Analysis of Gaze Trajectory and Skin Extension Pressure Data in

Blood Collection Technology

Kazuma Mihara

, Takeshi Matsuda

, Yukie Majima

, Seiko Masuda

, Masanori Akiyoshi

Kenji Adachi

and Naoki Taira

Graduate School of Humanities and Sustainable System Sciences, Osaka Prefecture University, Japan

Department of Information Security, University of Nagasaki, Japan

Department of Information Systems Creation, Faculty of Engineering, Kanagawa University, Japan

Department of Computer Science and Intelligent Systems, Graduate School of Engineering,

Osaka Prefecture University, Japan

Keywords: Blood Collection, Gaze, Nursing Skills, Skin Extension, Tacit Knowledge.

Abstract: Up to this time, research on tacit knowledge of blood collection technology has been conducted, but A method

for quantitatively evaluating skills related to blood collection technology and a system that implements them

have not been developed. For the present study, using a sensor that can measure eye gaze movement and

pressure. Collect finger pressure for skin extension and eye gaze trajectory data during blood collection, and

analyze characteristics of pressure distribution during puncture and movement range of the eye gaze and then

to examine a method to quantitatively evaluate a part of blood collection technique procedure.

1 INTRODUCTION

Blood collection, a fundamentally important medical

practice for clinical tests using blood as a sample, is

an invasive nursing technique that percutaneously

punctures a vein and extracts blood (Japanese

Committee for Clinical Laboratory Standard, 2018).

Therefore, it requires skill sufficient for precision and

consideration for the patient simultaneously (Miki, S.

et al., 2011). In fact, the Japan Nursing Association

has pointed out the necessity of further strengthening

education on intravenous injections (Japanese

Nursing Association, 2003). However, because

injection techniques entail so much tacit knowledge

that cannot be expressed in words, it is difficult to

inherit tacit knowledge in learning. An important

difficulty is that nursing students have not mastered

skilled skills (Naoki, U. et al., 2018). Therefore, it is

necessary to formalize tacit knowledge,

quantitatively evaluate skills related to blood

collection technology, and improve technical skills.

In recent years, sensing technology has become an

indispensable technology along with the trend of

Internet of Things. Many devices used worldwide

already have sensing technology.

One of them is a

gaze measuring instrument. In earlier studies have

attempted to pass on tacit knowledge of skilled

workers using gaze measurement (Harumasa, N. et al.,

2014) (Mamiko, S. et al., 2005) .These studies are

also used in the nursing field, such as feature analysis

of skills of skilled nurses (Yasuko, M. et al., 2013)

and nursing skills education (Yukie, M. et al., 2018).

As sensor technology innovations continue to

increase, their accuracy is expected to become greater

and devices are expected to become cheaper, and able

to acquire widely diverse data. Unprecedented

development of systems is expected to occur from

effective utilization of these data.

For the present study, the subject wears glasses-

type tobii pro / glasses 2 (manufactured by Tobey

Technology) as a gaze measuring instrument and a

sensor capable of measuring pressure on the finger

and then a blood collection experiment was

conducted using an arm model. For this experiment,

a simulation model arm was used. We attempted to

extract the characteristic data necessary to evaluate

each procedure by analyzing the pressure data for

skin extension and eye gaze data during puncture. As

a result, it was found that a characteristic shape

appeared in the pressure graph during skin extension

and that the gaze movement range during skin

extension was smaller than that of other procedures.

By looking at these two data in combination, each

procedure may be identified automatically.

Mihara, K., Matsuda, T., Majima, Y., Masuda, S., Akiyoshi, M., Adachi, K. and Taira, N.

Analysis of Gaze Trajectory and Skin Extension Pressure Data in Blood Collection Technology.

DOI: 10.5220/0009164606870692

In Proceedings of the 13th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2020) - Volume 5: HEALTHINF, pages 687-692

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

687

2 RELATED RESEARCH

2.1 Tacit Knowledge during Blood

Collection

In earlier studies, we explored formalization of tacit

knowledge from various viewpoints such as

mathematical analysis of finger movement in

injection technology (Yutaro, Y. et al., 2017) and

calculation method of needle bends (Takeshi, M. et

al., 2017). Tacit knowledge was clarified by

comparing the skills of nurses and nursing students,

particularly addressing skin extension and pressure,

which are movements of nurses’ auxiliary fingers

during blood collection (Naoki, U. et al., 2018)

(Takeshi, M. et al., 2018). These studies, we collected

data of nursing students and skilled workers. And the

studies mainly focus on only skin extension pressure.

But in this study, we collected gaze trajectory data

and skin extension pressure data. This time, we tried

to extract tacit knowledge that is unlikely to make a

difference for active nurses.

2.2 Automatic Evaluation of Medical

Procedures

Earlier studies examined the development of

automated medical procedure evaluation systems

using Deep Learning for objective evaluation of

medical procedures. Reference (Nao,S. et al. 2018)

analyzes the data obtained from kinecct, but this study

is different in that the objective is to evaluate blood

collection objectively by analyzing skin extension

pressure and eye gaze data.

3 EXPERIMENT

3.1 Outline of Experiment

For the present study, we focused on “blood

collection technology”, which is frequently

performed in daily work, among nursing techniques

and measure the pressure for skin extension and eye

gaze trajectory during puncture. Blood vessel samples

were collected using a simulation arm model. A

patient role-player was set in front of subjects to give

a sense of realism of blood collection. subjects were

able to collect blood while talking to the patient.

Table 1 shows the experimental locations and periods.

Fig. 1 shows the experimental environment. In this

experiment, we are also concurrently with a study to

see the difference in brain activity between skilled

nurses and new nurses based on cerebral blood flow

(Takahito, T. et al. 2019).

This research was conducted with the approval of

the Ethics Committee of the Graduate School of

Sustainable System Sciences, Osaka Prefecture

University.

Table 1: Experiment place and period.

Implementation

period

4 days from November to

December 2018

Implementation

location

Practice room in Hospital A

Figure 1: Experimental environment.

3.2 Test Subjects

Subjects were 19 active nurses working in hospitals

who had consented to research cooperation. Table 2

shows “the ladder levels” and numbers of nurses.

“The ladder level” is a nurse development and

evaluation system established by the Japan Nursing

Association. The five levels, from I–V, represent the

nurse ability and career. Higher numbers reflect

higher the nursing practice ability.

Table 2: Nurse ladder level.

3.3 Equipment Used

3.3.1 Pressure Sensor

The equipment used to acquire pressure data for the

present study includes the following.

• Arduino Uno

• Pressure sensor FSR402 (Fig. 2)

HEALTHINF 2020 - 13th International Conference on Health Informatics

688

•

A/D converter

•

1/4 carbon film resistance

•

Small universal board (ICB-90)

•

Heat-resistant electrical wire

Figure 2: Pressure sensor FSR402.

The pressure sensor was fixed on the finger that

extension skin when fixing the blood vessel. As values

of 0–1023 acquired from the sensor increase, a stronger

force is applied and acquired as time series data.

3.3.2 Eye Gaze Measuring Instrument

For gaze measurement, gaze was measured using a

wearable eye tracking system (tobii pro / glasses 2 G2-

100; Tobii Technology) (Fig. 3).

Figure 3: Tobii pro sensors.

3.3.3 Arm Model for Blood Collection

Simulation

For the blood collection simulation, we used an arm

model (LM-086; Koken Co. Ltd.). For selecting a

blood vessel to collect blood, blood vessel models of

several types exist. Fig. 4 shows the arm model with

the blood vessel model. Fig. 5 shows the type and

difficulty of the blood vessel model.

Figure 4: The arm model with the blood vessel model.

Figure 5: The type and difficulty of the blood vessel model.

We explained in advance that exchanging multiple

simulation models. A sample was presented.

However, participants do not know from which blood

vessel model they will collect blood.

4 ANALYTICAL METHOD

As described in this paper, among the 19 subjects, 4

nurses were selected from nurses who had no data

loss and who had pressure sensors on the same finger.

After graphing the pressure data and the eye gaze data

obtained from the experiment, we tagged the

procedures and actions at the time of blood collection

one by one while watching the video, and observed

their mutual relation. The recorded video is the

following (Figs. 6 and 7). The procedure for blood

collection is presented below.

(1) Wrap the tourniquet

(2) Blood vessel selection

(3) Wipe the puncture site with gauze

(4) Hold a syringe

(5) Remove the syringe cap

(6) Puncture

(7) Pull the inner cylinder

(8) Take the tourniquet

(9) Hold the puncture site with gauze

(10) Pull out the needle

The procedures were almost identical, but with some

differences in the order, depending on the person.

Figure 6: Image of camera.

Analysis of Gaze Trajectory and Skin Extension Pressure Data in Blood Collection Technology

689

Figure 7: Wide angle camera from tobii.

5 RESULTS AND DISCUSSION

This study examines characteristics of the changes in

pressure during skin extension and the trajectory data

of the eye gaze, comparing each procedure and

action.

5.1 Skin Extension Pressure Graph

First, we introduce a graph of pressure during skin

extension. Figs. 8–10 show changes in the skin

extension pressure with a scatter diagram (smooth

line).

Figure 8: Nurse ID15, Changes in skin extension pressure.

Figure 9: Nurse ID11, Changes in skin extension pressure.

The pressure data include data other than skin

extension. They include states such as “having a

syringe” and “unscrewing the cap of the syringe”. As

shown in Figs. The characteristic shape was visible.

Regardless of the success or failure of blood

collection, it looked like a rectangle

Figure 10: Nurse ID18, Changes in skin extension pressure.

5.2 Gaze Locus Graph

Next, we introduce a gaze locus graph. Figures 11–14

show scatter plots (smooth lines) of the eye gaze

trajectory until skin extension. Similar to the pressure

graph, this graph includes items other than the eye

gaze during skin extension, and also includes

behavioral and procedural states. The circled range

represents the range of motion of the eye gaze when

the skin extension. Throughout the procedure, the

range of gaze movement during skin extension and

puncture was smaller than that for other procedures.

However, As shown in Figures 11 and 12, it can be

characterized that failure is bigger than success.

Figures 13 and 14 both show data from nurse 18.

Nevertheless, but more than individual differences,

differences in the range of gaze movement during

skin extension are failure is bigger than success. This

time, the gaze patterns of four nurses were analyzed

based on the dominant arm and the finger attached to

the pressure sensor. The relationship between nurses

and ladder levels is shown in Table 3. Results showed

that the gaze momentum at the time of skin extension

or puncture increased when it fails. In addition, it was

found that the higher the ladder level has a tendency

to the longer moving distance of eye gaze when

selecting blood vessels. The table 4 shows the moving

distance of the four nurses' eye gaze. This trend that

the higher the ladder level, the more moving distance

of eye gaze, because the higher the ladder level, the

more pondered the injection location after selecting

the blood vessels. However, it has been pointed out in

post-experiment interviews that the sense of a

patient's arm is vastly different from the sensation

200

400

600

800

1000

1200

0 5 10 15 20 25 30 35

Time-axis

（s）

Pressure

Skin extension

200

400

600

800

1000

0 102030405060

Skin extension

Time-axis

（s）

Pressure

200

400

600

800

1000

1200

0 1020304050

Skin extension

Time-axis

（s）

Pressure

HEALTHINF 2020 - 13th International Conference on Health Informatics

690

with this arm model. So, it is likely that there simply

is variability in gaze distance among participants, not

that there is a trend. Presumably, simple comparison

is difficult. In the future study, we will examine the

arm model validity and it need to do analysis with

more participants.

Figure 11: Nurse ID11-Success, the eye gaze movement

range for skin extension.

Figure 12: Nurse ID10-Failure, the eye gaze movement

range for skin extension.

Figure 13: Nurse ID18- Success, the eye gaze movement

range for skin extension.

Figure 14: Nurse ID18-Failure, the eye gaze movement

range for skin extension.

Table 3: Experiment place and period.

NurseID Ladder level

10 Ⅲ

11 Ⅳ

15 Ⅰ

18 Ⅴ

Table 4: moving distance of the eye gaze.

Ladder Level NurseID The Moving Distance (pixel)

Ⅰ 15 1399

Ⅲ 10

2654

3324

Ⅳ 11

11597

18114

Ⅴ 18 7726

6 CONCLUSION

In this analysis, a characteristic pressure distribution

appeared in the skin development pressure during

blood collection.

We also found that the range of

motion of the gaze at the time of skin extension and

the puncture was smaller than the range of motion of

the eye gaze during other procedures.

7 FUTURE WORK

Based on these results, we can consider the

construction of a general-purpose evaluation system

for procedures. Results of the nurses analyzed this

time tended to show similar characteristic shapes in

pressure during skin extension. However, some

400

600

800

1000

600 800 1000 1200 1400 1600 1800

600

650

700

750

800

850

900

950

1000

1050

1100

200 400 600 800 1000 1200 1400

200

400

600

800

1000

1200

500 700 900 1100 1300 1500

200

400

600

800

1000

1200

600 800 1000 1200 1400 1600

Analysis of Gaze Trajectory and Skin Extension Pressure Data in Blood Collection Technology

691

nurses have shapes appearing twice in a single

procedure. In such cases, it is impossible to determine

which of the two is the skin extension pressure just

before puncture. but the range of motion of the eye

gaze during skin extension and puncture is smaller

than that for other procedures such as “wrapping a

tourniquet” and “blood vessel selection.” Therefore,

it is possible to estimate the skin extension and

puncture work by combining the two data of the

pressure data and the gaze data. Furthermore, the

range of gaze movement at the time of skin extension

and insertion be larger at the time of failure than at

the time of success. This feature will be important for

judging success or failure of the technique.

8 SUMMARY

This study was designed to extract the characteristic

data necessary to evaluate each blood collection

procedure, and to clarify the characteristics of

pressure and the eye gaze movement range for skin

extension during blood collection. Results show that

each procedure can be identified automatically

examining these two data types in combination.

Further study is required for developing

quantitatively evaluating the relationship between

pressure and gaze data in order to realize an automatic

evaluation.

ACKNOWLEDGMENTS

This research was supported by JSPS KAKENHI

19K10808, 19K22774, 17K19845. We deeply

appreciate the nurses, including those of Hospital A,

for their cooperation with this study.

REFERENCES

Japanese Committee for Clinical Laboratory

Standard.2018. Standard Phlebotomy Guideline.

Miki Sato, Hiroko Otsu, Yoko Sota, Arisa Nishio, Tomoko

Tanaka, and Tetsuji Minoura., 2011. Differences

Between Nurse and Student in Eye Tracking During

Venous Blood Collection. Bulletin of Aichi Prefectural

University School of Nursing & Health,17, pp.1-14.

Japanese Nursing Association.,2003. Guidelines for

conducting intravenous injections.

Naoki Ueda, Masao Izumi, Yukie Majima, Takeshi

Matsuda and Yasuko Maekawa., 2018.

Correlation between left hand contact force and skin

development in injection technique.

The Institute of Electronics, Information and

Communication Engineers, IEICE technical report

117(419), pp.11-16.

Harumasa Nishina, Masashi Kume, and Yuka Takai., 2014.

Features of gaze and chasing of expert in metal-fittings-

of-flag. Studies in science and technology, Vol3, NO.1,

pp23-28.

Mamiko Sakata, Yuka Marumo., 2005. Quantitative

Analysis of the Use of Eye Movements in Japanese

Traditional Dance. IPSJ SIG Computers and the

Humanities,2005, pp.9-14.

Yasuko Maekawa, Yukie Majima.,2009. Quantifying Eye

Tracking for Learning the Tacit Knowledge of Nursing

Skill: Differences between Skilled Nurses and Nursing

Students in Intravenous Injection. Japanese Society for

Information and Systems in Education.Vol.27(7),

pp.122-129.

Yukie Majima, Takeshi Matsuda, and Seiko Masuda.,

Development of Wearable Learning System for

Education of Nursing Skills. Studies in Health

Technology and Informatics, vol.264, pp.1720-1721.

Yutaro Yoshida, Takeshi Matsuda, Yukie Majima and

Kousuke Ootani., 2017. Mathematical Analysis and

Consideration on the Hand Motion Data at an Injection

Technique. IPSJ SIG Technical Report, 2017-5J-06.

Takeshi Matsuda, Yukie Majima., 2017. Consideration on

Feature Extraction of Skill Level by Puncture Angle of

Injection Technique. 23

Int’l Conf on Parallel and

Distributed Processing Techniques and Applications

(Accepted)

Takeshi Matsuda, Yukie Majima, and Kousuke Ootani.,

2018. Pressure data analysis for vascular fixation in

venipuncture. The Institute of Electronics, Information

and Communication Engineers,IEICE technical report

117(419), pp.1-3.

Nao Sato, Kenju Akai, Makoto Hirose, Satoru Okamoto,

and Kenji Karino., 2018. Visualization of Acquisition

Experience in Sternal Compression Maneuver Using

Kinect Sensoring: For Co-Creation of Medical

Technique Experiential Values. International Journal

of Automation Technology, Vol.12, No.4, pp.542-552.

Takahito Tamai, Yukie Majima and Tsuneo Kawano., 2019.

A Study on Brain Activities during Blood Draw

Technique

-Viewpoint from cerebral blood flow-. The Institute of

Electronics, Information and Communication

Engineers,IEICE technical report Vol.118, No.509,

PP.11-15.

HEALTHINF 2020 - 13th International Conference on Health Informatics

692