Automatic detection of one-on-one tackles and ruck events using microtechnology in rugby union.

OBJECTIVES: To automate the detection of ruck and tackle events in rugby union using a specifically-designed algorithm based on microsensor data. DESIGN: Cross-sectional study. METHODS: Elite rugby union players wore microtechnology devices (Catapult, S5) during match-play. Ruck (n=125) and tackle (n=125) event data was synchronised with video footage compiled from international rugby union match-play ruck and tackle events. A specifically-designed algorithm to detect ruck and tackle events was developed using a random forest classification model. This algorithm was then validated using 8 additional international match-play datasets and video footage, with each ruck and tackle manually coded and verified if the event was correctly identified by the algorithm. RESULTS: The classification algorithm's results indicated that all rucks and tackles were correctly identified during match-play when 79.4±9.2% and 81.0±9.3% of the random forest decision trees agreed with the video-based determination of these events. Sub-group analyses of backs and forwards yielded similar optimal confidence percentages of 79.7% and 79.1% respectively for rucks. Sub-analysis revealed backs (85.3±7.2%) produced a higher algorithm cut-off for tackles than forwards (77.7±12.2%). CONCLUSIONS: The specifically-designed algorithm was able to detect rucks and tackles for all positions involved. For optimal results, it is recommended that practitioners use the recommended cut-off (80%) to limit false positives for match-play and training. Although this algorithm provides an improved insight into the number and type of collisions in which rugby players engage, this algorithm does not provide impact forces of these events.

Classification of team sport activities using a single wearable tracking device.

Wearable tracking devices incorporating accelerometers and gyroscopes are increasingly being used for activity analysis in sports. However, minimal research exists relating to their ability to classify common activities. The purpose of this study was to determine whether data obtained from a single wearable tracking device can be used to classify team sport-related activities. Seventy-six non-elite sporting participants were tested during a simulated team sport circuit (involving stationary, walking, jogging, running, changing direction, counter-movement jumping, jumping for distance and tackling activities) in a laboratory setting. A MinimaxX S4 wearable tracking device was worn below the neck, in-line and dorsal to the first to fifth thoracic vertebrae of the spine, with tri-axial accelerometer and gyroscope data collected at 100Hz. Multiple time domain, frequency domain and custom features were extracted from each sensor using 0.5, 1.0, and 1.5s movement capture durations. Features were further screened using a combination of ANOVA and Lasso methods. Relevant features were used to classify the eight activities performed using the Random Forest (RF), Support Vector Machine (SVM) and Logistic Model Tree (LMT) algorithms. The LMT (79-92% classification accuracy) outperformed RF (32-43%) and SVM algorithms (27-40%), obtaining strongest performance using the full model (accelerometer and gyroscope inputs). Processing time can be reduced through feature selection methods (range 1.5-30.2%), however a trade-off exists between classification accuracy and processing time. Movement capture duration also had little impact on classification accuracy or processing time. In sporting scenarios where wearable tracking devices are employed, it is both possible and feasible to accurately classify team sport-related activities.