Can Simple Electronic Instrumentation Associated with Basic

Training Help Users of Assistive Devices?

Presenting and Verifying the Effectiveness of a Biofeedback Module for an

Instrumented Crutch

Renato Varoto

, Aline Midori Rodrigues Sato

, Carolina Lins

and Alberto Cliquet Jr.

1,2

Department of Electrical Engineering, University of São Paulo (USP), São Carlos, Brazil

Department of Orthopedics and Traumatology, University of Campinas (UNICAMP), Campinas, Brazil

Keywords: Crutches, Biofeedback, Rehabilitation.

Abstract: Crutches are prescribed towards compensating lower limb overload and adding sensory information through

upper limb. However, adequate loads are required to avoid upper limb lesions and further lower limb

injuries. Therefore, this work describes the development and application of a biofeedback module for a

Lofstrand crutch, based on a simple electronic instrumentation. The goal is to train the user to apply proper

load on the crutch. Basic training was performed by healthy subjects before and after static and dynamic

activities. Results showed the feasibility of the device and the effectiveness of the training to reach the

target (load on the crutch of 20% of body weight).

1 INTRODUCTION

The most active form of human mobility is gait,

being characterized by the gait cycle.

Gait cycle

(stride) is the continuous repetitive pattern of

walking or running, including stance (single and

double supports) and swing phases.

It starts when

one foot makes contact with the floor and ends when

the same foot makes contact again

(Agarwal et al.,

2012; Simoneau, 2011; Wall, 2001)

. During the

stance phase,

the foot is in contact with the floor;

and the leg moves freely above the floor in the

swing phase (Baker, 2012).

Assistive devices for mobility are prescribed to

compensate orthopaedics problems such as pain,

joint instability and lower limb overload (Cook and

Hussey, 2002).

In addition to reducing the load on the lower

limbs, the crutches are used towards increasing the

support base, adding sensory information and

allowing acceleration control during the gait (Saad,

2007; Delisa and Gans, 1983). Applied loads less

than 20% of body weight of the user are adequate

for this device (Chen et al., 2001; Melis et al., 1999).

To verify loads on the crutch, Leite and Cliquet

(2002) developed a system based on the

instrumented Lofstrand crutch and user friendly

software to analyse and save the data. The crutch

was instrumented with strain gauges, being

characterized by threshold of 105N. The system was

validated with force plate equipment, and

simultaneous measurements using both systems

present values with correlation of 0.98.

This paper describes the development and

application of a biofeedback module for the

instrumented Lofstrand crutch described previously.

This device based on simplistic electronic

instrumentation and coupled to the crutch sends an

audio signal when the user exerts more than 20% of

body weight on the crutch. The aim is to familiarize

the user with the proper load, thus avoiding upper

limb lesions and further lower limb injuries. To

verify the effectiveness of the biofeedback module,

healthy subjects performed rapid training and

executed pilot trials based on static and dynamic

activities.

2 MATERIALS AND METHODS

This work was done at the Laboratory of

Biocybernetics and Rehabilitation Engineering -

USP and at Laboratory of Biomechanics and

259

Varoto R., Midori Rodrigues Sato A., Lins C. and Cliquet Jr. A..

Can Simple Electronic Instrumentation Associated with Basic Training Help Users of Assistive Devices? - Presenting and Verifying the Effectiveness of

a Biofeedback Module for an Instrumented Crutch.

DOI: 10.5220/0004914502590264

In Proceedings of the International Conference on Biomedical Electronics and Devices (BIODEVICES-2014), pages 259-264

ISBN: 978-989-758-013-0

 2014 SCITEPRESS (Science and Technology Publications, Lda.)

Rehabilitation of the Locomotor System -

UNICAMP. Instrumentation was designed at USP,

and pilot trials performed by healthy subjects were

carried out on both laboratories.

The instrumented Lofstrand crutch has four

strain gauges in Wheatstone bridge configuration,

compensating temperature variation. The voltage

across the centre of the bridge is applied to

instrumentation amplifier, assuring adequate range

of signal and isolation of measurement circuit.

Besides, the instrumentation amplifier presents rail-

to-rail output (range of 4.8V) and makes the

connection between the bridge and the biofeedback

module (Leite and Cliquet, 2002).

2.1 Biofeedback Module

The main components of the biofeedback module

are microcontroller, binary-coded decimal (BCD) to

7-segment decoder, 8-bit monolithic digital-to-

analog converter (DAC), comparator circuit and

non-retriggerable monostable multivibrator (Figure

1).

Figure 1: Block diagram of the biofeedback module,

including the instrumented Lofstrand crutch.

The microcontroller used was PIC16F84

(Microchip Technology Inc., Chandler, AZ, USA)

and it was programmed to determine the value

corresponding to 20% of body weight (N) from body

mass (kg) of the crutch user. Furthermore, through

the calibration equation of the crutch, this value is

converted into a digital electrical signal; and then,

applied to the DAC.

The comparator circuit, which used an

operational amplifier as active component, receives

electrical signals from the crutch instrumentation

amplifier and the DAC. It compares the desired load

exerted on the crutch with the actual load and, if the

load on the crutch is greater than the desired one for

longer than 1s, the multivibrator is activated.

Based on the 555 monolithic timing circuit, the

non-retriggerable monostable multivibrator was

configured to generate an audio signal with duration

of 1s.

The whole electronic circuit, including the

original circuit of the crutch, is powered by two

rechargeable batteries (9V, 150mAh).

2.2 Pilot Trials

Five healthy subjects were recruited to participate in

this study (Table 1). Inclusion criteria were body

mass above 50kg and normal gait pattern. Exclusion

criteria were based on the presence of any upper

extremity musculoskeletal disorders, and not being

able to understand the instructions for the trials.

Subject C had no experience with assistive devices

for ambulation, and others had previous experience

(less than 3 months of use). Informed consent and

Ethical Committee approvals were obtained.

Table 1: Subjects characteristics.

Subjects Gender Age (year) Body mass (kg)

M 23.9 80.1

M 22.7 76.7

F 22.0 68.6

M 23.7 73.9

M 26.1 82.4

For each subject, before initiating the trials, the

body mass was determined using a bathroom scale

equipped with high precision sensor (Accumed

Produtos Médico Hospitalares Ltda., Duque de

Caxias, RJ, Brazil). The crutch was fitted according

to the user height, such that the handle was

approximately at the level of the greater trochanter,

leaving the elbow flexed about 30

(Edelstein, 2013;

Moriana et al., 2013; Laufer, 2003). Thus, the use of

the crutch is not influenced by user height.

Pilot trials were based on two activities (static

and dynamic) acquiring force values on the crutch,

and a period of training using the biofeedback

module. Each activity was repeated 3 times. Left

lower limb injury was simulated by the subjects;

thus, they used the crutch on the right forearm

(contra lateral side) (Melis et al., 1999). For all

trials, subjects were instructed to exert 20% of body

weight on the crutch.

During static activity, the subjects remained

standing, with the feet aligned. The tip of the crutch

was 100mm lateral and 150mm anterior to the right

foot (Edelstein, 2013). This activity lasted 10s, and

marks were put on the floor to help the subject and

standardize the trials (Figure 2).

BIODEVICES2014-InternationalConferenceonBiomedicalElectronicsandDevices

260

Figure 2: Subject during static activity.

The subjects performed a route of 8m in straight

line aided by crutch during dynamic activity. They

were instructed to advance the crutch and the pseudo

injured lower limb (left) and then step forward with

the healthy lower limb (right), with simultaneous

contra-lateral support of heel and crutch (Saad,

2007) (Figure 3). Gait speed was a free choice,

according to the natural pattern of the subjects.

Figure 3: Subject during dynamic activity.

The period of training was based on the use of

the biofeedback module, in order to familiarize the

subject towards applying 20% of body weight on the

crutch. Thus, the body mass data was entered in the

module, and every time the load exceeded the value

of 20% of body weight, the audible signal was

emitted. The subjects were free to use any training

strategy for as long as they felt like, not exceeding 5

minutes for each subject.

After the period of training, static and dynamic

activities were repeated to verify the training

effectiveness.

2.3 Data Processing and Analysis

All acquired data were low pass filtered at 10Hz

(finite impulse response) to smooth the signal.

For static activity, mean and standard deviation

(SD) were calculated for the force values above

105N, and maximum (max) and minimum (min)

forces were determined in relation to 30s of activity

duration. Mean and SD values are related to

accuracy and precision, respectively, i.e., the user's

ability to apply appropriate loads repetitively on the

crutch.

In relation to dynamic activity, the number of

gait cycles (strides) was counted during the whole

route (3x8m). Therefore, the number of gait cycles

corresponds to the number of contacts of the crutch

on the floor. The peak value of forces was

determined for each contact of the crutch and mean

and SD, maximum and minimum of values of peak

above 105N were calculated.

3 RESULTS

Figure 4 shows the biofeedback module and the final

version of the instrumented Lofstrand crutch, whose

mass is 1.1kg.

Figure 4: Biofeedback module and the final version of

instrumented Lofstrand crutch.

On the front panel, the module has a 10-digit

keypad in which the value of body mass of the

crutch user is entered. The module accepts values at

range of 12.0 to 99.9kg. Thus, this value must be

typed with 3 digits, in other words, with a resolution

of 0.1kg. Each digit is shown on the 7-segment

display sequentially.

CanSimpleElectronicInstrumentationAssociatedwithBasicTrainingHelpUsersofAssistiveDevices?-Presentingand

VerifyingtheEffectivenessofaBiofeedbackModuleforanInstrumentedCrutch

261

Table 2: Target forces and forces applied on the crutch during static activity.

Subjects

Before the training

Target [N]

After the training

Min [N] Mean(SD) [N] Max [N] Min [N] Mean(SD) [N] Max [N]

▲ ▲ ▲ 157.0 106.4 121.2(7.6) 139.9

▲ ▲ ▲ 150.3 107.3 125.9(9.8) 146.0

▲ ▲ ▲ 134.5 ▲ 122.4(13.5) 156.3

▲ 117.4(6.5) 135.3 144.8 114.9 132.1(8.3) 153.1

▲ 115.4(8.1) 139.8 161.5 ▲ 127.1(19.9) 188.6

The lateral side presents a pushbutton to reset the

microcontroller (if necessary), the buzzer which

receives the output of the non-retriggerable

monostable multivibrator, and a toggle switch to set

one of two functions of the final version of

instrumented Lofstrand crutch: acquiring signals

corresponding to forces applied to the crutch or

training the user with biofeedback. The second

function is independent of the computer, allowing

the user to train anywhere (outside clinical

environment).

In relation to the static and dynamic activities,

the target forces applied on the crutch for subjects A,

B, C, D and E were 157.0N, 150.3N, 134.5N,

144.8N and 161.5N, respectively.

Before the training, during static activity, two

subjects applied forces above 105N, and even then,

the minimum force was not detected. After the

training, all subjects applied forces that were

detected by the crutch. Table 2 presents the force

values for each subject.

For dynamic activity, table 3 presents the number

of gait cycles and the number of detected contacts of

the crutch on the floor.

Table 3: Gait cycles and Subjects characteristics.

Subjects

Before the training

After the training

Gait

cycles

Detected

contacts

Gait

cycles

Detected

contacts

21 21 21 19

21 15 19 17

24 4 24 24

22 22 21 21

21 21 21 16

Figure 5 shows peak value of forces for each

subject during the dynamic activity, before and after

the training with biofeedback module.

4 DISCUSSION

In relation to the biofeedback module

instrumentation, the use of a microcontroller with an

analog-to-digital converter (ADC) and perform a

comparison with the firmware is a feasible

alternative. However, solid state DAC enables better

adjustment of the parameters coming from the

microcontroller, in this case, the percentage of body

weight of the user. This adjustment, which was done

once, allows to set load limit (based on body weight)

on the crutch through hardware.

The patients that have gone through orthopaedics

surgical procedures are not allowed to put any load

on the operated limb during the first weeks after

surgery, and in the following months they are

required to exert around 20% of body weight on the

operated limb towards bone remodelling due to

piezoelectric effect. Therefore, the use of assistive

devices such as crutches, canes and walkers are

recommended.

Loads from 15% to 50% of body weight can be

applied on crutches (Melis et al., 1999). However, in

relation to dynamic activities, the crutch becomes

unstable when more than 20% of body weight is

applied on the device (considering only one crutch)

(Chen et al., 2001; Melis et al., 1999). Proper loads

avoid upper limbs lesion such as carpal tunnel

syndrome and, at the same time, relief loads on the

hip and on the injured lower limb (Waring and

Werner, 1989; Blount, 1956). Besides, it is

important in the case of lower limb implant of plates

and screws to stabilize bone fracture site during

osteosynthesis in order to avoid the risk of bone

refractures and consequent loosing of the implant.

According to the results, load on the crutch

substantially changed after the training performed

using the biofeedback module. In relation to the

static activity, the load on the crutch increases,

becoming closer to the target; thus, the pseudo

injured limb was preserved without compromising

the upper limb.

Biofeedback training did show improvement on

both accuracy (subjects A, B and E) and precision

(subject E) related to the awareness of the actual

upper limb load.

Pilot trials demonstrated the effectiveness of

training with instrumented Lofstrand crutch and

BIODEVICES2014-InternationalConferenceonBiomedicalElectronicsandDevices

262

biofeedback module for healthy subjects simulating

left lower limb injury. Thus, the application of

training with instrumented crutch becomes feasible

for orthopaedics patients.

5 CONCLUSIONS

Based on simple construction, the biofeedback

module can help subjects to apply more adequate

loads on the crutch through basic training. Such

innovation is a feasible alternative for patients of

outpatient clinic that have gone through

orthopaedics surgical procedures such as implants

towards osteosynthesis.

ACKNOWLEDGEMENTS

We thank the support by grants from São Paulo

Research Foundation (FAPESP) and National

Council for Scientific and Technological

Development (CNPq).

We are grateful to the volunteers, undergraduate

medical students.

REFERENCES

Agarwal, S., Abidin, A. Z., Chattopadhyay, S., Acharya,

U. R., 2012. Engineering interventions to improve

impaired gait: a review. In: Yu, W., Chattopadhyay,

S., Lim, T. C.; Acharya, U. R. Advances in

Therapeutic Engineering. Boca Raton: CRC

PressTaylor & Francis Group, pp. 335-363.

Baker, R., 2012. Clinical gait analysis. In: Winkelstein, B.

A. Orthopaedic Biomechanics. Boca Raton: CRC

Press-Taylor & Francis Group, pp. 419-443.

Blount, W. P., 1956. Don't throw away the cane. The

Journal of Bone and Joint Surgery, 38, pp. 695-708.

Chen, C. L., Chen, H. C., Wong, M. K., Tang, F. T., Chen,

R. S., 2001. Temporal stride and force analysis of

cane-assisted gait in people with hemiplegic stroke.

Archives of Physical Medicine and Rehabilitation, 82,

pp. 43-48.

Cook, A. M., Hussey, Susan, M., 2002. Assistive

technologies: principles and practice. Missouri:

Mosby.

Delisa, J. A., Gans, B. M., 1993. Rehabilitation Medicine -

Principles and Practice. Philadelphia: J. B. Lippincott.

Edelstein, J. N., 2013. Assistive devices for ambulation.

Physical Medicine & Rehabilitation Clinics of North

America, 24, pp. 291-303.

Laufer, Y., 2003. The effect of walking aids on balance

and weight-bearing patterns of patients with

hemiparesis in various stance positions. Physical

Therapy, 83, pp. 112-122.

Leite, F., I. L. 2002. Development of an Electronic Crutch

to Medical Assistance. Master of Science. University

of São Paulo.

Melis, E. H., Torres-Moreno, R., Barbeau, H., Lemaire, E.

D., 1999. Analysis of assisted-gait characteristics in

persons with incomplete spinal cord injury. Spinal

Cord, 37, pp. 430-439.

Moriana, G. C., Roldán, J. R., Rejano, J. J., Martínez, R.

C., Serrano, C. S., 2013. Design and validation of

GCH System 1.0 which measures the weight-bearing

exerted on forearm crutches during aided gait. Gait

Posture, 37, pp. 564-569.

Saad, M., 2007. Meios auxiliares de marcha. In: Greve, J.

M. D. Tratado de Medicina de Reabilitação. São

Paulo: Roca, pp. 330-333.

Simoneau, G. G., 2011. Cinesiologia da marcha. In:

Neumann, D. A. Cinesiologia do Apararelho

Musculoesquelético - Fundamentos para Reabilitação.

Rio de Janeiro: Elsevier, pp. 627-676.

Figure 5: Peak value of forces for the dynamic activity.

CanSimpleElectronicInstrumentationAssociatedwithBasicTrainingHelpUsersofAssistiveDevices?-Presentingand

VerifyingtheEffectivenessofaBiofeedbackModuleforanInstrumentedCrutch

263

Wall, J. C., 2001. Marcha. In: Durward, B. R., Baer, G.

D., Rowe, P. J. Movimento Functional Humano:

Mensuração e Análise. São Paulo: Manole, pp. 93-

105.

Waring, W. P., Werner, R. A., 1989. Clinical management

of carpal tunnel syndrome in patients with long-term

sequelae of poliomyelitis. Journal of Hand Surgery

(American Volume), 14, pp. 865-869.

BIODEVICES2014-InternationalConferenceonBiomedicalElectronicsandDevices

264