Research of Motion Artefacts in Eye Blood Filling Diagnostics by

Photoplethysmographic Methods

Y. S. Kadochkin

, P. V. Luzhnov

and E. N. Iomdina

Bauman Moscow State Technical University, 5, 2-nd Baumanskaya St., Moscow, Russian Federation

Hemlholtz National Medical Research Center of Eye Diseases, 14/19, Sadovaya-Chernogryazskaya St.,

Moscow, Russian Federation

Keywords: Photoplethysmography, Blood Filling, Motion Artefacts, Eye.

Abstract: The analysis of the blood filling process is necessary to study the physiological characteristics of the blood

circulation of the eye in normal and pathological conditions. In this paper, we studied the possibility of

constructing a photoplethysmographic (PPG) diagnostic channel for assessing the hemodynamic

characteristics of an eye and an eyelid. The main objective of this paper is to study the possibility of using

PPG together with other diagnostic channels, as well as the study of motion artefacts when the PPG sensor is

installed on the upper eyelid. The research group consisted of 6 volunteers without ophthalmologic disease.

The red and infrared emission ranges for the PPG sensor were selected. Registration of PPG signals consisted

of the following: PPG measurement in red and infrared light in the presence of artefacts in the horizontal

plane, in the resting state, in the presence of artefacts in the vertical plane. It is shown that the infrared PPG

channel has a greater signal-to-noise ratio (SNR) for both vertical and horizontal motion artefacts of the eye.

As a result of this work, studies have shown that the infrared PPG signal is less affected by eye motion

artefacts than red PPG. Moreover, the recorded signals in the conditions of vertical eyes movements have a

lower SNR than in the conditions of horizontal eyes movements. The method of video PPG has shown better

value of SNR by the diagnostics in the eyelid.

1 INTRODUCTION

The analysis of the blood filling process in the eye is

necessary to compare diagnostic data obtained by

various methods. Ultrasound methods, laser Doppler

flowmetry, optical coherence tomography

angiography and others methods are used in ocular

blood flow research (Kurysheva, 2017). As a rule,

these methods do not allow to make a general

estimation of a blood flow status in eye vessels,

investigating blood circulation in the each vessel

separately. To quantify blood flow the transpalpebral

rheoophthalmography (TP ROG) method is used

(Luzhnov, 2015; Luzhnov, 2017). TP ROG allows to

investigate an eye hemodynamic in physiological

conditions. With TP ROG it is possible to test the

state of hemodynamics in the ciliary body, and,

integrally, in the anterior segment of the eye

(Luzhnov, 2017; Lazarenko, 1999) in case of myopia,

diabetic retinopathy, glaucoma and other

ophthalmologic diseases. Blood flow in the eyelid

https://orcid.org/0000-0003-1111-7063

renders the major factor on the measurements

accuracy for TP ROG (Shamaev, 2017; Shamaev,

2018). The TP ROG method, supplemented by an

analysis of the blood flow in the upper eyelid,

provides more accurate diagnostic data. The addition

of TP ROG data with photoplethysmography method

(PPG) in the eyelid is of great diagnostic significance.

It is necessary to notice that with the help TP ROG

and PPG research techniques it is possible to size up

blood supply on different depth that will allow to

estimate the blood flow contribution of blood vessels

of a certain rank and a location into ocular vascular

system as a whole. Essential advantages of non-

invasive methods, such as PPG and ROG, are also an

absence of contact with eye surface, small duration of

research (2-5 min), mobility and the low cost price of

the equipment.

The main task of this work was development of

PPG channel design, capable of working together

with TP ROG channel, and also researching of motion

artefacts in PPG channel at its use on an eyelid.

288

Kadochkin, Y., Luzhnov, P. and Iomdina, E.

Research of Motion Artefacts in Eye Blood Filling Diagnostics by Photoplethysmographic Methods.

DOI: 10.5220/0009175302880291

In Proceedings of the 13th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2020) - Volume 1: BIODEVICES, pages 288-291

ISBN: 978-989-758-398-8; ISSN: 2184-4305

2 MATERIALS AND METHODS

PPG technology is used in a wide range of medical

devices available for measuring blood oxygen

saturation, measuring blood pressure and cardiac

output, evaluating the autonomic function of the

heart, and detecting peripheral vascular disease

(Tamura, 2014; Allen, 2007; Liu, 2016). To

implement the PPG channel, only a few

optoelectronic components are required: a light

source for illuminating biological tissue and a

photodetector with a recording channel for measuring

intensity changes modulated by pulse waves.

The PPG sensor layout on the patients face is

presented in the Fig.1. There are two possible

locations for ocular blood filling research: for PPG

signal registration from eyelid (1) and for registration

of PPG signal from temporal area (2) are shown.

Figure 1: Two possible locations of the PPG sensor: on the

eyelid (1) and on the temporal area (2), together with video

PPG area (3).

The integrated circuit MAX30100 was chosen as

basic element of PPG sensor. The full-functional

design of the sensory module is incarnated in the

chosen microcircuit for creation of portable

pulsoximetry systems with high demands to accuracy

of measurements. The minimum of additional

external components is required for integration into a

wearable full-function measuring system. Two light-

emitting diodes and the photodiode form a part of this

microcircuit, and also optical elements are built into

it. The electronic circuit of a signal processing

situated in a structure of the sensor is characterised by

low level of own noise. The PPG sensor has two light-

emitting diodes: red and infrared. Light radiated by

light-emitting diodes is close to the monochromatic.

Light radiated by the infrared diode is located in the

wave band with length from 870 nm to 900 nm and

with a peak wave length about 880 nm, and for the

red diode – from 650 nm to 670 nm with a peak wave

about 660 nm.

The measurement was carried out using the

developed model of the PPG sensor. The signals were

recorded with a sampling frequency of 100 Hz. The

currents supplying the LEDs were chosen so that the

amplitudes of the PPG pulse waves of the red and

infrared channels were equal. Studies for all subjects

were carried out at the same temperature of 23°C, in

order to exclude the influence of thermoregulation

mechanisms on the results.

The study group included 6 volunteers who did

not have cardiovascular and ophthalmological

disease. The average age of test persons is 25 years ±

2 years. Before starting the measurements, each

volunteer was informed about the objectives of the

research and was instructed on the measurement

procedure.

The scheme for conducting one registration was

as follows. The test person occupied a sitting position,

after which our PPG sensor was installed in the

position 1 (see the Fig.1). After resting for three

minutes, the registration of the PPG signal began. It

consisted of the following steps:

1) Measurement of PPG in red and infrared light in

the presence of horizontal plane movement artefacts.

2) Measurement of PPG in red and infrared light at

rest.

3) Measurement of PPG in red and infrared light in

the presence of vertical plane movement artefacts.

Each stage lasted 15 seconds and followed one

after another without pauses and stops. A typical view

of signal trends recorded using the developed sensor

is shown in the Fig.2. It is an example of PPG signal

with four seconds duration, recorded at rest, in units

of ADC samples.

Figure 2: The typical trends of the PPG signals from the

closed eyelid at rest.

In our study, motion artifacts were created as

follows. According to the research plan, the subject

changed the view direction with closed eyelids

throughout the entire time of the corresponding

measurement stage. For the case of horizontal

movements, it was a change in the view direction to

the right and left. For the case of vertical movements,

Research of Motion Artefacts in Eye Blood Filling Diagnostics by Photoplethysmographic Methods

289

it was a change in the direction of looking up and

down. At each of these stages (step 1 and step 3), the

test subject made three movements in one second. It

was corresponded to a frequency of motion artifacts

introduced into the recorded signal, equal to 3 Hz.

The frequency of the introduced artifacts was set for

the subjects by the sound signals of a computer

program simulating a metronome. The frequency of 3

Hz was chosen as the largest range of comfortable

movements for the subjects. Moreover, it was higher

than the heart rate (about 1 Hz at 60 beats per minute).

It allowed us to separate the pulse fluctuations in

blood flow and motion artifacts in the subsequent

digital processing of signals using band-pass filters.

The analysis of the received signals can be carried

out both in the time and in the frequency domain. In

the first case, the maximum signal amplitude is

estimated for one period of the pulse blood volume or

for the entire signal recording interval. Then it

becomes possible to compare the changes in the

amplitude of the signal at rest and with motion

artifacts.

For analysis in the frequency domain, it can be

possible to use digital filters to highlight the

frequency ranges of the useful signal (pulse blood

supply) and motion artifacts. Based on the spectral

characteristics of these ranges, the signal-to-noise

ratio (SNR) for each recorded signal can be

calculated.

We used eighth-order band-pass Chebyshev

filters for analysis of registered PPG signals. The first

filter was with a lower cut-off frequency of 0.1 Hz

and an upper cut-off frequency of 2.5 Hz for blood

flow pulsative waves. The second filter was with a

lower cut-off frequency of 2.5 Hz and an upper cut-

off frequency of 3.5 Hz for motion artefacts diapason.

The signal spectrum was obtained in the

MATLAB software environment. From each

obtained signal spectrum, the SNR was calculated.

SNR was calculated as the ratio of the signal for the

pulse wave range to the signal for the range of motion

artefacts. The power in each of the frequency bands

was calculated as the value of the sum of samples in

the interval with boundaries corresponding to the

values of the boundaries of the selected ranges. Then

the analysis and statistical processing of the results

was carried out.

In addition to the PPG method, a contactless video

PPG method was used in our research (Rubins, 2010;

Rubins, 2016). In this case, there is no contact with

both the surface of the eyelid and deeper lying tissues.

In our research, the rectangular area for analysis on

the image was highlighted along anatomical lines

defining the edges of the upper eyelid. A study area

selection diagram is given in Fig.3.

Figure 3: The study area selection diagram for video PPG

method in the upper eyelid.

In our study, the first area (see Fig.3) was chosen

to calculate the video PPG (vPPG) signal parameters.

3 RESULTS

As result, the following SNR ratios were obtained. To

simplify the visual comparison of the experimental

results, the Fig.4 shows the range of SNR values.

Figure 4: Diagram of the SNR range for the results of

measurements.

The span diagrams with HM_(IR) and HM_(R)

provide the results of SNR ratios for measuring the

PPG signal under horizontal oscillation conditions for

the infrared and red ranges, respectively. Span

diagrams VM_(IR) and VM_(R) - for PPG signals of

infrared and red ranges at vertical oscillation (see

Fig.4). The results of the study showed that the

infrared PPG signal is less affected by motion

BIODEVICES 2020 - 13th International Conference on Biomedical Electronics and Devices

290

artefacts than the red PPG signal. In this case, the

recorded signals in the conditions of eye vertical

oscillations have a lower SNR ratio than for signals

that were recorded under the conditions of eye

horizontal oscillations. This can be explained by a

change in the thickness of the eyelid layers in which

the PPG signal is recorded.

Even higher SNR ratios can be achieved using the

vPPG method. This method demonstrates the spectral

component of the signal, similar to the previous PPG

method. When the signal vPPG is decomposed into

components, it is characterized similarly to the signal

recorded from the eyelid skin surface.

4 CONCLUSIONS

As a result of the studies, it was shown that using the

proposed method, it is possible to carry out a

quantitative assessment of blood flow both in the

eyelid skin and in deeper tissues. This allows

supplementing the non-invasive diagnostic method

with a new research algorithm. In this case, the doctor

receives additional diagnostic information about the

blood flow both in the eyelid and in eye structures. It

also becomes possible to calculate blood flow

parameters in each structure of the study area, which

increases the diagnostic value of such studies.

The conducted researches confirm possibility of

using PPG sensor on the closed eyelid. Authors note

a possibility of simultaneous research TP ROG and

PPG signals for the control of individual eyelid

features and for the rising accuracy of ocular blood

filling determining during transpalpebral diagnostics

in the future.

CONFLICT OF INTEREST

The authors declare that they have no conflict of

interest. The paper was supported by a grant from

RFBR (No.18-08-01192).

REFERENCES

Allen, J., 2007. Photoplethysmography and its application

in clinical physiological measurement. Physiol. Meas.

28(3): R1–R39.

Kurysheva, N.I., Parshunina, O.A., Shatalova, E.O., et al.,

2017. Value of structural and hemodynamic parameters

for the early detection of primary open-angle glaucoma.

Curr. Eye Res. 42(3): 411–417.

Lazarenko, V.I., Kornilovsky, I.M., Ilenkov, S.S. et al.,

1999. Our method of functional rheography of eye.

Vestnik oftalmologii 115(4): 33–37.

Liu, J., Ping-Yen Yan, B., Wen-Xuan Dai, et al., 2016.

Multi-wavelength photoplethysmography method for

skin arterial pulse extraction. Biomedical optic express

7(10): 4326. DOI: 10.1364/BOE.7.004313

Luzhnov, P.V., Shamaev, D.M., Iomdina, E.N. et al., 2015.

Transpalpebral tetrapolar rheoophtalmography in the

assessment of parameters of the eye blood circulatory

system. Vestn Ross Akad Med Nauk 70(3): 372–377.

DOI: 10.15690/vramn.v70i3.1336

Luzhnov, P.V., Shamaev, D.M., Iomdina, E.N., et al., 2017.

Using quantitative parameters of ocular blood filling

with transpalpebral rheoophthalmography. IFMBE

Proceedings 65: 37–40. DOI:10.1007/978-981-10-

5122-7_10

Rubins, U., Erts, R., and Nikiforovs, V., 2010. The blood

perfusion mapping in the human skin by

photoplethysmography imaging. IFMBE Proceedings

29: 304–306.

Rubins, U., Spigulis, J., Miscuks, A., 2016.

Photoplethysmography imaging algorithm for

continuous monitoring of regional anesthesia. Proc. of

14th ACM/IEEE Symposium on Embedded Systems for

Real-Time Multimedia : 67-71.

Shamaev, D. M., Luzhnov, P. V., Iomdina, E. N., 2018.

Mathematical modeling of ocular pulse blood filling in

rheoophthalmography. IFMBE Proceedings 68(1):

495–498. DOI: 10.1007/978-981-10-9035-6_91

Shamaev, D.M., Luzhnov, P.V., Iomdina, E.N., 2017.

Modeling of ocular and eyelid pulse blood filling in

diagnosing using transpalpebral rheoophthalmography.

IFMBE Proceedings 65: 1000–1003. DOI:10.1007/9

78-981-10-5122-7_250

Tamura, T., Maeda, Y., Sekine, M., and Yoshida, M., 2014.

Wearable photoplethysmographic sensors – past and

present. Electronics 3(2): 282–302.

Research of Motion Artefacts in Eye Blood Filling Diagnostics by Photoplethysmographic Methods

291