Marinsek M., Slana U. SMART PHONE AS A STANDING BALANCE ASSESMENT DEVICE
Vol. 6 Issue 3: 39 - 47
SMART PHONE AS A STANDING BALANCE ASSESSMENT
DEVICE
Miha Marinsek1, Uros Slana2
1 Faculty of Education, University of Maribor, Slovenia 2 Faculty of Sports, University of Ljubljana, Slovenia
_Original article
Abstract
Balance plays important role in postural control and force production in artistic gymnastics and can be used as injury prevention. Therefor balance should be monitored during the training process. Many different protocols have been used to assess balance. Recently there were proposals made to include new technologies such as smart phones in various assessment protocols. In present research, we compared a balance assessment protocol on a T shaped tilt board with smart phone application and G-weight goniometer balance assessment protocol. Thirty different positions in x-axis and thirty different positions in y-axis were used as a criteria acquired with two smart phones and one G-weight goniometer. We acquired also one real time measurement for 10 seconds on the T shaped tilt board. We found out a very strong and positive correlation between smart phone application and G-weight goniometer. Cronbach's alpha showed very high reliability of the smart phone measures. The limits of agreement showed that the measurement with the smart phone could be 0.85° below or 3.23° above in x-axis and 0.09° below or 2.96° above in y-axis the goniometer. Both smart phones share a very similar displacement-time curve. To sum up, the smart phones with its measurement characteristics are reliable and valid enough for monitoring balance on T shaped tilt board for practical use but are not precise enough for the research use.
Keywords: Artistic Gymnastics, Evaluation, Panel Judging, Bias.
INTRODUCTION
Humans are from balance point of view quiet unstable system with multiple segments to control over a relatively small base of support. Balance can be defined as the ability to maintain the body's centre of gravity (COG) within the base of support that involves the use of sensory information and its integration with muscle contractions (Kirby, Price & MacLeod, 1987).
Artistic gymnastics (AG) is a sport that requires a great sense of balance. It
plays important role in postural control and force production while performing difficult gymnastic elements and landings. Many factors influence gymnastics postural control and force production performance, which include sensory information, motor responses and attention demands (Horak, 1987, Vuillerme & Nougier, 2004). The gymnasts are mostly exposed to above-mentioned factors when performing difficult gymnastics skills and/or at competitions. Difficult gymnastics skills
Science of Gymnastics Journal
39
Science of Gymnastics Journal
Marinsek M., Slana U. SMART PHONE AS A STANDING BALANCE ASSESMENT DEVICE
Vol. 6 Issue 3: 39 - 47
demand high level of sensory utilization for good postural orientation, high level of coordination, strength and joint range of motion in order to produce appropriate force direction and magnitude. Gymnasts have to focus their attention on the skill execution and not let other environmental factors distract them.
The aesthetic note of the AG demands from the athletes balanced and controlled execution. Landings are present in every gymnastics discipline and are being assessed by judges not only in artistic gymnastics, but also in acrobatic gymnastics, trampoline, rhythmic gymnastics, aerobic gymnastics and gymnastics for all. Landings in artistic gymnastics are parts of gymnastics routines where most of mistakes happen because of lack of balance (Marinšek, 2009, 2010). Somersault twists are mostly the reason, which causes landing asymmetries and thus lack of balance (Marinšek & Čuk, 2013). In most of gymnastics disciplines, static as well as dynamic balance is demanded. Motionless balance skills, holds on reduced base of support, such as various static holds in pyramids (Figure 1) and other leaping and tumbling skills are part of every gymnastics disciplines.
Figure 1: Stand in hand or on shoulder (Acrobatic Gymnastics Code of Points 2013-2016, FIG, 2013)
Balance training can improve postural control (Heitkamp, Horstmann, Mayer, Weller & Dickhuth, 2001, Granacher, Gollhofer & Strass 2006, Myer, Ford,
Brent & Hewett, 2006, Yaggie & Campbell 2006, Beck et al., 2007, Gruber et al. 2007a, Taube et al., 2007a), has also a great impact on strength and jumping abilities (Bruhn, Kullmann & Gollhofer, 2004, Gruber & Gollhofer 2004, Kean, Behm & Young, 2006, Gruber et al., 2007b, Taube et al., 2007b, Tsopani et al., 2014) and can be used as injury prevention in elite sport.
Balance can be developed with different training devices such as tilt boards, half discs, soft mats (Heitkamp et al., 2001, Gruber & Gollhofer 2004, Taube, Gruber & Gollhofer, 2008), and sport specific movements on all gymnastics apparatus (Bressel, Yonker, Kras & Heath, 2007).
Numerous assessment protocols have been developed to provide important information on balance progress. Various protocols on tilt boards and force plates have been used as well as other tests such as Star Excursion Balance Test (Filipa, Byrnes, Paterno, Myer & Hewett, 2010) and Y Balance Test (Plisky et al., 2009).
Previous research indicates attempts to use other popular technologies for postural control and force production assessment such as Wii Balance Board (Clark et al., 2010), EquiTest t computerized dynamic posturography system (Tsopani et al., 2014), accelerometer-embedded springboard (Čuk, Penič & Križaj, 2011). These technologies can come into wide use because of their portability, inexpensiveness and availability. One of the most popular and widespread electronic devices nowadays which is equipped with a large set of embedded sensors is a smart phone.
Increasing number of smart phones enhances the number of applications used in smart phones. Recently, the application for the balance test has been developed. The application measures angle displacements on a T shaped tilt board in frontal and sagittal plane. The data can be
Science of Gymnastics Journal
40
Science of Gymnastics Journal
Marinsek M., Slana U. SMART PHONE AS A STANDING BALANCE ASSESMENT DEVICE
Vol. 6 Issue 3: 39 - 47
exported to the excel worksheet and used for further analysis.
The aim of this study was to compare measurement reliability between a G-weight goniometer and the balance test application on two different smartphones for various positions on a T shaped tilt board. Our hypothesis was that a smart phone measurement would provide reliable output in comparison to the G-weight goniometer measurements regardless of the smart phone model used.
METHODS
Apparatus
Two smart phones Huawei Ascend P6 (OS: Android OS, v4.2.2 CPU: Quad-core 1.5 GHz) and Samsung I9100 Galaxy S II (OS: Android OS, v2.3.4 CPU: Dual-core 1.2 GHz Cortex-A9) with balance test application and G-weight goniometer (Winkelmesser BMI) were used.
For determining the orientation of the device, one can use the rotational vector sensor, which is either of software or hardware type. Usually the rotation vector sensor fuses data from three different sensors: accelerometer, gyro and magnetic field sensor. Additional rotation vector sensors can be available on the Android device. By default, our program chooses the software sensor, whose vendor is Google Inc. and its version is 3. This sensor is part of The Android Open Source Project. If this sensor is not present then the default rotation vector sensor is chosen. Based on the data provided by the sensor we calculate the rotation matrix from which we derive the orientation of the device. When the device is lying on the horizontal plane with screen up the angle in the x- and y-axes is 0°. The application can be downloaded from https://play.google.com/store/apps/details7i d=org.slani.balancetest.
Procedures
Thirty different positions (15 in positive and 15 in negative side) in x-axis and thirty different positions (15 in
positive and 15 in negative side) in y-axis were used as a criteria. We acquired position (in degrees) in x- and y-axis randomly for G-weight goniometer and both smart phones. The acquired data was used to examine the eventual measurement errors of smart phones.
After initial position acquirement, we acquired one real time measurement for 10 seconds on the T shaped tilt board. The tilt board was moved randomly simulating balance assessment protocol and recorded with both smart phones. The acquired data was used for comparison of displacement-time curve in x- and y- axis between both smart phones.
Statistical analysis and data processing
Kolmogorov Smirnov test was used to test the data distribution. The measurement error was evaluated by the relative absolute error (RAE). The data gathered by smart phones was compared to the G-weight goniometer representing the gold standard for measuring range of motion. The t-test for paired data was calculated in order to find out the differences between G-weight goniometer and both smart phones. Additionally Pearson's correlation coefficient (PCC) was used to find out correlations between G-weight goniometer, Huawei, and Samsung smart phones. With respect to internal consistency, Cronbach's alpha was calculated in order to evaluate the reliability of the survey measures. The alpha value of 0.70 to 0.80 are regarded satisfactory for group comparison and 0.90 to 0.95 for individual comparison (Bland and Altman, 1997). As proposed by Bland and Altman (1986) an analysis of 95% limits of agreement (LoA) was performed to compare absolute reliability between the G-weight goniometer and smart phone measurements.
RESULTS
Kolmogorov-Smirnov test showed that distributions of the variables significantly do not deviate from normal
Science of Gymnastics Journal
41
Science of Gymnastics Journal
Marinsek M., Slana U. SMART PHONE AS A STANDING BALANCE ASSESMENT DEVICE
Vol. 6 Issue 3: 39 - 47
distribution (p>0.01, Table 1). The differences in the mean values of the measurements were statistically significant in x- (p<0.001) and y- axis (p<0.001). Samsung smart phone showed smaller difference to the G-weight goniometer (diff=0.753°) in comparison to Huawei smart phone (diff=1.18°) in x-axis. In y-axis the difference to the G-weight goniometer was smaller for the Huawei smart phone (diff=1.043°) than for the Samsung (diff=1.477°) (Table 1).
SEM for goniometer and smart phones ranged from 2.050° to 2.243° in x-axis and from 2.023° to 2.210° in y-axis (Table 1), suggesting consistency of the measurements. Relative absolute error compared to goniometer, which can be considered as gold standard is 0.85° and 1.48° for Samsung in x- and y-axis, and 1.19° and 1.07° for Huawei, respectively (Table 1)
Table 1. Descriptive statistics, distribution and difference between G-weight goniometer and smart phone measurements._
		N	Mean±SD (°)	RAE	SEM	P (K-S)	t	P(t)
Pair 1	xG	30	1.667±11.742	/	2.144	0.200	-5.974	0.000
	xSPsam	30	2.420±12.286	0.85	2.243	0.200		
Pair 2	xSPsam	30	2.420±12.286	0.85	2.243	0.200	7.485	0.000
	xSPhua	30	0.487±11.229	1.19	2.050	0.200		
Pair 3	xG	30	1.667±11.742	/	2.144	0.200	6.165	0.000
	xSPhua	30	0.487±11.229	1.19	2.050	0.200		
Pair 4	yG	30	-0.667±11.513	/	2.102	0.200	10.794	0.000
	ySPsam	30	-2.144±12.107	1.48	2.210	0.200		
Pair 5	ySPsam	30	-2.144±12.107	1.48	2.210	0.200	-2.269	0.031
	ySPhua	30	-1.710±11.078	1.07	2.023	0.200		
Pair 6	yG	30	-0.667±11.513	/	2.102	0.200	8.916	0.000
	ySPhua	30	-1.710±11.078	1.07	2.023	0.200		
Legend: N - numerous, Mean - mean value, SD - standard deviation, RAE - relative absolute error, SEM - standard error mean, P(K-S) - Sig. Kolomogorov Smirnov test, t - t value, P(t) - Sig. paired t test
Table 2. Corelations between G-weight goniometer and smart phone measurements in x- and y- axis.	_
	xSPsam	xSPhua	ySPsam	ySPhua
xG	.999	.997		
yG			.999	.999
Science of Gymnastics Journal
42
Science of Gymnastics Journal
Marinsek M., Slana U. SMART PHONE AS A STANDING BALANCE ASSESMENT DEVICE
Vol. 6 Issue 3: 39 - 47
Figure 2. Bland - Altman plots for the comparison between G-weight goniometer and Samsung smart phone in (A) x-axis and (B) y-axis.
Figure 3. Bland - Altman plots for the comparison between G-weight goniometer and Huawei smart phone in (A) x-axis and (B) y-axis.
Science of Gymnastics Journal
43
Science of Gymnastics Journal
Marinsek M., Slana U. SMART PHONE AS A STANDING BALANCE ASSESMENT DEVICE
Vol. 6 Issue 3: 39 - 47
Figure 4. Comparison of displacement (degrees) in (A) x-axis and in (B) y-axis for Huawei and Samsung smart phones.
All Person's correlations are very high in x-axis (rxG-xSPsam=.999; rxG-xSPhua=.997; rxSPhua-xSPsam=.997) and in y-axis (ryG-ySPsam=.999; ryG-ySPhua=.999; rySPhua-ySPsam=1.000) which is showing a strong positive correlation (Table 2). Despite the fact that results in nominal mean value are different they have the same function when angle of the tilt board is getting higher or lower.
Cronbach's alpha (.999 for x- and y-axis) showed very high reliability of the survey measures. Reliability index for the devices is very high. In order to compare absolute reliability between the G-weight goniometer and smart phone measurements the Bland and Altman plots for the degrees are presented in Figure 2 and 3.
The limits of agreement show that the measurement with the Samsung smart phone can be 0.15° below or 1.95° above the goniometer in x-axis and 0.00° below or 2.96° above in y-axis (Figure 2).
The limits of agreement show that the measurement with the Huawei smart phone can be 0.85° below or 3.23° above the goniometer in x-axis and 0.09° below or 2.23° above in y-axis (Figure 3).
Smart phones have different frequencies of recording. Despite selecting the same frequencies of recording (Hz), we the sampling rate was different. For the Huawei it was sampling rate 933 samples per 10 seconds, while for Samsung 950 samples per 10 seconds.
For Huawei we had to eliminate every 22nd measure to equalize time in y- and x-axis with Samsung smart phone. Figure 4 A and B show very similar displacement-time curve in x- and y- axis.
Correlation between both smart phones in x-axis was 0.94 and 0.97 in y-axis respectively. Thus, smart phones share 88% and 94% of variance in x- and y-axis respectively. This means very good to excellent reliability and validity of measurement.
DISCUSSION
The aim of this study was to find out if the smart phone could be reliable as a balance assessment device. There has been many devices used to assess balance, one of them being T shaped tilt board. The protocol to assess balance on the T shaped tilt board can be measured in time elapsed from the start of the protocol until the touchdown of the tilt board or in range of motion during protocol.
In the present study, we used two different models of smart phones to assess angle displacement on the T shaped tilt board and compare them to the G weight goniometer. After the comparison, we conducted two 10-seconds standing protocols with both smart phones.
In our research, we proved that two smart phone devices had a very strong positive correlation with G-weight
Science of Gymnastics Journal
44
Science of Gymnastics Journal
Marinsek M., Slana U. SMART PHONE AS A STANDING BALANCE ASSESMENT DEVICE
Vol. 6 Issue 3: 39 - 47
goniometer, which is considered a gold standard in range of motion measurements. SEM for goniometer and smart phones showed consistent measurements. Relative absolute error of two smart phones used compared to goniometer was between 0.85° and 1.19° in x-axis and between 1.07° and 1.48° in y-axis, showing slightly higher error in y-axis.
Internal consistency was confirmed by Cronbach's alpha values of 0.999 for x-axis and 0.999 for y-axis.
In measurement comparison between new and established measurement technique limits of agreement should be calculated (Bland & Altman, 1986). Bland - Altman's limits of agreement revealed that the measurement with smart phone could be as much as 0.85° below or 3.23° above in x-axis and 0.09° below or 2.96° above in y-axis the goniometer measurement, depending on the smart phone model. Figure 2 and 3 displayed lack of agreement between goniometer and smart phone that would be unacceptable for clinical purpose, but would be acceptable for practical use in monitoring balance progress of an individual.
Both smart phones share a very similar displacement-time curve. Despite selecting the same frequencies of recording (Hz), we found that the smart phone's operating system dynamically manages its sampling rate. As mentioned in a research by Mellone, Tacconi & Chiari (2012) occasionally the sampling rate intervals undergo large changes because of the concurrent processes with higher priority.
It seems that smart phones with its measurement characteristics are reliable and valid enough for monitoring balance on T shaped tilt board for practical use but are not good enough for research or clinical use.
However, we do not have to overlook advantages of smart phones in assessment procedures. As mentioned in other studies (Nishiguchi et al., 2012; Mellone, Tacconi & Chiari, 2012; Shin, Ro, Lee, Oh & Kim, 2012) smart phone's applications are relatively easily to improve. The data acquired through assessment can be
presented in intelligible graphical way in smart phone's application, transmitted fast, and convenient to another device for statistical use. Because of their cost-efficiency, smart phones are available to broader public and therefore useful for monitoring balance progress during individual physical therapy or training.
CONCLUSION
Smart phone can be a valid and reliable balance assessment device for practical use in monitoring balance progress. Popular technologies such as smart phones can come into wide use as assessment devices because of their portability, inexpensiveness and availability. They come with a large set of embedded sensors, which can be used in many assessment circumstances.
REFERENCES
Beck, S., Taube, W., Gruber, M., Amtage, F., Gollhofer, A. & Schubert, M. (2007). Task-specific changes in motor evoked potentials of lower limb muscles after different training interventions. Brain research, 1179, 51-60.
Bland, J. M. & Altman, D. (1986). Statistical methods for assessing agreement between two methods of clinical measurement. The lancet, 327(8476), 307310.
Bland, J. M. & Altman, D. G. (1997). Statistics notes: Cronbach's alpha. Bmj, 314(7080), 572.
Bressel, E., Yonker, J. C., Kras, J. & Heath, E. M. (2007). Comparison of static and dynamic balance in female collegiate soccer, basketball, and gymnastics athletes. Journal of athletic training, 42(1), 42.
Bruhn, S., Kullmann, N. & Gollhofer, A. (2004). The effects of a sensorimotor training and a strength training on postural stabilisation, maximum isometric contraction and jump performance. International journal of sports medicine, 25(01), 56-60.
Clark, R. A., Bryant, A. L., Pua, Y., McCrory, P., Bennell, K. & Hunt, M.
Science of Gymnastics Journal
45
Science of Gymnastics Journal
Marinsek M., Slana U. SMART PHONE AS A STANDING BALANCE ASSESMENT DEVICE
Vol. 6 Issue 3: 39 - 47
(2010). Validity and reliability of the Nintendo Wii Balance Board for assessment of standing balance. Gait & posture, 31(3), 307-310.
Čuk, I., Penič, S. & Križaj, D. (2011). Towards a smart springboard (case study). Science of Gymnastics Journal, 3(3).
Čuk, I. & Marinšek, M. (2013). Landing quality in artistic gymnastics is related to landing symmetry. Biology of Sport, 30(1), 29-33.
Fédération Internationale de Gymnastique. (2013). Acrobatic Gymnastics Code of Points 2013-2016.
Filipa, A., Byrnes, R., Paterno, M. V., Myer, G. D. & Hewett, T. E. (2010). Neuromuscular training improves performance on the star excursion balance test in young female athletes. Journal of orthopaedic & sports physical therapy, 40(9), 551-558.
Granacher, U., Gollhofer, A. & Strass, D. (2006). Training induced adaptations in characteristics of postural reflexes in elderly men. Gait & posture, 24(4), 459-466.
Gruber, M. & Gollhofer, A. (2004). Impact of sensorimotor training on the rate of force development and neural activation. European journal of applied physiology, 92(1-2), 98-105.
Gruber, M., Gruber, S. B., Taube, W., Schubert, M., Beck, S. C. & Gollhofer, A. (2007a). Differential effects of ballistic versus sensorimotor training on rate of force development and neural activation in humans. The Journal of Strength & Conditioning Research, 21(1), 274-282.
Gruber, M., Taube, W., Gollhofer, A., Beck, S., Amtage, F. & Schubert, M. (2007b). Training-specific adaptations of Hand stretch reflexes in human soleus muscle. Journal of motor behavior, 39(1), 68-78.
Heitkamp, H. C., Horstmann, T., Mayer, F., Weller, J. & Dickhuth, H. H. (2001). Gain in strength and muscular balance after balance training. International Journal of Sports Medicine, 22(4), 285-290.
Horak, F. B. (1987). Clinical measurement of postural control in adults. Physical therapy, 67(12), 1881-1885.
Kirby, R. L., Price, N. A. & MacLeod, D. A. (1987). The influence of foot position on standing balance. Journal of biomechanics, 20(4), 423-427.
Kean, C. O., Behm, D. G. & Young, W. B. (2006). Fixed foot balance training increases rectus femoris activation during landing and jump height in recreationally active women. Journal of sports science & medicine, 5(1), 138.
Marinšek, M. (2009). Landing characteristics in men's floor exercise on European championships 2004. Science of Gymnastics Journal, 1(1), 31-39.
Marinšek, M. (2010). Basic landing characteristics and their application in artistic gymnastics. Science of Gymnastics Journal, 2(2), 59-67.
Marinšek, M. & Čuk, I. (2013). The influence of different twists in the forward and backward somersaults on increased landing asymmetries. Kineziologija, 45(1), 73-81.
Martin Bland, J. & Altman, D. (1986). Statistical methods for assessing agreement between two methods of clinical measurement. The lancet, 327(8476), 307310.
Mellone, S., Tacconi, C. & Chiari, L. (2012). Validity of a Smartphone-based instrumented Timed Up and Go. Gait & posture, 36(1), 163-165.
Myer, G. D., Ford, K. R., Brent, J. L. & Hewett, T. E. (2006). The effects of plyometric vs. dynamic stabilization and balance training on power, balance, and landing force in female athletes. The Journal of Strength & Conditioning Research, 20(2), 345-353.
Nishiguchi, S., Yamada, M., Nagai, K., Mori, S., Kajiwara, Y., Sonoda, T. & Aoyama, T. (2012). Reliability and validity of gait analysis by android-based smartphone. Telemedicine and e-Health, 15(4), 292-296.
Plisky, P.J., Gorman, P. P., Butler, R. J., Kiesel, K. B., Underwood, F. B. & Elkins, B. (2009). The reliability of an instrumented device for measuring components of the Star Excursion Balance
Science of Gymnastics Journal
46
Science of Gymnastics Journal
Marinsek M., Slana U. SMART PHONE AS A STANDING BALANCE ASSESMENT DEVICE
Vol. 6 Issue 3: 39 - 47
Test. North American Journal of Sports Physical Therapy, 4(2), 92-99.
Santos, J. R. A. (1999). Cronbach's alpha: A tool for assessing the reliability of scales. Journal of extension, 37(2), 1-5.
Shin, S. H., Ro, D. H., Lee, O., Oh, J. H. & Kim, S. H. (2012). Within-day reliability of shoulder range of motion measurement with a smartphone. Manual therapy, 17(4), 298-304.
Taube, W., Gruber, M., Beck, S., Faist, M., Gollhofer, A. & Schubert, M. (2007a). Cortical and spinal adaptations induced by balance training: correlation between stance stability and corticospinal activation. Acta Physiologica, 189(4), 347-358.
Taube, W., Kullmann, N., Leukel, C., Kurz, O., Amtage, F. & Gollhofer, A. (2007b). Differential reflex adaptations following sensorimotor and strength training in young elite athletes. International journal of sports medicine, 28(12), 999-1005.
Taube, W., Gruber, M. & Gollhofer, A. (2008). Spinal and supraspinal adaptations associated with balance training and their functional relevance. Acta Physiologica, 193(2), 101-116.
Tavakol, M. & Dennick, R. (2011). Making sense of Cronbach's alpha. International journal of medical education, 2, 53-55.
Tsopani D., Dallas, G., Tsiganos, G., Papouliakos, S., Di Cagno, A., Korres, G., Riga, M. & Korres, S. (2014). Short-term effect of whole-body vibration training on balance, flexibility and lower limb explosive strength in elite rhythmic gymnasts. Human movement science, 33, 149-158.
Vuillerme, N. & Nougier, V. (2004). Attentional demand for regulating postural sway: the effect of expertise in gymnastics. Brain Research Bulletin, 63(2), 161-165.
Yaggie, J. A. & Campell, B. M. (2006). Effects of balance training on selected skills. The Journal of Strength & Conditioning Research, 20(2), 422-428.
Corresponding author:
Miha Marinsek Faculty of Education University of Maribor Slovenska 36 2000 Maribor Slovenia •
miha.marinsek@uni-mb.si
Science of Gymnastics Journal
47
Science of Gymnastics Journal
Marinsek M., Slana U. SMART PHONE AS A STANDING BALANCE ASSESMENT DEVICE
Vol. 6 Issue 3: 39 - 47
Science of Gymnastics Journal	48	Science of Gymnastics Journal