RO,
sensor.
Rostrup M, Smit AJ: Autonomic function in hypertensive and
While we obtained good results using the proposed method,
normotensive subjects: the importance of gender. Hypertension.
there are some issues to be considered. A larger number of sub-
37(6), pp. 13511356, 2001.
jects is required to enhance the credibility of these results statisti- 7. Sugita N, Yoshizawa M, Abe M, Tanaka A, Watanabe T, Chiba S,
cally and to determine individual differences. For example, the Yambe T, Nitta S: Evaluation of adaptation to visually induced
best site to measure PPG for each subject should be investigated. motion sickness based on the maximum cross-correlation be-
Furthermore, we should conduct an experiment to confirm that tween pulse transmission time and heart rate. J Neuroeng Reha-
max(GPTTi) reflects the change in autonomic nervous activity. bil. 4, p. 35 (online journal; [Link]
For example, we have to ascertain whether it reflects the change content/4/1/35), 2007.
in sympathetic activity with exercises, Valsalva maneuver and 8. Sugita N, Yoshizawa M, Tanaka A, Abe K, Chiba S, Yambe T,
other activities. Since the depth of penetration of green LED is Nitta S: Quantitative evaluation of effects of visually-induced
lower than that of near-infrared LED, we should examine the ef- motion sickness based on causal coherence functions between
fects of this difference on max. blood pressure and heart rate. Displays. 29, pp. 167175, 2008.
9. Abe M, Yoshizawa M, Sugita N, Tanaka A, Homma N, Yambe T,
[Link] Nitta S: Estimation of baroreflex function using independent
component analysis of photoplethysmography (in Japanese).
To evaluate baroreflex function based on autonomic nervous ac-
IEEJ Trans Electron Inf Syst. 131(9), pp. 15401546, 2011.
tivity, we propose a method to estimate BP-related parameters
10. Abe M, Yoshizawa M, Sugita N, Tanaka A, Homma N, Yambe T,
using a green light PPG instead of a near-infrared PPG. Experi-
 Makoto ABE, et al: Evaluation of Baroreflex Function Using Green Light PPG (5)

Nitta S: Physiological Evaluation of Visually Induced Motion Makoto YOSHIZAWA

Sickness Using Independent Component Analysis of Pho- He received the B.S., M.S. and Ph.D. degrees in
toplethysmogram. Adv Biomed Eng. 2, pp. 2531, 2013. Electrical and Communication Engineering from
11. Hayes MJ, Smith PR: Artifact reduction in photoplethysmogra- Tohoku University in 1978, 1980 and 1983, re-
phy. Appl Opt. 37(31), pp. 74377446, 1998.
spectively. He was a Research Associate from
12. Hayes MJ, Smith PR: A new method for pulse oximetry possess-
1983 to 1991 in the same department. Since 1991
ing inherent insensitivity to artifact. IEEE Trans Biomed Eng.
48(4), pp. 452461, 2001. to 1994, he was an Associate Professor in Toyo-
13. Maeda Y, Sekine M, Tamura T: Relationship between measure- hashi University of Technology, Toyohashi, Japan. In 1994, he returned
ment site and motion artifacts in wearable reflected photoplethys- to Tohoku University. He became a Visiting Scientist, Research Insti-
mography. J Med Syst. 35(5), pp. 969976, 2011. tute of Medicine, Johns Hopkins University, Baltimore and Baylor Col-
14. Fallow BA, Tarumi T, Tanaka H: Influence of skin type and wave- lege of Medicine, Houston, U.S.A. in 1999. Since 2001, he has been a
length on light wave reflectance. J Clin Monit Comput. 27(3), Professor in the Research Division on Advanced Information Technol-
pp. 313317, 2013.
ogy, Information Synergy Center (currently, Cyberscience Center), To-
15. Lee J, Matsumura K, Yamakoshi K, Rolfe P, Tanaka S, Yama-
hoku University. He engages in application of virtual reality to medi-
koshi T: Comparison between red, green and blue light reflection
photoplethysmography for heart rate monitoring during motion. cine, intelligent control of artificial hearts, assessment of effects of
Conf Proc IEEE Eng Med Biol Soc. 2013, pp. 17241727, 2013. visual stimulation on humans and tele-healthcare. He was a member of
16. Sandberg M, Zhang Q, Styf J, Gerdle B, Lindberg LG: Non-inva- AdCom of IEEE EMBS From 2009 to 2011. He has been a member of
sive monitoring of muscle blood perfusion by photoplethysmog- Editorial Committee and a councilor of Journal of the Japanese Society
raphy: evaluation of a new application. Acta Physiol Scand. for Medical and Biological Engineering.
183(4), pp. 335343, 2005.
17. Hagblad J, Lindberg LG, Kaisdotter Andersson A, Bergstrand S,
Kazuma OBARA
Lindgren M, Ek AC, Folke M, Lindn M: A technique based on
laser Doppler flowmetry and photoplethysmography for simulta- He received the B.S. degree in Engineering from
neously monitoring blood flow at different tissue depths. Med Tohoku University in 2013. He is currently a stu-
Biol Eng Comput. 48(5), pp. 415422, 2010. dent of a master s course in the Graduate School of
18. Gribbin B, Steptoe A, Sleight P: Pulse wave velocity as a mea- Engineering at Tohoku University. He is studying
sure of blood pressure change. Psychophysiology. 13(1), pp. 86 extraction of biological signals from video images.
90, 1976.

Norihiro SUGITA
Makoto ABE He received the B.S., M.S. and Ph.D. degrees in
He received the B.S., M.S. and Ph.D. degrees in engineering from Tohoku University, Sendai, Ja-
Electrical and Communication Engineering from pan, in 1998, 2001 and 2004, respectively. He was
Tohoku University in 2004, 2006 and 2009, re- a COE Research Fellow from 2004 to 2006 and an
spectively. He was a Postdoctoral Fellow from Assistant Professor from 2006 to 2010 in the De-
2009 to 2010 in Cyberscience Center, Tohoku Uni- partment of Electrical and Communication Engi-
versity. Since 2011, he has been an Assistant Pro- neering, Graduate School of Engineering, Tohoku University. He is
fessor in Graduate School of Engineering, Tohoku University. He en- currently an Associate Professor in the Department of Management
gages in evaluation of autonomic nervous activity using information of Science and Technology, Graduate School of Engineering, Tohoku
circulation system and development of a detection algorithm of fatal University. His research interests include application of virtual reality
arrhythmias for the implantable cardioverter-defibrillator. He has been to medicine, assessment of effects of visual stimulation on humans and
a member of the society of IEEE and the Japanese Society for Medical tele-healthcare. He is a member of Japanese Telemedicine and Telecare
and Biological Engineering. Association and the Society of Instrument and Control Engineers of
Japan.
 (6) Advanced Biomedical Engineering. Vol. 4, 2015.

Noriyasu HOMMA Tomoyuki YAMBE

He received the B.S., M.S. and Ph.D. degrees in He received the M.D. and Ph.D. degrees in Medi-
electrical and communication engineering from cal Science from Tohoku University, Sendai, Ja-
Tohoku University in 1990, 1992 and 1995, re- pan, in 1986 and 1989, respectively. He was a Re-
spectively. From 1995 to 1998, he was a lecturer at search Associate from 1992 at the Division of
the Tohoku University, Japan. From 2000 to 2001, Medical engineering and Clinical investigation and
he was a visiting professor at the Intelligent Sys- Department of Medical Engineering and Cardiolo-
tems Research Laboratory, University of Saskatchewan, Canada. From gy, Institute of Development, Aging and Cancer, Tohoku University.
2008 to 2013, he was an associate professor of the Cyberscience Center He has been a Professor in the some department from 2004. He engag-
at the Tohoku University. He is currently a Professor in Graduate es in artificial heart, autonomic nervous system analysis and telemedi-
School of Medicine, Tohoku University. His current research interests cine. He is a member of Japanese Society for Artificial Organs, the
include neural networks, complex and chaotic systems, soft-comput- Japanese Society for Medical and Biological Engineering, and Japa-
ing, cognitive sciences, medical systems and brain sciences. He has nese Society of Neurovegetative Research.
been an associate editor of Journal of Intelligent & Fuzzy Systems
since 2006 and a member of NNTC of IEEE Computational Intelli-
gence Society since 2007.