Effectiveness of Three-Dimensional Kinematic Biofeedback on the

Performance of Scapula-focused Exercises

Ana Antunes

, Inês Filipe

, Sara Cordeiro

, Joana Rosa

, Filomena Carnide

and Ricardo Matias

1,2,3

Neuromechanics of Human Movement Research,

Group of the Interdisciplinary Centre for the Study of Human Performance, University of Lisbon, Lisbon, Portugal

School of Healthcare, Setúbal Polytechnic Institute, Campus do IPS, Estefanilha - Edifício da ESCE, Setúbal, Portugal

Pattern and Image Analysis Group of the Instituto de Telecomunicações, Instituto Superior Técnico,

University of Lisbon, Lisbon, Portugal

Keywords: Biofeedback, Scapulothoracic Stability, Motor Relearning, Physiotherapy.

Abstract: Three-dimensional (3D) kinematic biofeedback can help identify scapular movement disorders and assist

the subjects' motor relearning process by facilitating changes in physiological and biomechanical function

through real-time knowledge of performance and result during or immediately after a task execution. This

study assessed the effectiveness of 3D kinematic biofeedback on the quality of the scapula-focused

exercises execution, and motor learning transfer during shoulder flexion and a daily activity. Thirty healthy

adults with no history of shoulder pain or dysfunction were randomly distributed into two groups. Skin-

mounted sensors allowed tracking of the thorax, scapula and humerus, and scapulothoracic and

glenohumeral 3D angles were computed after reconstructing upper-extremity motions during daily activities

and exercises for different phases of a motor relearning process. The results of this study demonstrate that

the execution quality of scapula-focused exercises benefits of real-time 3D kinematic biofeedback and that

transfer of learning occurs with a specific motor training intervention.

1 INTRODUCTION

Shoulder pain and dysfunction are among the most

frequent problems of patients with mechanical

musculoskeletal disorders seeking health

professionals (e.g. Physical therapists), which

usually affect functional ability and life quality,

resulting in a significant economic impact (Cunha-

Miranda et al., 2010).

Scapular dysfunction (or dyskinesis) seems to be a

common denominator across the most prevalent

shoulder dysfunctions. Although a consistent body

of literature addresses how shoulder impingement

symptoms are affected by scapular dysfunction, the

role of the latter is not clearly defined in creating or

exacerbating shoulder dysfunctions (Kibler et al.,

2013; Struyf et al., 2013). Potential biomechanical

contributors have been divided into two main

groups: musculoskeletal alterations at rest (postural)

and movement alterations (dynamic); such as pain,

soft tissue tightness, muscle dynamics, muscle

strength imbalances and fatigue, and thoracic

posture (Michener et al., 2003).

Some emerging evidence suggests that a scapula-

-focused treatment is effective to restore an accurate

scapular motion and stability, considered essential

for a normal shoulder function (Hanratty et al.,

2012; Kibler et al., 2013; Struyf et al., 2013). The

conclusions on the effectiveness of exercise in the

treatment of people with shoulder dysfunctions are

challenged by the heterogeneity of the exercise

interventions. Still, it has been extensively accepted

the assumption that these patients need to go through

a motor relearning process such as the proposed by

Fitts and Posner (1967). Real-time biofeedback has

been used to enhance the learning ability of an

individual (Holtermann et al., 2008). 3D kinematic

biofeedback is a valid method that can reliably

identify scapular movement disorders (Tate et al.,

2009) and can facilitate the physiological and

biomechanical function through the reception of

feedback information in real-time during or

immediately after a task (Vedsted et al., 2011).

173

Antunes A., Filipe I., Cordeiro S., Rosa J., Carnide F. and Matias R..

Effectiveness of Three-Dimensional Kinematic Biofeedback on the Performance of Scapula-focused Exercises.

DOI: 10.5220/0004928701730178

In Proceedings of the International Conference on Physiological Computing Systems (PhyCS-2014), pages 173-178

ISBN: 978-989-758-006-2

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

The biofeedback is effective in multiple contexts,

showing satisfactory results by obtaining maximum

performance (Egner and Gruzelier, 2003; Huang et

al., 2013; Markovska-Simoska et al., 2008; Pop-

Jordanova and Cakalaroska, 2008; Tsao and Hodges,

2007). Scapula motion analysis contributes to the

understanding of the shoulder dysfunction and has

been considered very suitable to the daily clinical

context (Tate et al., 2009).

The aim of this study was to assess the

effectiveness of 3D kinematic biofeedback during

scapula-focused exercises on: (i) motor learning

transfer during shoulder flexion task and one daily

activity (simulating drinking a glass of water); (ii)

the quality of the exercise execution during a

cognitive phase and (iii) associative phase of a

shoulder motor relearning process.

2 MATERIALS AND METHODS

2.1 Participants

A non-probabilistic sample of 30 participants (10

male and 20 female), 26 right-handed and 4 left-

handed. Subjects were randomly and equally

distributed into two groups: control (CG) and

experimental (EG) groups. The mean age was 21.57

± 4.14 years, with a mean weight and height of 63 ±

10.37 kg and 1.68 ± 0.08 m, respectively.

Participants were selected following specific criteria

(checklist), applied by physical therapists properly

instructed and aware of the study purposes. Healthy

young participants were included with no history of

pain or dysfunction of the shoulder. The exclusion

criteria were: to be aged over 60 years; to have signs

of complete rupture of the rotator cuff tendon or

acute inflammation; to have been submitted to a

physical therapy or any other treatment during the

study and/or in the past 12 months that could have

effect on dependent variables; to perform regular

sport activity in the last six months (at least three

times per week); to have diagnosis of cervical

radiculopathy or neurological changes, visceral or

systemic pain, positive Thoracic Outlet syndrome,

any rheumatic diseases, history of shoulder, neck or

spine high surgery and history of dislocation,

subluxation or shoulder fractures.

The ethics committee of the School of Healthcare,

Setúbal Polytechnic Institute, approved the study

and all participants gave their written informed

consent.

2.2 Instrumentation

Bony segment landmarks 3D coordinates were

collected using an electromagnetic system

trackSTAR (Ascension Technology, Burlington,

Vermont) and “The MotionMonitor” software

(Innovative Sports Training, Chicago, Illinois). This

allowed synchronising the tracking of four sensors

with a sampling rate of 100 Hz per sensor. Static

accuracy has been reported at 1.8 mm and 0.5°

(Milne et al., 1996).

2.3 Procedures

All research procedures were conducted by two

experienced and trained physical therapists. After

reading a letter explaining the study procedures and

goals, data collection began with a questionnaire of

the subject characteristics, including age, gender,

height, weight and dominant side. The participants

were randomly distributed in CG and EG. The figure

1 summarises the study procedure.

Figure 1: Study flow chart.

TrackSTAR sensor 1 was attached on a transparent

acrylic – stylus - and three electromagnetic sensors

were attached over the skin: of the spinous process

of the first thoracic vertebra (sensor 2); of the flat

surface on the superior acromion (sensor 3) and on

the lateral side of the humerus (sensor 4), using

double-sided tape. The humeral sensor fixation was

30 participants

Control Group

n= 15

Experimental Group

n=15

Randomiza

ion

Shoulder flexion up to 45°

Mimic drinking a glass of water

Baseline for transfer assessment

Scapula relocation towards ST-NP

Cognitive Phase exercises

Shoulder flexion up to 45°

Mimic drinking a glass of water

Transfer assessment

Scapula relocation towards ST-NP + Flexion

Associative Phase exercises

Outcome Data

BP distance, ST-NP distance and Execution time

Statistical Methods

Intra- and Inter-group outcomes analysis

PhyCS2014-InternationalConferenceonPhysiologicalComputingSystems

174

reinforced with a velcro strap. Tape was also placed

strategically over the sensor cables to reduce any

motion artefacts.

The protocol of 3D kinematics data collection of the

thorax, scapula and humerus followed previously

published procedures (Matias and Pascoal, 2006)

and the International Society Biomechanics

recommendations for reporting upper extremity

joints motion (Wu et al., 2005).

Scapulothoracic neutral position (ST-NP) was

defined according to Mottram (1997) and scapula's

Euler angles (internal rotation, upward rotation and

posterior tilt) recorded.

ST-NP was set out for each subject, providing visual

and tactile feedback. This position was assumed as

the ST target position during the exercises.

Each subject performed the following sequence of

movements:

a) Shoulder flexion up to 45°;

b) Mimic drinking a glass of water;

c) From the initial (postural) position relocate the

scapula towards the ST-NP;

d) From the initial (postural) position relocate the

scapula towards the ST-NP and while holding

this position preform shoulder elevation in the

scapular plane;

e) Repeat a);

f) Repeat b).

This exercise sequence was intended to study the

effectiveness of 3D kinematic biofeedback on the

performance of scapula-focused exercises in a

cognitive (c) and associative (d) phases of motor

relearning and if any immediate transfer occurs to an

control shoulder flexion task (e) and to a daily

activity task (f).

In each movement, a set of trigger signals

synchronised with the kinematic data were used to

identify (i) the beginning of data collection, (ii)

when the subject verbalised that it reached the ST-

NP and (iii) after 3 seconds from the latter.

In the experimental group, participants were asked

to focus on the data show to access the biofeedback

information.

This process happens in real-time, where the patient

tries to move the yellow cross (that responds

instantaneously to the shoulder movements,

represented in Figure 2) into a static square (Figure

2), which appears always in the same position,

according to the number of variables (coordinates)

relevant to the study.

Figure 2: Example of Biofeedback display.

In the control group (group 1) participants have not

access to the biofeedback system information. For

both cases, the exercise ended when the participants

reached the goal (i.e. when the ST reached the NZ)

or after 15 seconds even if the participant could not

achieved the objective.

Five repetitions for each exercise were performed,

with an established interval of two minutes rest

between them, in order to prevent muscle fatigue

(Wilmore et al., 2008). The same physiotherapist

provided a verbal instruction for each repetition.

The effectiveness of the exercise execution was

determined by the following parameters:

-The distance to the BP: computed as the mean of

the root mean square distances of scapula orientation

values in each time frame to the BP;

-The distance to the ST-NP: transformed by

the mean of the root mean square distances of the

scapula orientation values, also in each time frame to

the ST-NP.

-Execution time: time spends from the first to the

second trigger.

2.4 Statistical Methods

All data were analysed using the MATLAB (version

2012a) and IBM-SPSS (version 21.0).

A visual pre-screening of the records was made, to

rule out abnormal data acquisition sessions for a

Gaussian distribution and additionally tested with

the Shapiro-Wilk test. Appropriate descriptive

statistics were calculated for each variable (mean,

standard deviation and mode).

Differences between groups were performed through

Student’s t or Mann Whitney tests (when normal

distribution was not verified). The within-groups

association analysis was calculated with paired t-

test’s or Wilcoxon’s test (for non-normal

distribution of variables). The level of significance

used in this study was set for p < 0.05 (two-tailed).

EffectivenessofThree-DimensionalKinematicBiofeedbackonthePerformanceofScapula-focusedExercises

175

3 RESULTS

It was found statistical significant differences within

the experimental group on the Euclidean distance to

the best path results (z= -2.22; p= 0.027), with an

execution quality increased by 1.25º when compared

to the pre-exercise condition (table 1).

Table 1: Descriptive statistics (mean±sd) and associations

test results between pre and pos- exercise conditions

(FLEX- flexion; ADL-activity daily living).

n=30

Pre

exercise

Pos

exercise

Distance to the

CG 8.41±3. 48 9.21±3.33

EG 9.16±3.04

8.22±2.43

Distance to the

ST-NP mode

CG 8.07±4.76 7.79±4.37

EG 9.73±6.54 9.33±7.06

Distance to the

CG 5.11±1.27 4.79±1.40

EG 4.73±1.52 4.89±1.68

Distance to the

ST-NP mode

CG 8.41±4.52 8.16±6.20

EG 9.49±7.16 9.56±7.33

*p<0.05

No statistical significant differences were found in

any of the measured variables related to the

cognitive phase exercises. On the contrary, for the

associative phase exercises, statistical significant

differences were found between groups in Euclidean

distance to the best path results (t= 3.91; p= 0.001),

and the mean Euclidean distance to the ST neutral

position (u= 58.0; p= 0.014) (table 2).

Table 2: Descriptive statistics (mean±sd) and associations

tests results between control and experimental groups.

n=30 CG EG

COGNITIVE

Distance to the BP 1.56±0.81 1.70±0.86

Distance to the ST-NP 3.44±2.00 2.95±1.88

Time 3.25±1.25 3.93±1.25

ASSOCIATIVE

Distance to the BP 4.19±1.17 2.68±0.89*

Distance to the ST-NP 5.00±2.00 3.63±1.63*

Time 2.81±0.66 2.62±1.00

*p<0.05

4 DISCUSSION

In a motor relearning process the variability

of the practice consists in repeating variations of the

same task in which the parameters are changed,

providing variations around the same skill. This

study rely its intervention model in the extensive

known, considering three stages motor relearning

process that consists on the sequential cognitive,

associative and autonomous phases (Fitts and

Posner, 1967). The duration of the first phase is

limited from a few minutes to a few days, while the

second learning phase can last for weeks or even

months (Schmidt, 2003; Sherwood and Lee, 2003;

Summers and Anson, 2009) . According to this

model for patients to improve their motor

performance the exercise intervention must go

through scapula-focused exercises aiming the

awareness of the scapulothoracic neutral zone and

normalisation of the scapulohumeral rhythm, by

facilitating the central nervous system's ability to

efficiently control the inter-segmental motion of the

upper limb (Hess, 2000; Cools et al., 2003; Cowan

et al., 2003).

In this study the exercise exposure period was the

same of an average time of a physiotherapy

session at an early stage of motor relearning process

with emphasis on the cognitive and associative

stages.

The results of this study demonstrate that real-time

3D kinematic biofeedback is an effective solution to

correctly perform scapula-focused exercises during

the progressive phases of learning a new skill,

notably when the complexity of the task increases

(Jones and French, 2007). The latter become

particularly relevant to the associative phase

where the quality of the movement becomes more

important than the amount of practice itself

(Schmidt, 2003; Sherwood and Lee, 2003; Summers

and Anson, 2009). The results also provide

preliminary evidence that after one physiotherapy

session, immediate transfer of learning (Issurin,

2013; Maslovat et al., 2009) occurs when

performing a similar task, with an increase in the

execution quality. Given the obtained results, we

believe that successful learning can be more

expressive if we raise the time and volume of

practice.

These preliminary findings corroborate other

published results on the immediate effect of a

specific motor training intervention (Tsao and

Hodges 2007; Sturmberg et al. 2013).

Future studies are needed to address the effect of

exercises volume and specificity on the execution

performance quality, and its effects on the learning

retention and transfer. In addition, it would be

relevant to determine whether the magnitude of

PhyCS2014-InternationalConferenceonPhysiologicalComputingSystems

176

these changes are clinical meaningful, particularly in

patients' functionality.

5 CONCLUSIONS

The real-time 3D kinematic biofeedback, proved to

be an effective tool for improving the quality of the

exercises’ execution. Based on this study results

subjects who had access to kinematic biofeedback

improved 3D motion control of the ST during the

analyzed tasks, corroborating results from previous

studies. Such a tool can help subjects achieve

rehabilitation motor (re)learning goals, and improve

rehabilitation decision-making process by

quantifying human movement performance.

ACNOWLEDGEMENTS

This study was a non-profit project and had no

financial support. The authors would like to

acknowledge all the participants that kindly accepted

to participate in this study.

REFERENCES

Cools, A. M., Witvrouw, E. E., Declercq, G.A., Danneels,

L. A., Cambier, D. C., 2003. Scapular muscle

recruitment patterns: trapezius muscle latency with

and without impingement symptoms. Am. J. Sports

Med. 31, 542–549.

Cowan, S. M., Bennell, K.L., Hodges, P.W., Crossley, K.

M., McConnell, J., 2003. Simultaneous feedforward

recruitment of the vasti in untrained postural tasks can

be restored by physical therapy. J. Orthop. Res. Off.

Publ. Orthop. Res. Soc. 21, 553–558.

Cunha-Miranda, L., Carnide, F., Lopes, M.F., 2010.

Prevalence of rheumatic occupational diseases -

PROUD study. Acta Reum. Port. 35, 215–226.

Egner, T., Gruzelier, J. H., 2003. Ecological validity of

neurofeedback: modulation of slow wave EEG

enhances musical performance. Neuroreport 14,

1221–1224.

Fitts, P.M., Posner, M.I., 1967. Human performance.

Brooks/Cole Pub. Co.

Hanratty, C.E., McVeigh, J.G., Kerr, D.P., Basford, J.R.,

Finch, M.B., Pendleton, A., Sim, J., 2012. The

effectiveness of physiotherapy exercises in

subacromial impingement syndrome: a systematic

review and meta-analysis. Semin. Arthritis Rheum.

42, 297–316.

Hess, S.A., 2000. Functional stability of the glenohumeral

joint. Man. Ther. 5, 63–71.

Holtermann, A., Søgaard, K., Christensen, H., Dahl, B.,

Blangsted, A.K., 2008. The influence of biofeedback

training on trapezius activity and rest during

occupational computer work: a randomized controlled

trial. Eur. J. Appl. Physiol. 104, 983–989.

Huang, H.-Y., Lin, J.-J., Guo, Y. L., Wang, W.T.-J., Chen,

Y.-J., 2013. EMG biofeedback effectiveness to alter

muscle activity pattern and scapular kinematics in

subjects with and without shoulder impingement. J.

Electromyogr. Kinesiol. Off. J. Int. Soc.

Electrophysiol. Kinesiol. 23, 267–274.

Issurin, V. B., 2013. Training transfer: scientific back-

ground and insights for practical application. Sports

Med. Auckl. NZ 43, 675–694.

Jones, L. L., French, K. E., 2007. Effects of contextual

interference on acquisition and retention of three

volleyball skills. Percept. Mot. Skills 105, 883–890.

Kibler, W. B., Ludewig, P. M., McClure, P.W., Michener,

L. A., Bak, K., Sciascia, A .D., Bak, K., Ebaugh, D.,

Kibler, W.B., Ludewig, P., Kuhn, J., McClure, P.,

Mazzocca, A., Michener, L., Bailey, L., Sciascia, A.

D., Borstad, J., Seitz, A., Cools, A., Uhl, T., Cote, M.,

2013. Clinical implications of scapular dyskinesis in

shoulder injury: the 2013 consensus statement from

the “scapular summit”. Br. J. Sports Med. 47, 877–

885.

Markovska-Simoska, S., Pop-Jordanova, N., Georgiev, D.,

2008. Simultaneous EEG and EMG biofeedback for

peak performance in musicians. Pril. Makedon. Akad.

Na Nauk. Umet. Oddelenie Za Biološki Med. Nauki

Contrib. Maced. Acad. Sci. Arts Sect. Biol. Med. Sci.

29, 239–252.

Maslovat, D., Brunke, K. M., Chua, R., Franks, I.M.,

2009. Feedback effects on learning a novel bimanual

coordination pattern: support for the guidance

hypothesis. J. Mot. Behav. 41, 45–54.

Matias, R., Pascoal, A.G., 2006. The unstable shoulder in

arm elevation: a three-dimensional and

electromyographic study in subjects with

glenohumeral instability. Clin. Biomech. Bristol Avon

21 Suppl 1, S52–58.

Michener, L.A., McClure, P.W., Karduna, A.R., 2003.

Anatomical and biomechanical mechanisms of

subacromial impingement syndrome. Clin. Biomech.

Bristol Avon 18, 369–379.

Milne, A.D., Chess, D.G., Johnson, J.A., King, G.J., 1996.

Accuracy of an electromagnetic tracking device: a

study of the optimal range and metal interference. J.

Biomech. 29, 791–793.

Mottram, S .L., 1997. Dynamic stability of the scapula.

Man. Ther. 2, 123–131.

Pop-Jordanova, N., Chakalaroska, I., 2008. Comparison of

Biofeedback Modalities for Better Achievement in

High School Students.

Schmidt, R.A., 2003. Motor schema theory after 27 years:

reflections and implications for a new theory. Res. Q.

Exerc. Sport 74, 366–375.

Sherwood, D.E., Lee, T.D., 2003. Schema theory: critical

review and implications for the role of cognition in a

new theory of motor learning. Res. Q. Exerc. Sport 74,

EffectivenessofThree-DimensionalKinematicBiofeedbackonthePerformanceofScapula-focusedExercises

177

376–382.

Struyf, F., Nijs, J., Mollekens, S., Jeurissen, I., Truijen, S.,

Mottram, S., Meeusen, R., 2013. Scapular-focused

treatment in patients with shoulder impingement

syndrome: a randomized clinical trial. Clin.

Rheumatol. 32, 73–85.

Sturmberg, C., Marquez, J., Heneghan, N., Snodgrass, S.,

van Vliet, P., 2013. Attentional focus of feedback and

instructions in the treatment of musculoskeletal

dysfunction: A systematic review. Man. Ther.

Summers, J.J., Anson, J.G., 2009. Current status of the

motor program: Revisited. Hum. Mov. Sci. 28, 566–

577.

Tate, A.R., McClure, P., Kareha, S., Irwin, D., Barbe,

M.F., 2009. A clinical method for identifying scapular

dyskinesis, part 2: validity. J. Athl. Train. 44, 165–

173.

Tsao, H., Hodges, P.W., 2007. Immediate changes in

feedforward postural adjustments following voluntary

motor training. Exp. Brain Res. Exp. Hirnforsch.

Expérimentation Cérébrale. 181, 537–546.

Tsao, H., Hodges, P.W., 2007. Immediate changes in

feedforward postural adjustments following voluntary

motor training. Exp. Brain Res. Exp. Hirnforsch.

Expérimentation Cérébrale 181, 537–546.

Vedsted, P., Søgaard, K., Blangsted, A.K., Madeleine, P.,

Sjøgaard, G., 2011. Biofeedback effectiveness to

reduce upper limb muscle activity during computer

work is muscle specific and time pressure dependent.

J. Electromyogr. Kinesiol. Off. J. Int. Soc.

Electrophysiol. Kinesiol. 21, 49–58.

Wilmore, J.H., Costill, D.L., Kenney, W.L., 2008.

Physiology of Sport and Exercise. Human Kinetics.

Wu, G., van der Helm, F.C.T., Veeger, H.E.J.D.,

Makhsous, M., Van Roy, P., Anglin, C., Nagels, J.,

Karduna, A.R., McQuade, K., Wang, X., Werner,

F.W., Buchholz, B., International Society of

Biomechanics, 2005. ISB recommendation on

definitions of joint coordinate systems of various

joints for the reporting of human joint motion--Part II:

shoulder, elbow, wrist and hand. J. Biomech. 38, 981–

992.

PhyCS2014-InternationalConferenceonPhysiologicalComputingSystems

178