Novel Use of a Smartphone to Measure Standing Balance

Abstract

BackgroundBalance assessment and training is utilized by clinicians and their patients to measure and improve balance. There is, however, little consistency in terms of how clinicians, researchers, and patients measure standing balance. Utilizing the inherent sensors in every smartphone, a mobile application was developed to provide a method of objectively measuring standing balance.ObjectiveWe aimed to determine if a mobile phone application, which utilizes the phone’s accelerometer, can quantify standing balance.MethodsThree smartphones were positioned simultaneously above the participants’ malleolus and patella and at the level of the umbilicus. Once secured, the myAnkle application was initiated to measure acceleration. Forty-eight participants completed 8 different balance exercises separately for the right and left legs. Accelerometer readings were obtained from each mobile phone and mean acceleration was calculated for each exercise at each ankle and knee and the torso.ResultsMean acceleration vector magnitude was reciprocally transformed to address skewness in the data distribution. Repeated measures ANOVAs were completed using the transformed data. A significant 2-way interaction was revealed between exercise condition and the body position of the phone (P<.001). Post-hoc tests indicated higher acceleration vector magnitude for exercises of greater difficulty. ANOVAs at each body position were conducted to examine the effect of exercise. The results revealed the knee as the location most sensitive for the detection of differences in acceleration between exercises. The accelerometer ranking of exercise difficulty showed high agreement with expert clinical rater rankings (kappa statistic>0.9).ConclusionsThe myAnkle application revealed significantly greater acceleration magnitude for exercises of greater difficulty. Positioning of the mobile phone at the knee proved to be the most sensitive to changes in accelerometer values due to exercise difficulty. Application validity was shown through comparison with clinical raters. As such, the myAnkle app has utility as a measurement tool for standing balance.

Reliability, Validity and Utility of Inertial Sensor Systems for Postural Control Assessment in Sport Science and Medicine Applications: A Systematic Review

“…These findings are in line with previously published work which has demonstrated that males possess poorer PC than females, while PC improves with the transition from youth to young adulthood [74]. A further nine studies compared PC performance during different static stance conditions [29,32,41,43,44,[46][47][48]75]. Two studies failed to quantitatively compare conditions [32,43], while the seven remaining studies demonstrated significant differences between conditions, with more challenging stances causing increased instability [25,29,41,44,46,48,76].…”

“…The likely explanation for the popularity of this methodology is the lumbosacral regions proximity to the COM, with previous reports demonstrating that it closely matches the acceleration of the COM [66,67]. While seven studies investigated the use of multiple sensor mounting locations, only two studies with differing methodologies directly compared locations to determine which provided the most valuable measure of PC [13,25]. As such, the authors draw no conclusions pertaining to optimal sensor mounting location.…”

“…Only two studies compared multiple mounting locations directly, with Brown and colleagues [13] reporting that the head mounted tri-axial accelerometer/gyroscope had the best agreement with the clinical BESS, when compared to the sternum, waist, wrist and shank sensors. Additionally, Shah and colleagues [25] reporting that the thigh mounted tri-axial accelerometer possessed the greatest sensitivity compared to shank and waist tri-axial accelerometers, when differentiating static stance conditions.…”

“…Twelve studies assessed discriminant validity in healthy cohorts, assessing the ability of inertial sensor systems to discriminate between known stance, visual, hearing, fatigue or taping conditions [25,28,32,41,[43][44][45][46][47][48][49][50].…”

“…A further nine studies compared PC performance during different static stance conditions [29,32,41,43,44,[46][47][48]75]. Two studies failed to quantitatively compare conditions [32,43], while the seven remaining studies demonstrated significant differences between conditions, with more challenging stances causing increased instability [25,29,41,44,46,48,76]. As such, it is clear that the inertial sensor based measures of PC are capable of discriminating age group and/or sex differences, reproducing findings which have previously been observed in studies using 'gold-standard' comparisons [77].…”