Location on the Body of a Wearable Accelerometer Affects Accuracy of Data for Identifying Equine Gaits

Thompson, Carol J.; Luck, L.; Keshwani, Jennifer; Pitla, Santosh K.; Karr, Lisa

doi:10.1016/j.jevs.2017.12.002

Cited by 12 publications

(13 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For this reason, previous studies have analyzed different anatomical locations at the same time. 42–44 Özdemir 43 placed the devices on head, chest, waist, right wrist, right thigh and right ankle to evaluate the fall risk in elderly people, finding the best results in the wrist. Boerema et al 42 used five accelerometers on different parts of the waist to measure physical activity, with the best values from the most lateral position of the participant’s waist belt.…”

Section: Discussionmentioning

confidence: 99%

“…41 To the best of the authors' knowledge, only a few non-sport-related studies have analyzed the behavior of accelerometers in different body locations. [42][43][44] Moreover, the WIMU PRO accelerometers have not been tested previously. Therefore, the aim of this investigation was to evaluate the reliability of triaxial accelerometers containing inertial devices to measure external loads for sport-related activity in the laboratory and under conditions of intermittent and continuous efforts when placed in different anatomical locations.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Static and dynamic reliability of WIMU PRO™ accelerometers according to anatomical placement

Gómez-Carmona

Bastida-Castillo

García-Rubio

et al. 2018

Proceedings of the Institution of Mechanical Engineers, Part P:

View full text Add to dashboard Cite

Currently, the use of accelerometry to analyze training and competition is on the increase. Thus, accelerometers must be reliable when calculating different variables from raw data. The purpose of this investigation was to assess the reliability of triaxial accelerometers that contain inertial devices for measuring external loads in sports. Four and eight WIMU PROä inertial devices (RealTrack Systems, Almeria, Spain) were assessed in laboratory and specific sport conditions, respectively. The laboratory test was performed in static (with and without stress) and dynamic (10 and 30 Hz vibrations) conditions. In addition, two tests were performed during continuous and intermittent activity: (1) an incremental progressive running test on the treadmill and (2) 5 min of a Soccer-Specific Aerobic Field Test (SAFT 90), where the devices were placed on ankle, knee, lower back and scapulae. Direct assessment of the accelerometers by calculating a vector magnitude, expressed as total acceleration (AcelT), was used for reliability using bias, 695% limits of agreement and coefficient of variation. A t-test and Pearson's r were performed for test-retest reliability. In the laboratory assessments, an excellent within-and between-device static (with and without stress) and dynamic reliability were found (coefficient of variation = 0.23%-0.78%). A very high reliability was also observed in the incremental treadmill test (coefficient of variation = 2.20%) and SAFT 90 (coefficient of variation = 2.96%) with a nearly perfect correlation between devices (r = 0.99-1.00). Finally, in the between-sessions reliability analysis, excellent results (p = 0.46-0.98; t = 0.01-0.73) and a very strong correlation (r = 0.86-0.96) was found with p values greater than 0.05 indicating no differences between the tests. In conclusion, the accelerometers of the tested device have shown excellent results for within-and betweendevice reliability and in test-retest analysis. Thus, this device can calculate all variables that depend on accelerometry measurement such as PlayerLoadä or impacts in different ranges, and in different anatomical locations.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Static and dynamic reliability of WIMU PRO™ accelerometers according to anatomical placement

Gómez-Carmona

Bastida-Castillo

García-Rubio

et al. 2018

Proceedings of the Institution of Mechanical Engineers, Part P:

View full text Add to dashboard Cite

show abstract

“…The orientation of the right accelerometer when the horse is standing is shown in Figure 1 and the three colored axis indicate the orientation of the accelerometer axis. This orientation was respected for all horses since a study (Thompson et al, 2018) revealed that the highest accuracies for detecting gaits could be reached at this location using an accelerometer. For successful data capturing the AX3 is securely fastened with the use of VELCRO stick on circles to the tendon boot with minimal room for vibration, slip or twist; to preserve that only the motions of the horse are captured.…”

Section: Data Collection Proceduresmentioning

confidence: 99%

Automatic equine activity detection by convolutional neural networks using accelerometer data

Eerdekens

Deruyck

Fontaine

et al. 2020

Computers and Electronics in Agriculture

View full text Add to dashboard Cite

In recent years, with a widespread of sensors embedded in all kind of mobile devices, human activity analysis is occurring more often in several domains like healthcare monitoring and fitness tracking. This trend did also enter the equestrian world because monitoring behaviours can yield important information about the health and welfare of horses. In this research, a deep learning-based approach for activity detection of equines is proposed to classify seven activities based on accelerometer data. We propose using Convolutional Neural Networks (CNN) by which features are extracted automatically by using strong computing capabilities. Furthermore, we investigate the impact of the sampling frequency, the time series length and the type of underground on which the data is gathered on the recognition accuracy and evaluate the model on three types of experimental datasets that are compiled of labelled accelerometer data gathered from six different subjects performing seven different activities. Afterwards, a horse-wise cross validation is carried out to investigate the impact of the subjects themselves on the model recognition accuracy. Finally, a slightly adjusted model is validated on different amounts of 50 Hz sensor data. A 99% accuracy can be reached for detecting seven behaviours of a seen horse when the sampling rate is 25 Hz and the time interval is 2.1 s. Four behaviours of an unseen horse can be detected with the same accuracy when the sampling rate is 69 Hz and the time interval is 2.4 s. Moreover, the accuracy of the model for the three datasets decreased on average with about 4.75% when the sampling rate was decreased from 200 Hz to 25 Hz and with 5.27% when the time interval was decreased from 3 s to 0.6 s. In addition, the classification performance of the activity "walk" was not influenced by the type of underground the horse was performing this movement on and even the model could conclude from which underground the data was gathered for three out of four undergrounds with accuracies above 93% at time intervals higher than 1.2 s. This ensures the evaluation of activity patterns in real world circumstances. The performance and ability of the model to generalise is validated on 50 Hz data from different horse types, using tenfold cross validation, reaching a mean classification accuracy of 97.84% and 96.10% when validated on a lame horse and pony, respectively. Moreover, in this work we show that using data from one sensors is at the cost of only 0.24% reduction in accuracy (99.42% vs 99.66%).

show abstract

“…17 This suggests that these technologies remain isolated from day-to-day performance management structures in the applied field, which is the putative future for these types of technologies. 12,27,28 When discussing the 'poor moving racehorse' Clements 29 stated that the Lameness Locator is still of limited value in racehorses. The author outlines that the lack of consistency and precise definition around the 'poor mover' presentation increases the difficulty of applying sensible diagnostic testing.…”

Section: Limits and Benefits Of Technology For The Collection Of Biommentioning

confidence: 99%

End-user practices in equine movement analysis: The potential of objective analysis tools to meet their needs

Egan

Brama

McGrath

2020

Proceedings of the Institution of Mechanical Engineers, Part P:

View full text Add to dashboard Cite

The use of objective gait analysis tools is becoming increasingly popular in equine research. The advancement of these technologies seeks to promote objective, mobile solutions which can be applied in equine training, monitoring or clinical contexts. This has led to the introduction of various tools in the commercial sphere. However, there has been slow adoption of objective technologies in the applied field. The purpose of this study was to understand existing equine movement analysis processes. We undertook qualitative interviews with industry stakeholders in the elite field to understand what key information they gather through their own, subjective analysis of the horse. We then mapped this information onto the capabilities of existing technologies to see what opportunities exist for user-driven technology solutions in this area. Our study highlights the need for knowledge exchange in the development and deployment of technologies within the equine industry. Commonalities in the important movement features identified by the stakeholders are presented that could be usefully considered and interpreted by future technology developers and researchers with respect to the underpinning biomechanical parameters. Objective technologies designed to capture and integrate these user-based parameters could be developed to augment the current subjective practices in equine analysis and monitoring.

show abstract

Location on the Body of a Wearable Accelerometer Affects Accuracy of Data for Identifying Equine Gaits

Cited by 12 publications

References 12 publications

Static and dynamic reliability of WIMU PRO™ accelerometers according to anatomical placement

Static and dynamic reliability of WIMU PRO™ accelerometers according to anatomical placement

Automatic equine activity detection by convolutional neural networks using accelerometer data

End-user practices in equine movement analysis: The potential of objective analysis tools to meet their needs

Contact Info

Product

Resources

About