Automatic detection of passing and shooting in water polo using machine learning: a feasibility study.

There is currently no efficient way to quantify overhead throwing volume in water polo. Therefore, this study aimed to test the feasibility of a method to detect passes and shots in water polo automatically using inertial measurement units (IMU) and machine-learning algorithms. Eight water polo players wore one IMU sensor on the wrist (dominant hand) and one on the sacrum during six practices each. Sessions were filmed with a video camera and manually tagged for individual shots or passes. Data were synchronised between video tagging and IMU sensors using a cross-correlation approach. Support vector machine (SVM) and artificial neural networks (ANN) were compared based on sensitivity and specificity for identifying shots and passes. A total of 7294 actions were identified during the training sessions, including 945 shots and 5361 passes. Using SVM, passes and shots together were identified with 94.4% (95%CI = 91.8-96.4) sensitivity and 93.6% (95%CI = 91.4-95.4) specificity. Using ANN yielded similar sensitivity (93.0% [95%CI = 90.1-95.1]) and specificity (93.4% [95%CI = 91.1 = 95.2]). The results suggest that this method of identifying overhead throwing motions with IMU has potential for future field applications. A set-up with one single sensor at the wrist can suffice to measure these activities in water polo.

Sport-Specific Agility and Change of Direction in Water Polo: The Reliability and Validity of Two Newly Developed Tests.

ABSTRACT: Dong, L, Paradelo, D, Delorme, A, Oliveira, J, Parillo, B, Croteau, F, Romeas, T, Dubé, E, Bieuzen, F, Billaut, F, and Berryman, N. Sport-specific agility and change of direction in water polo: The reliability and validity of two newly developed tests. J Strength Cond Res 35(12S): S111-S118, 2021-There is a gap in water-based agility testing that considers both the change-of-direction (COD) and perceptive-reactive components of agility. This study sought to develop easily implementable, sport-specific in-water agility tests for water polo and to verify the reliability and validity of these new tests: the in-water Stop and Go (SG) and Jump and Go (JG). Female water polo athletes at the Senior (n = 12, age = 22.1 ± 2.1 years), Junior (n = 19, age = 18.5 ± 1.0 years), and Youth (n = 11, age = 16.5 ± 0.8 years) national levels performed 3 trials of each of the SG, JG, and the existing Functional Test for Agility Performance (FTAP). Senior athletes performed an additional experimental session to assess reliability parameters. Relative reliability for agility and COD versions of the SG and JG was high or very high (intraclass correlation coefficient [ICC] = 0.76-0.95). For construct validity analyses, significant between-group differences for each of the new tests (p < 0.05) were found. In contrast, the FTAP was moderately reliable (ICC = 0.57) and was unsuccessful in discriminating between playing levels. Considering the favorable metrological properties of the SG and JG, their fidelity to in-game demands, and their accessible setups, these new tests represent viable options to implement at grassroots and elite levels for the assessment and training of water polo-specific agility.

Risk Factors for Shoulder Injuries in Water Polo: a Cohort Study.

BACKGROUND: Very limited investigations have been conducted exploring risk factors for injury in water polo players. A gap remains in the literature regarding identification of variables that should be considered as part of player screening evaluations. PURPOSE: To estimate whether previous injury, changes in strength, range of motion (ROM) or upward scapular rotation (UR) are related to shoulder injuries in water polo players. STUDY DESIGN: Descriptive cohort study. METHODS: Thirty-nine international-level players participated (19 males). Shoulder internal (IR) and external rotation (ER) peak torque was measured using an isokinetic device (CONtrex MJ). Shoulder ROM was measured passively using standard goniometry. Scapular UR was measured using a laser digital inclinometer. At baseline players were divided into groups: those with and without previous shoulder injuries. Independent t-tests and Mann-Whitney U tests were used to compare the study variables between groups. After nine months, a second analysis compared the same athletes, who were then grouped by those who had or had not sustained new injuries. Effect sizes were calculated with a Hedge's g. Chi squared analysis compared proportion of injured players with and without previous injury. RESULTS: Eighteen participants (46%) had previous injuries at baseline. Players with a previous injury showed higher peak torques for IR (0.62±0.15 vs 0.54±0.13N/kg, p=0.04, g=0.60); larger loss of IR ROM (9.9±9.1 vs 4.1±7.5°, p=0.04, g=0.68), but no statistical difference in UR (p=0.70). After nine months, there were no statistical strength differences between groups. Loss of IR ROM was significantly higher in the injured group (9.8±9.8 vs 4.0±6.7°, p=0.04, g=0.68), as well as UR (13.0±3.0 vs 10.4±3.3°, p=0.01, g=0.81). History of previous injury was significantly related to developing a new injury (OR 6.5, p=0.02). Logistic regression found previous injury and UR most important contributors to injury risk. CONCLUSIONS: Previous injury, changes in IR ROM and UR are related to new shoulder injuries in water polo, but further variables such as rest, training load, or psychosocial factors may explain the incidence of new injuries. LEVEL OF EVIDENCE: Level 3.