A Left-Right-Asymmetric Pedaling Machine for Medical

Rehabilitation of Lower Limbs

Milun Liu

, Fajian Wu

, Jinhua She

1, 2, 3

, Hiroshi Hashimoto

and Min Wu

2, 3

Graduate School of Bionics, Computer and Media Sciences, Tokyo University of Technology,

Hachioji, Tokyo 192-0982, Japan

School of Automation, China University of Geosciences, Wuhan, Hubei 430074, China

Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems,

Wuhan, Hubei 430074, China

Master Program of Innovation for Design & Engineering, Advanced Institute of Industrial Technology Shinagawa-ku,

Tokyo 140-0011, Japan

Keywords: EMG Sensor, Left-Right Asymmetry, Lower-Limb Rehabilitation, Human Centered, Pedaling Machine.

Abstract: This paper explains a pedaling machine of a left-right-asymmetric type for lower-limb rehabilitation. Since

most commercially available machines for the rehabilitation of lower limbs are symmetric, people with

lower-limb injuries have to adapt themselves to the machines to do exercises. To solve this problem, we

have been developing a new kind of pedaling machine that can easily be used to adapt the requirements for

left-right asymmetry of lower limbs. Main points in the design of a prototype of a half model for one leg of

the machine are summarized. Preliminary tests with a tread force sensor and some electromyogram (EMG)

sensors are carried out and are showed the feasibility of the machine.

1 INTRODUCTION

Maintaining or improving walking ability is

essential to ensure a person’s mental and physical

soundness. However, many people cannot walk

normally due to diseases and/or injuries of lower

limbs, brain damage, or aging. Lower-limb

rehabilitation is an important way to regain the

ability of walking. According to statistics, the

number of people in Japan who are issued a

certification of needed long-term care or a

certification of needed support has been rapidly

increasing in the last decade, and the number of

people was 5.457 million in 2012 (Ministry of

Health, Labour and Welfare, Japan, 2016). As

shown in the same report, there will be more than

one third of the Japanese to become the elderly in

2035. A large number of people who need

rehabilitation will cost large manpower to take care

of them. And the increasing number of people who

need rehabilitation will become to a serious social

problem in the near future. So, developing

rehabilitation machines to help people to train their

walking muscles not only has a positive influence on

people’s physiology and psychology, but also

contributes to the whole society.

Nowadays, a great number of rehabilitation

machines, which are mainly remodelled from

training machines, are often used in rehabilitation.

However, those machines have some problems, such

as bisymmetric pedaling, fixed structure of machines

(for example, Anzai, 2014). Since all those machines

are machine-centered, they are hard to meet all kinds

of requirements of users for the rehabilitation of

lower limbs. They not only may lead to tremendous

pain for users, but also may lead to the degradation

of motivation for rehabilitation. In order to solve

above problems, it is necessary to develop a truly

effective machine for rehabilitation.

Riding a bicycle is effective to train walking

muscles, but it is not suitable for a user with lower-

limb injury. To complete a training task, a user with

different degrees of damage in left and right legs

subconsciously uses the powerful leg with a great

effort. This causes great inadequacy in rehabilitation.

We designed a new type of an asymmetric

pedaling machine for lower-limb rehabilitation and

built a prototype of a half model for one leg to solve

the above mentioned problems (She et al., 2016,

2017a, 2017b). Unlike other ones, it is a human-

652

Liu, M., Wu, F., She, J., Hashimoto, H. and Wu, M.

A Left-Right-Asymmetric Pedaling Machine for Medical Rehabilitation of Lower Limbs.

DOI: 10.5220/0006471306520657

In Proceedings of the 14th International Conference on Informatics in Control, Automation and Robotics (ICINCO 2017) - Volume 1, pages 652-657

ISBN: 978-989-758-263-9

centered machine that the structure of the machine is

easy to be adjusted to suit the different requirements

for the lower-limb injuries. The left and right

pedaling loads and strokes can also be adjusted

independently. In order to obtain information about

rehabilitation, we built a measuring system to carry

out the interaction between exercises and the

computer-based supervision (She et al., 2017a).

This paper summarizes the main points in the

design of the machine. The results of preliminary

tests using a force sensor and some electromyogram

(EMG) sensors are presented to show the validity of

the pedaling machine for lower-limb rehabilitation.

2 KEY POINTS IN DESIGN OF

ASYMMETRIC PEDALING

MACHINE

Pedaling is widely used in training the walking

muscles. There are basically two kinds of pedaling:

rotational and linear. Since a linear type can easily

be used to design a left-right-asymmetrical

mechanism, we used it to build a new type of a

pedaling

machine to suit different requirements for

Figure 1: Optimal pedaling region. (a) Top view and (b)

Side view (Upper case: heel; lower case: toe) (Sato, 1994;

She et al., 2017b).

lower-limb injuries.

To design a pedaling machine, first, we need to

determine the specifications of the mechanism.

As the first step, we selected a pedaling load for

a leg-extension force. According to Sato (1994), the

maximum of the average leg-extension force of one

leg for 20-year-old man is about 2900 N. It

decreases with aging from 20s, and the force of

people in 60s is only about 50% of that of their 20s.

Considering that people who need rehabilitation

have very weak legs, we chose the maximum

pedaling load to be

N. 2000

max



(1)

Then, we determined the range of an adjusting

angle for the pedaling machine. There is an optimal

pedaling region for a normal person (Figure 1), and

the definitions are given in Figure 2. The angle

between the femur and the lower leg is in the range

[15

, 90

] when the knee is at the closest position to

the body, and [30

, 90

] when the knee at the farthest

position from the body. Considering that a person

who needs rehabilitation may not sit and/or pedal as

a normal person does, we chose the angle to be

]90 ,0[







(2)

so that it can provides a larger region than the

optimal one does to satisfy the different

requirements for users. Based on a preliminary test,

we chose the length of stroke of the linear pedaling

mechanism to be

mm. 150



(3)

Figure 2: Definitions (Sato, 1994; She et al., 2017b).

Max:

Seat reference

point











  

Distance forward of

seat reference point [cm]

Distance below

seat reference point [cm]

Floor

(a)

(b)

Height of

knee

Hip-knee

distance

Seat reference

point

Height of

seat

Horizontal line

Near high point

Near low point

Far high poin

Far low point

Seat reference

point

A Left-Right-Asymmetric Pedaling Machine for Medical Rehabilitation of Lower Limbs

653

Lef

lateral view Frontal view Photo (

igh

lateral view)

Figure 3: Prototype of pedaling machine for one leg.

Figure 4: Measuring system (She et al., 2017).

Table 1: Parameters of oil damper, KINECHECK Super K.

Model Overall len

th Stroke Force ran

5001-31-4 356 mm 102 mm 23 ~ 5440 N

2 PEDALING MACHINE AND

MEASURING SYSTEM

We selected components to build a pedaling

machine that satisfies (1)-(3).

An oil damper, KINECHECK Super K (Meiyu

Aimatic Co. Ltd., Japan) (Table 1), was selected to

produce a pedaling load. It has the longest stroke

and produces the largest damping force in small-size

oil dampers. The maximum force is more than twice

max

in (1), but the stroke is about half of

(3). So, we designed a pulley mechanism to enlarge

the stroke two times and to reduce the force to half.

A prototype of the machine for one leg was built

(Figure 3). And an inclined angle of the adjusting

part was designed to be changed from 0

to 90

ensure (2).

As shown in Figure 3, an inclined panel is fixed

to the base by two bolts. The angle from 0

to 90

equally divided by 10

. Resetting and fastening the

bolt turn the inclined panel at a desired angle to suit

a user’s need. The pedal is directly connected to the

oil damper by a steel-wire rope in a pulley

mechanism. The pedaling force is manually adjusted

by turning a nob on the oil damper. The user pushes

Tread force sensor

(LPR-C-1KNS15)

Displacement sensor

(DTS-A-100)

Heart rate meter

(neo HR-40)

Compact recording system

(EDX-10A)

Wireless EMG sensor

(SX230-1000)

9-axes wireless

motion sensor

Wireless set

Transmitter (PH8310)

Amplifier (PH8320)

Data receiver (PH8020)

Data base

Interface

(Windows 8.1)

Display

Hinge joint

Displacement

sensor

Oil damper

Pedal

Pulley

mechanism

Guide rail

Inclined

panel

Base

Adjusting bolt

ICINCO 2017 - 14th International Conference on Informatics in Control, Automation and Robotics

654

Figure 5: A photo of components of measuring system.

the pedal down in a linear motion for exercise. The

pedal returns back to the up position by elasticity

produced by the oil damper.

The following points are considered in the

design of the measuring system:

 It suitably stores measured data in a real-time

fashion.

 It ensures easy access to measured data.

 It displays measured data in a real-time

fashion, and easily switches the display to

interested data.

 It is easy to synchronize data if needed.

A measuring system (Figures 4 and 5) was

constructed to collect data. The measuring system

consists of a heart rate meter (neo Hr-40) (NISSEI

Co. Ltd., Japan), a force sensor (LPR-C-1KNS15)

and a displacement sensor (DTS-A-100) (Kyowa

Electronic Instruments Co. Ltd., Japan), one set of

wireless EMG sensor (SX230-1000) and one set of

9-axes wireless motion sensor (XYZ geomagnetism,

XYZ acceleration, and XYZ angular acceleration)

(DKH, Japan).

The measuring system ensures the possibility of

interaction between exercises and computer-based

supervision and control of medical rehabilitation.

3 PRELIMINARY TESTS

Preliminary tests were carried out for the prototype

for three subjects. A subject sat on a fixed chair in

front of the prototype machine with a determined

distance. The inclined angle was set to be 20°, 50°,

and 70°; and the pedaling force was set from 0 to 60

It was found that pedaling was carried out

smoothly and comfortably for the inclined angle of

50° among the three setting of 20°, 50°, and 70°.

(a)

(b)

(c)

Figure 6: Pedaling force for different inclined angles

[inclined angle: (a) 20º, (b) 50º, and (c) 70º].

Some typical time responses of pedaling force for

different inclined angles are shown in Figure 6. As

can be seen from the figure, the variation of the

pedaling force is the smallest for the inclined angle

of 50º among the three angles. This shows that it

was the easiest position for the user who pedaled the

machine, and it also shows that adjusting the

inclined angle of the pedaling machine can easily

satisfy the needs of individual requirements.

The muscles of quadriceps femoris, biceps

femoris, tibialis anterior, and soleus (Figures 7 and

8) are most closely related to walking. They were

measured in preliminary tests. The EMG signals of

those muscles were recorded.

0 10 20 30 40 50 60

Time [s]

Force [N]

0 10 20 30 40 50 60

Time [s]

Force [N]

0 10 20 30 40 50 60

Time [s]

Force [N]

Displacement sensor

Force sensor

Measuring screen

(PC)

EMG sensor

Wireless module

A Left-Right-Asymmetric Pedaling Machine for Medical Rehabilitation of Lower Limbs

655

Figure 7: Walking muscles (She et al., 2006).

Figure 8: A photo of preliminary test with EMG sensor.

Since the characteristics of the EMG signals can

mainly be observed in the frequency range of 50-150

Hz (Marras, 1992; De Luca, 2002), we chose the

sampling frequency to be 500 Hz. The FFT (fast

Fourier transformation) was used to those signals

(Figure 9). The figure shows that the EMG signals

were recorded properly.

On the other hand, as pointed out by Carlo J. De

Luca (2002), the amplitudes of the EMG signals are

stochastic (random) in nature, and can be reasonably

represented by a Gaussian distribution function. The

usable energy of the signal is in the frequency range

of [0, 200] Hz, with the dominant energy in the

range of [50, 100] Hz. The usable signals are viewed

as those with energy above the electrical noise level.

The noise is usually inherent noise in the electronics

components in the detection and recording

equipment, ambient noise, motion artifacts, and

inherent instability of the signal. The amplitude of

an ambient noise may be one to three times larger

than that of the EMG signals. It is clear from Figure

9 that big amplitudes of spectra at 50 Hz, 100 Hz,

and 200 Hz are considered to be the power noise and

its harmonics. How to abstract true characteristics of

the pedaling motion from the noisy EMG signals is

one of the main tasks in this study, and will be

investigated in the near future.

Figure 9: EMG signals of for Load 5 (60 N) and inclined

angle of 50º.

4 CONCLUSION

A left-right-asymmetric rehabilitation machine and a

measuring system were designed and used in this

study. Some preliminary tests showed that this

machine was suitable for people carrying out

exercise for their walking muscles, and was able to

answer the needs for people with lower-limb

asymmetry.

On the other hand, preliminary tests also showed

that, while the stroke was long enough for exercise,

the maximum pedaling load (1) was too large.

Experiments showed that setting the maximum

0.0002

0.0004

0.0006

0.0008

0.0010

0.0012

50 100 150 200

Amplitude spectrum

(Quadriceps femoris)

[Hz]

0.0001

0.0002

0.0003

0.0004

0.0005

0.0006

0 50 100 150 200

[Hz]

Amplitude spectrum

(Biceps femoris)

Amplitude spectrum

(Tibialis anterior)

0.0020

0.0015

0.0010

0.0005

0 50 100 150 200 [Hz]

0.0005

0.0015

0 50 100 150 200

Amplitude spectrum

(Soleus)

[Hz]

0.0010

Biceps femoris

Soleus

Tibialis anterior

Quadriceps femoris

ICINCO 2017 - 14th International Conference on Informatics in Control, Automation and Robotics

656

pedaling load to be 200 N would be large enough for

rehabilitation. The part of load generation is planned

to be rebuilt in the near future.

To further verify the practicability of the

asymmetric pealing machine, we plan to test various

normal subjects for the comparison of individual

differences (males and females). Then we will

analysis the collected data and try to find a way to

carry out the control of medical rehabilitation. The

performance indexes used in (Smak et al., 1999;

Carpes et al., 2010) will be integrated to evaluate the

left-right-asymmetry and the effectiveness of

pedaling for rehabilitation.

ACKNOWLEDGEMENTS

The authors would like to thank Dr. Wangyong He

and Mr. Qi Shi for their contribution in this study.

This work was supported by Japan Society for the

Promotion of Science (JSPS) KAKENHI under

Grants 26350673 and 16H02883, by the National

Natural Science Foundation of China under Grants

61473313 and 61210011, by the Hubei Provincial

Natural Science Foundation of China under Grant

2015CFA010, and by the 111 Project, China under

Grant B17040.

REFERENCES

S. Anzai, 2014. Healthy Cycle: A Leg-Powered

Wheelchair from Japan Is Motivating the Disabled,

Highlighting Japan, no. 11, pp. 20-21. Available:

http://dwl.gov-online.go.jp/video/cao/dl/public_html/

gov/pdf/hlj/20141101/20-21.pdf

F. P. Carpes, C. B. Mota, and I. E. Faria, 2010. On the

bilateral asymmetry during running and cycling –--A

review considering leg preference, Physical Therapy

in Sport, no. 11, pp. 136-142.

C. J. De Luca, 2002. Surface Electromyography:

Detection and Recording, DelSys Incorporated,

Available: https://www.delsys.com/Attachments_pdf/

WP_SEMGintro.pdf

W. Marras, 1992. Selected Topics in Surface

Electromyography for Use in The Occupational

Setting: Expert Perspectives, U.S. Department of

Health and Human Services Public Health Service

Centers for Disease Control National Institute for

Occupational Safety and Health.

Ministry of Health, Labour and Welfare, Japan, 2016.

Annual Report on the Aging Society, Available:

http://www8.cao.go.jp/kourei/whitepaper/w-2015/

html/zenbun/s1_2_3.html (in Japanese), http://

www8.cao.go.jp/kourei/english/annualreport/2014/201

4pdf_e.html (in English)

M. Sato, 1994. Ningen Kougaku Kijun Suuchi Suushiki

Benran (Handbook of ergonomic standards, numerical

values, and formulas), Tokyo: Gihodo Shuppan Co.,

Ltd.

J. She, Y. Ohyama, and H. Kobayashi, 2006. Master-Slave

Electric Cart Control System for

Maintaining/Improving Physical Strength, IEEE Trans.

Robot., vol. 22, no. 3, pp. 481-490.

J. She, F. Wu, T. Mita, H. Hashimoto, and M. Wu, 2016.

Design of a New Human-Centered Rehabilitation Ma-

chine, Proc. 9th ACM Int. Conf. Pervasive

Technologies Related to Assistive Environments

(PETRA 2016), doi: 10.1145/2910674.2910684, Corfe,

Greece.

J. She, F. Wu, H. Hashimoto, T. Mita, and M. Wu, 2017a.

Design of a Bilaterally Asymmetric pedaling Machine

and its Measuring System for Medical Rehabilitation,

Proc. 1st Int. Conf. on Human Computer Interaction

Theory and Applications (HUCAPP 2017), Porto,

Portugal.

J. She, F. Wu, T. Mita, H. Hashimoto, M. Wu, and A. M.

Iliyasu, 2017b. Design of a New Lower-Limb

Rehabilitation Machine, Journal of Advanced

Computational Intelligence and Intelligent Informatics,

vol. 21, no. 3, pp. 409-416.

W. Smak, R. R. Neptune, and M. L. Hull, 1999. The

influence of pedaling rate on bilateral asymmetry in

cycling, J. Biomechanics, vol. 32, pp. 899-906.

A Left-Right-Asymmetric Pedaling Machine for Medical Rehabilitation of Lower Limbs

657