Purpose:To assess the reliability of triaxial accelerometers as a measure of physical activity in team sports.Methods:Eight accelerometers (MinimaxX 2.0, Catapult, Australia) were attached to a hydraulic universal testing machine (Instron 8501) and oscillated over two protocols (0.5 g and 3.0 g) to assess within- and between-device reliability. A static assessment was also conducted. Secondly, 10 players were instrumented with two accelerometers during Australian football matches. The vector magnitude was calculated, expressed as Player load and assessed for reliability using typical error (TE) ± 90% confidence intervals (CI), and expressed as a coefficient of variation (CV%). The smallest worthwhile difference (SWD) in Player load was calculated to determine if the device was capable of detecting differences in physical activity.Results:Laboratory: Within- (Dynamic: CV 0.91 to 1.05%; Static: CV 1.01%) and between-device (Dynamic: CV 1.02 to 1.04%; Static: CV 1.10%) reliability was acceptable across each test. Field: The between-device reliability of accelerometers during Australian football matches was also acceptable (CV 1.9%). The SWD was 5.88%.Conclusions:The reliability of the MinimaxX accelerometer is acceptable both within and between devices under controlled laboratory conditions, and between devices during field testing. MinimaxX accelerometers can be confidently utilized as a reliable tool to measure physical activity in team sports across multiple players and repeated bouts of activity. The noise (CV%) of Player load was lower than the signal (SWD), suggesting that accelerometers can detect changes or differences in physical activity during Australian football.

Purpose:To describe the external load of Australian football matches and training using accelerometers.Methods:Nineteen elite and 21 subelite Australian footballers wore accelerometers during matches and training. Accelerometer data were expressed in 2 ways: from all 3 axes (player load; PL) and from all axes when velocity was below 2 m/s (PLSLOW). Differences were determined between 4 playing positions (midfielders, nomadics, deeps, and ruckmen), 2 playing levels (elite and subelite), and matches and training using percentage change and effect size with 90% confidence intervals.Results:In the elite group, midfielders recorded higher PL than nomadics and deeps did (8.8%, 0.59 ± 0.24; 34.2%, 1.83 ± 0.39 respectively), and ruckmen were higher than deeps (37.2%, 1.27 ± 0.51). Elite midfielders, nomadics, and ruckmen recorded higher PLSLOW than deeps (13.5%, 0.65 ± 0.37; 11.7%, 0.55 ± 0.36; and 19.5%, 0.83 ± 0.50, respectively). Subelite midfielders were higher than nomadics, deeps, and ruckmen (14.0%, 1.08 ± 0.30; 31.7%, 2.61 ± 0.42; and 19.9%, 0.81 ± 0.55, respectively), and nomadics and ruckmen were higher than deeps for PL (20.6%, 1.45 ± 0.38; and 17.4%, 0.57 ± 0.55, respectively). Elite midfielders, nomadics, and ruckmen recorded higher PL (7.8%, 0.59 ± 0.29; 12.9%, 0.89 ± 0.25; and 18.0%, 0.67 ± 0.59, respectively) and PLSLOW (9.4%, 0.52 ± 0.30; 11.3%, 0.68 ± 0.25; and 14.1%, 0.84 ± 0.61, respectively) than subelite players. Small-sided games recorded the highest PL and PLSLOW and were the only training drill to equal or exceed the load from matches across positions and playing levels.Conclusion:PL differed between positions, with midfielders the highest, and between playing levels, with elite higher. Differences between matches and training were also evident, with PL from small-sided games equivalent to or higher than matches.

Objective assessment of an athlete's performance is of importance in elite sports to facilitate detailed analysis. The implementation of automated detection and recognition of sport-specific movements overcomes the limitations associated with manual performance analysis methods. The object of this study was to systematically review the literature on machine and deep learning for sport-specific movement recognition using inertial measurement unit (IMU) and, or computer vision data inputs. A search of multiple databases was undertaken. Included studies must have investigated a sportspecific movement and analysed via machine or deep learning methods for model development. A total of 52 studies met the inclusion and exclusion criteria. Data preprocessing, processing, model development and evaluation methods varied across the studies. Model development for movement recognition were predominantly undertaken using supervised classification approaches. A kernel form of the Support Vector Machine algorithm was used in 53% of IMU and 50% of vision-based studies. Twelve studies used a deep learning method as a form of Convolutional Neural Network algorithm and one study also adopted a Long Short Term Memory architecture in their model. The adaptation of experimental setup , data pre-processing, and model development methods are best considered in relation to the characteristics of the targeted sports movement(s).

No abstract

Myofascial trigger points (TrPs) have been clinically described as discrete areas of muscle tenderness presenting in taut bands of skeletal muscle. Using well-defined clinical criteria, prior investigations have demonstrated interrater reliability in the diagnosis of TrPs within a given muscle. No reports exist, however, with respect to the precision with which experienced clinicians can determine the anatomic locations of TrPs within a muscle. This paper details a study wherein four trained clinicians achieved statistically significant reliability (see below) in estimating the precise locations of latent TrPs in the trapezius muscle of volunteer subjects (n=20). To do so, the clinicians trained extensively together prior to the study. The precise anatomic location of each subject's primary TrP was measured in a blinded fashion using a 3 dimensional (3-D) camera system. Use of this measurement system permitted the anatomic co-ordinates of each TrP to be located without providing feedback to subsequent clinicians. The clinicians each used a pressure algometer along with patient feedback to document the sensitivity of each suspected TrP site, however unlike routine clinical practice, the algometry was performed with a double-blinded approach hence the results were only examined post-hoc. At the time of data collection (algometry readings unknown), 16 of the 20 subjects were judged to present with a latent TrP. Subsequently, when subjected to a criterion pressure threshold value of <3.0 kg.cm(-2), 12 of these TrPs were classified as being clinically sensitive. To assess the 3-D measurement precision, and the reliability of the TrP estimates, statistical measures of the SEM and the Generalizability coefficient (G-coeff) were determined for all suspected TrP sites in the superior-inferior, medial-lateral and anterior-posterior directions. The best results were determined by pooling the measurements of all 4 clinicians, however, based upon exceeding a criterion reliability threshold of 80%, the use of just two testers was found to produce reliable results. The two-tester condition yielded a precision of 7.5, 7.6 and 6.5 mm (SEM) with reliability (G-coeff) of 0.92, 0.86 and 0.83, respectively. Given the double-blinded methodology, the use of pressure algometry was also found to demonstrate internal validity. The algometer responses associated with TrP estimates varied inversely with respect to the clinical group's reliability in identify the TrP locations. To summarize, for the trapezius muscle, this study demonstrates that two trained examiners can reliably localize latent TrPs with a precision that essentially approaches the physical dimensions of the clinician's own fingertips. Finally, it should be recognized that the ability to precisely document TrP location appears critical to the success of future studies that may be designed to investigate the etiology and pathogenesis of this commonly diagnosed clinical disorder.

In this study, we examined the relationships between body sway, aim point fluctuation and performance in rifle shooting on an inter- and intra-individual basis. Six elite shooters performed 20 shots under competition conditions. For each shot, body sway parameters and four aim point fluctuation parameters were quantified for the time periods 5 s to shot, 3 s to shot and 1 s to shot. Three parameters were used to indicate performance. An AMTI LG6-4 force plate was used to measure body sway parameters, while a SCATT shooting analysis system was used to measure aim point fluctuation and shooting performance. Multiple regression analysis indicated that body sway was related to performance for four shooters. Also, body sway was related to aim point fluctuation for all shooters. These relationships were specific to the individual, with the strength of association, parameters of importance and time period of importance different for different shooters. Correlation analysis of significant regressions indicated that, as body sway increased, performance decreased and aim point fluctuation increased for most relationships. We conclude that body sway and aim point fluctuation are important in elite rifle shooting and performance errors are highly individual-specific at this standard. Individual analysis should be a priority when examining elite sports performance.

No abstract

The swimming start is typically broken into three sub-phases; on-block, flight, and underwater phases. While overall start performance is highly important to elite swimming, the contribution of each phase and important technical components within each phase, particularly with the new kick-start technique, has not been established. The aim of this study was to identify technical factors associated with overall start performance, with a particular focus on the underwater phase. A number of parameters were calculated from 52 starts performed by elite freestyle and butterfly swimmers. These parameters were split into above-water and underwater groupings, before factor analysis was used to reduce parameter numbers for multiple regression. For the above-water phases, 81% of variance in start performance was accounted for by take-off horizontal velocity. For the underwater water phase, 96% of variance was accounted for with time underwater in descent, time underwater in ascent and time to 10 m. Therefore, developing greater take-off horizontal velocity and focussing on the underwater phase by finding the ideal trajectory will lead to improved start performance.