Evaluation of postural stability in quietly standing humans and horses has emerged as a method for the assessment and monitoring of neuromuscular disorders. In humans and other animals, the body establishes static balance by maintaining the COM over the base of support.1 To accomplish this balance, sensory and neuromuscular systems integrate through visual, proprioceptive, and vestibular pathways.2 Integration of these systems and pathways allows the body to make small adjustments in muscle activity, which results in slight movements of the COM so that the body can remain in an upright position while the subject is standing quietly.3
Postural stability in quietly standing humans has been used to assess risk of falling for elderly adults,4 longitudinal effects of head injury,5 neurologic disorders (eg, Parkinson disease),6,7 effects of sports training,8 and effects of chronic joint pain.9 Improvements in postural stability determined by a decrease in the measured area of sway indicate that a subject has better neuromuscular control and, consequently, improvement in balance. Thus, assessment of postural stability can contribute to diagnosing and monitoring of neuromuscular diseases.
Postural stability has not been as thoroughly evaluated in horses as it has in humans. Changes in postural stability of horses as an assessment of neuromuscular control have been used to examine lameness,3 effects of rehabilitation,3 balance changes in growing foals,10 body mass,11 and effects attributable to wearing a blindfold.3,12
Use of a stationary force platform is the criterion-referenced standard for evaluation of postural stability in both humans and horses. Slight adjustments for maintaining the COM within the base of support can be measured as movement of the COP when a subject is standing on a stationary force platform.1,3 Strong agreement between COP and COM movement has been detected.13,14 Although stationary force platforms can be used to accurately detect differences in COP on the basis of limb position, visual input, and musculoskeletal injury,3,12 such devices are expensive, typically restricted to a laboratory environment, and not widely available. However, there has been increased use of body-mounted sensors to investigate the gait for both human and equine biomechanical analysis in nonresearch settings. In human medicine and rehabilitation, body-mounted inertial sensors, such as accelerometers, have been investigated for use in assessing postural stability.13,15–17 These devices are becoming widely available and less expensive. Because use of these sensors has enabled investigators to discriminate postural changes of people (as assessed by postural stability), these sensors may have a similar use for horses.
Currently, many handheld devices, such as smartphones and PMDs, contain inertial sensing components (accelerometers and gyroscopes). These smartphones are lightweight, easy to use, and widely available and are less expensive than other inertial sensing devices. In addition, there are numerous applications available to retrieve the sensor data from these devices. The accelerometers contained within several smartphones have been investigated in humans to evaluate gait,18–20 hand tremors associated with Parkinson disease,21 and the patellar reflex.22 Recently, 1 type of smartphone was investigated for its use in evaluation of hind limb asymmetry in trotting horses.23 Because more veterinarians are using smartphones and other handheld devices, it is reasonable that such devices could also be used as a diagnostic tool.
Many smartphones and PMDs are alike in numerous regards, including the processor, sensors, and applications for extracting sensor data. However, PMDs do not require a contract phone plan, and substantial damage to PMDs will not result in loss of the ability to communicate with clients. One commonly used PMD also has both Bluetooth and wireless capabilities, so data can be exported to another device for processing and analysis, regardless of the location of the PMD.
The objective of the study reported here was to determine the usefulness of a PMD for assessment of postural stability in quietly standing horses. Specifically, we wanted to compare usefulness of the PMD for assessing changes in postural stability with results for the criterion-referenced standard, a stationary force platform. We hypothesized that data from the accelerometer within the PMD could be used to differentiate between several stance conditions and that accelerations would be strongly correlated with postural sway variables collected by use of a stationary force platform system.
Materials and Methods
Horses
Seven skeletally mature horses (4 mares, 2 geldings, and 1 stallion) were used in the study. Mean age of horses was 3.2 years (range, 2 to 4 years), and mean body weight was 344 kg (range, 287 to 393 kg). All protocols were approved by an institutional animal care and use committee prior to the start of the study.
Data collection
All trials were performed in the authors’ gait analysis laboratory, and all horses were allowed to acclimate to the equipment prior to data collection. For data collection, each horse was positioned on 2 stationary force platformsa; both forelimbs were on 1 stationary force platform, and both hind limbs were on the other stationary force platform (Figure 1). A trial was considered successful when a horse stood squarely, without extraneous head and neck movement, for a minimum of 10 seconds.
Collection of PMD data
Each horse was fitted with a breastplate and surcingle. The surcingle was placed immediately caudal to the highest point of the shoulders (withers) over the spinous processes of T8–T10. One investigator (VJM) mounted a PMDb in a protective case with a belt clip attachmentc onto the surcingle at a location over the dorsal midline; the PMD was mounted by use of plastic cable ties (Figure 1). The PMD was oriented so that the long axis was in a craniocaudal direction (the distal aspect of the device was positioned cranially), and the midline of the PMD was centered over the dorsal midline of the horse. In addition, the PMD was assessed visually and deemed to be approximately parallel with the ground (craniocaudal and mediolateral directions). Data were collected from the triaxial accelerometerd (± 2 g) at 100 Hz with a resolution of ± 0.002 m/s2 by use of a commercial data logging application.24,e Data were exported via email and imported into a spreadsheetf for further processing, which included conversion of acceleration data (collected as gravitational acceleration) to meters per second squared. No other processing of acceleration data was performed.
Photograph of a sedated horse in a square stance at 5 minutes after administration of xylazine hydrochloride (0.35 mg/kg, IV). The horse is standing on 2 stationary force platforms; both forelimbs are on one force platform, and both hind limbs are on the other force platform. The horse is instrumented with a PMD, which is attached by use of a surcingle located immediately caudal to the highest point of the shoulder (withers). The long axis of the PMD is oriented craniocaudally, with the distal aspect of the device positioned cranially.
Citation: American Journal of Veterinary Research 78, 9; 10.2460/ajvr.78.9.1036
Accelerations in the craniocaudal and mediolateral directions were extracted from the PMD. Total range of acceleration and SDs for acceleration in both planes were determined.
Stationary force platform data collection
The COP data were collected from both stationary force platforms at 3,000 Hz and filtered at 15 Hz with a fourth-order, recursive Butterworth (low-pass) filter by use of a commercial motion analysis system,g as described elsewhere.3 A minimum of five 10-second trials were conducted for each stance condition. Trials were rejected when a horse did not remain in a static stance position with all 4 limbs in contact with the force platforms and the head and neck in a natural and comfortable position for each horse, which was maintained throughout a single stance condition.
The COP displacement and velocity in the craniocaudal and mediolateral directions were determined as described elsewhere.3 Displacements and velocities in both the craniocaudal and mediolateral directions represented the range of motion during each individual 10-second trial. The SDs were determined for COP displacement and velocity in both the craniocaudal and mediolateral directions.
Stance conditions
All horses were evaluated in 4 stance conditions, which were conducted in the same order for each horse: normal square stance, forelimb base-narrow stance, normal square stance at 5 minutes after sedation with xylazine hydrochloride (0.35 mg/kg, IV), and normal square stance at 10 minutes after sedation with xylazine hydrochloride. For the forelimb base-narrow stance, the medial aspects of both hooves of the forelimbs were in contact with each other.
Statistical analysis
Data were analyzed with commercial statistical software.h Data were first assessed for a normal distribution by use of the Shapiro-Wilk test of normality. Stationary force platform data were transformed by calculating the inverse of the square root of a value (ie, 1/square root of a value). The PMD data were transformed by calculating the inverse of a squared value (ie, 1/value2). Mean and 95% CI values were calculated for transformed data and were then back-transformed. A mixed-model ANOVA was used to compare each variable separately, with horse as a random variable and stance condition as a fixed categorical variable; normal square stance was set as the control condition. Pearson correlation coefficients were calculated to compare results for the PMD and stationary force platform system. Significance was set at values of P < 0.05.
Results
Horses
A minimum of 5 trials were conducted for each horse, and 4 to 6 trials/horse for the PMD and stationary force platform for each stance condition were used for analysis. The PMD remained firmly in position throughout the trials on each horse.
PMD variables
Significant differences were detected for the craniocaudal range of acceleration (P < 0.001), craniocaudal SD (P < 0.001), and mediolateral SD (P = 0.028) when stance condition was altered (Table 1). Differences attributable to specific stance conditions were examined, and mediolateral SD was the only variable that differed significantly (P = 0.024) between the normal square stance and forelimb base-narrow stance. Comparison between the normal square stance and the normal square stance at 5 minutes after xylazine administration revealed significant differences in the craniocaudal range of acceleration (P < 0.001), craniocaudal SD (P < 0.001), and mediolateral SD (P = 0.018).
Back-transformed mean and 95% CI values for PMD-derived variables in the craniocaudal and mediolateral directions for 7 horses examined while standing on a stationary force platform system in a normal square stance, forelimb base-narrow stance, and normal square stance at 5 and 10 minutes after sedation induced by administration of xylazine hydrochloride (0.35 mg/kg, IV).
| Normal square stance | Forelimb base-narrow stance | Stance at 5 minutes after xylazine | Stance at 10 minutes after xylazine | |||||
|---|---|---|---|---|---|---|---|---|
| Variable | Mean | 95% CI | Mean | 95% CI | Mean | 95% CI | Mean | 95% CI |
| Craniocaudal acceleration range (m/s2) | 0.3265 | 0.3058–0.3520 | 0.3451 | 0.3201–0.3770 | 0.4259* | 0.3852–0.4829 | 0.3759* | 0.3450–0.4170 |
| Mediolateral acceleration range (m/s2) | 0.2820 | 0.2657–0.3017 | 0.2914 | 0.2725–0.3149 | 0.2844 | 0.2620–0.3136 | 0.2735 | 0.2552–0.2964 |
| Craniocaudal acceleration SD (m/s2) | 0.0466 | 0.0446–0.0488 | 0.0469 | 0.0446–0.0497 | 0.0613* | 0.0560–0.0685 | 0.0547* | 0.0499–0.0611 |
| Mediolateral acceleration SD (m/s2) | 0.0384 | 0.0367–0.0404 | 0.0406* | 0.0389–0.0427 | 0.0410* | 0.0387–0.0439 | 0.0388 | 0.0367–0.0414 |
Results represent data for 4 to 6 successful trials/horse.
Within a row, value differs significantly (P < 0.05) from the value for normal square stance.
Stationary force platform variables
Significant (P < 0.001) differences were detected in total COP displacement, COP velocity, COP displacement SD, and COP velocity SD for both the craniocaudal and mediolateral directions (Table 2). Comparison between the normal square stance and forelimb base-narrow stance revealed significant (P = 0.004) differences in mediolateral total COP displacement and mediolateral COP SD. Comparison between the normal square stance and the normal square stance at 5 minutes after xylazine administration revealed significant (P < 0.001) differences for all variables.
Back-transformed mean and 95% CI values for variables derived by use of a stationary force platform system for 7 horses while standing in a normal square stance, forelimb base-narrow stance, and normal square stance at 5 and 10 minutes after sedation induced by IV administration of xylazine.
| Normal square stance | Forelimb narrow-base stance | Stance at 5 minutes after xylazine | Stance at 10 minutes after xylazine | |||||
|---|---|---|---|---|---|---|---|---|
| Variable | Mean | 95% CI | Mean | 95% CI | Mean | 95% CI | Mean | 95% CI |
| Craniocaudal COP displacement (mm) | 10.22 | 8.88–11.88 | 8.81 | 7.44–10.59 | 22.33* | 18.62–27.28 | 15.61* | 12.79–19.47 |
| Mediolateral COP displacement (mm) | 10.26 | 8.92–11.92 | 7.62* | 6.44–9.16 | 19.97* | 16.43–24.80 | 13.78* | 11.24–17.30 |
| Craniocaudal COP velocity (m/s) | 0.0050 | 0.0044–0.0057 | 0.0047 | 0.0041–0.0056 | 0.0088* | 0.0079–0.0099 | 0.0075* | 0.0064–0.0088 |
| Mediolateral COP velocity (m/s) | 0.0027 | 0.0025–0.0030 | 0.0025 | 0.0022–0.0028 | 0.0061* | 0.0052–0.0074 | 0.0040* | 0.0033–0.0050 |
| Craniocaudal COP displacement SD (mm) | 0.0024 | 0.0021–0.0028 | 0.0021 | 0.0018–0.0025 | 0.0061* | 0.0050–0.0076 | 0.0040* | 0.0032–0.0051 |
| Mediolateral COP displacement SD (mm) | 0.0029 | 0.0025–0.0033 | 0.0021* | 0.0017–0.0025 | 0.0054* | 0.0043–0.0069 | 0.0037* | 0.0030–0.0047 |
| Craniocaudal COP velocity SD (m/s) | 0.0045 | 0.0038–0.0053 | 0.0042 | 0.0034–0.0052 | 0.0066* | 0.0058–0.0075 | 0.0059* | 0.0051–0.0069 |
| Mediolateral COP velocity SD (m/s) | 0.0022 | 0.0020–0.0024 | 0.0021 | 0.0019–0.0023 | 0.0046* | 0.0039–0.0054 | 0.0032* | 0.0026–0.0039 |
Results represent data for 4 to 6 successful trials/horse.
See Table 1 for remainder of key.
Correlation between the stationary force platform and the PMD
Correlation coefficients between variables determined by use of the stationary force platform and the PMD ranged from r = 0.18 to 0.58 (Tables 3 and 4). Correlation coefficients between the SDs were higher than those calculated between ranges of motion (COP displacement to PMD acceleration range or COP velocity to PMD acceleration range) in the craniocaudal or mediolateral directions. Additionally, correlation coefficients were higher for the craniocaudal direction than the mediolateral direction.
Pearson correlation coefficient (r) between variables determined by use of data from a stationary force platform system and a PMD for 7 horses in various stances.
| PMD | |||
|---|---|---|---|
| Force platform | Craniocaudal acceleration range | Mediolateral acceleration range | P value |
| Craniocaudal COP displacement | 0.36 | — | < 0.001 |
| Mediolateral COP displacement | — | 0.18 | 0.039 |
| Craniocaudal COP velocity | 0.44 | — | < 0.001 |
| Mediolateral COP velocity | — | 0.22 | 0.010 |
— = Not applicable.
Pearson correlation coefficient (r) between SD of variables determined by use of data from a stationary force platform system and a PMD for 7 horses in various stances.
| PMD | |||
|---|---|---|---|
| Force platform | Craniocaudal acceleration SD | Mediolateral acceleration SD | P value |
| Craniocaudal COP displacement SD | 0.44 | — | < 0.001 |
| Mediolateral COP displacement SD | — | 0.25 | 0.003 |
| Craniocaudal COP velocity SD | 0.58 | — | < 0.001 |
| Mediolateral COP velocity SD | — | 0.33 | < 0.001 |
See Table 3 for key.
Discussion
Results of the study reported here supported the hypothesis that data from the accelerometer within a commonly used PMD could be used to differentiate among stance conditions in clinically normal horses. Changes to limb position can alter COP motion in horses standing quietly. Investigators of 1 study3 found that placing a horse in a base-narrow stance significantly altered postural sway in the mediolateral direction but not the craniocaudal direction. Similarly, lateral sway of humans is more affected than craniocaudal sway following changes to stance width.25 Results of the present study obtained by use of a stationary force platform system supported these findings because both COP displacement and COP displacement SD were significantly altered in the forelimb base-narrow stance in only the mediolateral direction. Interestingly, mediolateral COP velocity and COP velocity SD in the forelimb base-narrow stance were not significantly different from values for the normal square stance. Examination of the effects of limb position on COM motion revealed that data for the PMD were only useful for identifying significant changes in mediolateral acceleration SD between the normal square and forelimb base-narrow stances. This result supported previous findings for COP motion3,25 as well as results concurrently obtained by use of the stationary force platform system during the present study.
Additional methods for altering sensory input and their effects on postural stability have been investigated in people and horses. Altering visual input has been investigated by applying blindfold to horses, which resulted in significant changes to both craniocaudal and mediolateral COP sway.3,12 Investigators of 1 study25 also identified changes in both craniocaudal and mediolateral COP motion when people had their eyes closed. For the present study, we examined another method for altering sensory input (ie, a pharmacological agent). This method also resulted in significant changes to both COP and COM motion during the normal square stance.
Xylazine hydrochloride is an α2-adrenergic receptor agonist and has CNS effects, including sedation, altered consciousness, and analgesia.26 Various degrees of ataxia can be induced by xylazine, depending on the dose administered. The sedative effect of this class of drugs is thought to result from its depressive effects within the lower brainstem.27 In the present study, we identified changes to both the range and SD of craniocaudal body accelerations at both 5 and 10 minutes after xylazine administration, but the change in mediolateral acceleration SD was significantly different from that for the normal square stance only at 5 minutes after xylazine administration. Similarly, investigators of another study28 reported that mediolateral COP range was not significantly different from baseline values at 15 minutes after administration of detomidine hydrochloride (10 μg/kg), whereas craniocaudal range did not return to baseline values until 30 minutes after administration of the sedative. The results for that study28 as well as results for the study reported here supported the contention that changes in mediolateral sway may return to baseline values more quickly than would changes in craniocaudal sway.
In the past few years, body-mounted sensors, such as accelerometers, have been used in human sports medicine and rehabilitation to quantify postural stability. In 1 recent study,14 investigators found that in quietly standing human subjects, COP velocities were highly correlated with COM accelerations, and both variables were useful for discriminating young from elderly subjects. In another study29 that involved use of a body-mounted accelerometer, root mean square of the accelerations, which represents a measure of variability, could be used to discriminate people with an increased risk of falling from those with a low risk of falling. In the present study, SDs were examined as another measure of variability. We found that SDs in both the craniocaudal and mediolateral directions were significantly affected by stance condition when evaluated by use of a stationary force platform system or the PMD. In another study,10 SDs for craniocaudal and mediolateral amplitude were found to decrease as foals aged and developed more musculoskeletal control. Other equine studies30,31 have also revealed significant changes in SD following changes to sensory input, such as perineural anesthesia. Results for the study reported here and from previous investigations supported the analysis of variability (eg, SD) as a method for investigating neuromuscular conditions.
It is impossible to collect absolute COM accelerations by use of a horse-mounted inertial sensing device without invasively placing a sensor within the horse. Accelerations determined by use of data from the PMD were estimations of COM body motion. In humans, it has been suggested32 that a location external to the COM can be used, provided that location is moving parallel to the COM. The COM motion during exercise has been estimated by use of a sensor mounted over the spinous processes of T4-T6.33 Although the PMD of the present study was placed slightly more caudal than that location, it has been suggested34 that measuring overall trunk motion can be used as an estimate of COM motion. Thus, the method used in the present study should be adequate to estimate accelerations of the COM. It is important to mention that movement of the head and neck in horses results in measureable changes to COP motion owing to the long lever arm of the equine neck,28 and head movements in the craniocaudal direction result in large fluctuations in COP.35 The effects of small-magnitude head and neck movements on COM accelerations are not known, but it is recommended to conduct additional trials to replace those in which there was head and neck movement.
Correlation coefficients were calculated between results for the stationary force platform system and PMD to assess approximate agreement between the 2 systems. Although all correlations were significant, the correlation coefficients were only low to moderate in association (r = 0.18 to 0.58). Variables in the craniocaudal direction had higher correlation coefficients than did the corresponding mediolateral variables. This could have been related to differences in range of motion in the craniocaudal and mediolateral directions within each system. For COP, mean displacement was similar in both the craniocaudal and mediolateral directions. However, the mean range of acceleration of the COM for the PMD was larger in the craniocaudal than mediolateral direction. These differences indicated that estimated COM accelerations at the measured location did not perfectly reflect COP displacement. Additionally, correlation coefficients for SDs were higher than those for the displacement and acceleration ranges, with the highest values for comparisons of COP velocity and COM acceleration. A high correlation coefficient (r = 0.838) between COP velocity and COM acceleration in the anteroposterior direction has been reported for quietly standing people.14
Several potential explanations exist for the lower correlation coefficients in the present study. First, the PMD was attached to each horse with a surcingle; inherently, a small amount of motion would be expected from this attachment. However, we did not visually detect any motion of the PMD, which made this explanation less likely. Additionally, the accelerometer within a smartphone used in another study23 (which was made by the same manufacturer that made the PMD used in the present study) was identified as being located slightly to the left of midline. The horses of the present study were not moving; thus, the PMD remained stationary on each horse. Because the accelerations obtained by use of data from the PMD were ranges, we did not believe that positioning the accelerometer slightly off the midline would have an effect on the results. Nevertheless, if the PMD were not placed level on the surcingle, this could have accounted for the lower correlation coefficients. The PMD visibly appeared to be level; however, an external or internal leveling tool was not used to ensure that it was exactly level. When a PMD is used longitudinally over time, it would be important to precisely mount the device so that measurements could be compared; thus, use of a leveling tool would be recommended.
Another limitation was that although data were collected by use of both the stationary force platform system and PMD, the start of each trial was not perfectly synchronized; therefore, differences arising from timing were highly likely. Finally, the duration of time of quiet standing for each trial was short (10 seconds), compared with the duration of time recommended to collect reliable postural stability data for humans (20 to 30 seconds).36 Postural stability examinations by use of a stationary platform found that horses tended to move if required to stand quietly for > 10 seconds.35 During data collection for the nonsedated horses in the present study, it was noted that horses did not stand still for much longer than 10 seconds, so collecting data for a longer period was not feasible.
Analysis of results of the study reported here indicated that data from the accelerometer in a commonly used PMD mounted over the approximate COM could be used to identify significant differences between stance conditions in horses. Additional studies with this type of device should be conducted to determine its utility when used to differentiate neuromuscular function attributable to experimentally induced or naturally occurring causes. Further investigation of the interday reliability of the PMD would be necessary to determine its use in longitudinal diagnostic testing.
ABBREVIATIONS
| CI | Confidence interval |
| COM | Center of mass |
| COP | Center of pressure |
| PMD | Portable media device |
Footnotes
Model FP6090-15, Bertec Corp, Columbus, Ohio.
Fifth-generation 16-GB iPod Touch, Apple Inc, Cupertino, Calif.
Survivor All-Terrain for fifth-generation iPod Touch, Griffin Technology, Nashville, Tenn.
LIS331DLH (MEMS type), STMicroelectronics, Geneva, Switzerland.
Movement data logger, version 1.20711, Navigated Technologies LLC, Cleveland, Tenn.
Microsoft Excel 2010, Microsoft Corp, Redmond, Wash.
Vicon-Motus 9.2, Vicon Motion Systems Inc, Centennial, Colo.
STATA, version 13.1, StataCorp LP, College Station, Tex.
References
1. Røgind H, Lykkegaard JJ, Bliddal H, et al. Postural sway in normal subjects aged 20–70 years. Clin Physiol Funct Imaging 2003;23:171–176.
2. De Carli P, Patrizi M, Pepe L, et al. Postural control and risk of falling in bipodalic and monopodalic stabilometric tests of healthy subjects before, after visuo-proprioceptive vestibule-postural rehabilitation and at 3 months thereafter: role of the proprioceptive system. Acta Otorhinolaryngol Ital 2010;30:182–189.
3. King MR, Haussler KK, Kawcak CE, et al. Effect of underwater treadmill exercise on postural sway in horses with experimentally induced carpal joint osteoarthritis. Am J Vet Res 2013;74:971–982.
4. Melzer I, Benjuya N, Kaplanski J. Postural stability in the elderly: a comparison between fallers and non-fallers. Age Ageing 2004;33:602–607.
5. Lehmann JF, Boswell S, Price R, et al. Quantitative evaluation of sway as an indicator of functional balance in post-traumatic brain injury. Arch Phys Med Rehabil 1990;71:955–962.
6. Mancini M, Carlson-Kuhta P, Zampieri C, et al. Postural sway as a marker of progression in Parkinson's disease: a pilot longitudinal study. Gait Posture 2012;36:471–476.
7. Mancini M, Salarian A, Carlson-Kuhta P, et al. ISway: a sensitive, valid and reliable measure of postural control. J Neuroeng Rehabil 2012;9:59.
8. James EG. Short-term differential training decreases postural sway. Gait Posture 2014;39:172–176.
9. Baierle T, Kromer T, Petermann C, et al. Balance ability and postural stability among patients with painful shoulder disorders and healthy controls. BMC Musculoskelet Disord 2013;14:282.
10. Nauwelaerts S, Malone SR, Clayton HM. Development of postural balance in foals. Vet J 2013;198:e70–e74.
11. Clayton HM, Buchholz R, Nauwelaerts S. Relationships between morphological and stabilographic variables in standing horses. Vet J 2013;198:e65–e69.
12. Clayton HM, Nauwelaerts S. Effect of blindfolding on centre of pressure variables in healthy horses during quiet standing. Vet J 2014;199:365–369.
13. Whitney SL, Roche JL, Marchetti GF, et al. A comparison of accelerometry and center of pressure measures during computerized dynamic posturography: a measure of balance. Gait Posture 2011;33:594–599.
14. Masani K, Vette AH, Abe MO, et al. Center of pressure velocity reflects body acceleration rather than body velocity during quiet standing. Gait Posture 2014;39:946–952.
15. Mayagoitia RE, Lotters JC, Veltink PH, et al. Standing balance evaluation using a triaxial accelerometer. Gait Posture 2002;16:55–59.
16. Turcato A, Ramat S. Predicting losses of balance during upright stance: evaluation of a novel approach based on wearable accelerometers, in Proceedings. 32nd Annu Int Conf IEEE Eng Med Biol Soc 2010; 4918–4921.
17. Deshmukh PM, Russell CM, Lucarino LE, et al. Enhancing clinical measures of postural stability with wearable sensors, in Proceedings. 34th Annu Int Conf IEEE Eng Med Biol Soc 2012; 4521–4524.
18. LeMoyne R, Mastroianni T, Cozza M, et al. Implementation of an iPhone as a wireless accelerometer for quantifying gait characteristics, in Proceedings. 32nd Annu Int Conf IEEE Eng Med Biol Soc 2010; 3847–3851.
19. Nolan M, Mitchell JR, Doyle-Baker PK. Validity of the Apple iPhone/iPod Touch as an accelerometer-based physical activity monitor: a proof-of-concept study. J Phys Act Health 2014;11:759–769.
20. Furrer M, Bichsel L, Niederer M, et al. Validation of a smartphone-based measurement tool for the quantification of level walking. Gait Posture 2015;42:289–294.
21. LeMoyne R, Mastroianni T, Cozza M, et al. Implementation of an iPhone for characterizing Parkinson's disease tremor through a wireless accelerometer application, in Proceedings. 32nd Annu Int Conf IEEE Eng Med Biol Soc 2010; 4954–4958.
22. LeMoyne R, Mastroianni T, Grundfest W, et al. Implementation of an iPhone wireless accelerometer application for the quantification of reflex response, in Proceedings. 35th Annu Int Conf IEEE Eng Med Biol Soc 2013; 4658–4661.
23. Pfau T, Weller R. Comparison of a standalone consumer grade smartphone with a specialist inertial measurement unit for quantification of movement symmetry in the trotting horse. Equine Vet J 2017;49:124–129.
24. del Rosario MB, Redmond SJ, Lovell NH. Tracking the evolution of smartphone sensing for monitoring human movement. Sensors (Basel) 2015;15:18901–18933.
25. Day BL, Steiger MJ, Thompson PD, et al. Effect of vision and stance width on human body motion when standing: implications for afferent control of lateral sway. J Physiol 1993;469:479–499.
26. Valverde A. Alpha-2 agonists as pain therapy in horses. Vet Clin North Am Equine Pract 2010;26:515–532.
27. England GCW, Clarke KW. Alpha2 adrenoceptor agonists in the horse—a review. Br Vet J 1996;152:641–657.
28. Bialski D, Lanovaz JL, Bohart GV, et al. Effect of detomidine on postural sway in horses. Equine Comp Exerc Physiol 2004;1:45–50.
29. Doheny EP, McGrath D, Greene BR, et al. Displacement of centre of mass during quiet standing assessed using accelerometry in older fallers and non-fallers, in Proceedings. 34th Annu Int Conf IEEE Eng Med Biol Soc 2012; 3300–3303.
30. Peham C, Licka T, Girtler D, et al. The influence of lameness on equine stride length consistency. Vet J 2001;162:153–157.
31. Moorman VJ, Reiser RF II, Mahaffey CA, et al. Use of an inertial measurement unit to assess the effect of forelimb lameness on three-dimensional hoof orientation in horses at a walk and trot. Am J Vet Res 2014;75:800–808.
32. O'Sullivan M, Blake C, Cunningham C, et al. Correlation of accelerometry with clinical balance tests in older fallers and non-fallers. Age Ageing 2009;38:308–313.
33. Parsons KJ, Pfau T, Ferrari M, et al. High-speed gallop locomotion in the Thoroughbred racehorse. II. The effect of incline on centre of mass movement and mechanical energy fluctuation. J Exp Biol 2008;211:945–956.
34. Buchner HH, Obermuller S, Scheidl M. Body centre of mass movement in the sound horse. Vet J 2000;160:225–234.
35. Clayton HM, Bialski DE, Lanovaz JL, et al. Assessment of the reliability of a technique to measure postural sway in horses. Am J Vet Res 2003;64:1354–1359.
36. Le Clair K, Riach C. Postural stability measures: what to measure and for how long. Clin Biomech (Bristol, Avon) 1996;11:176–178.
