Machine Learning Approach for Pitch Type Classification Based on Pelvis and Trunk Kinematics Captured with Wearable Sensors.

The large stream of data from wearable devices integrated with sports routines has changed the traditional approach to athletes' training and performance monitoring. However, one of the challenges of data-driven training is to provide actionable insights tailored to individual training optimization. In baseball, the pitching mechanics and pitch type play an essential role in pitchers' performance and injury risk management. The optimal manipulation of kinematic and temporal parameters within the kinetic chain can improve the pitcher's chances of success and discourage the batter's anticipation of a particular pitch type. Therefore, the aim of this study was to provide a machine learning approach to pitch type classification based on pelvis and trunk peak angular velocity and their separation time recorded using wearable sensors (PITCHPERFECT). The Naive Bayes algorithm showed the best performance in the binary classification task and so did Random Forest in the multiclass classification task. The accuracy of Fastball classification was 71%, whilst the accuracy of the classification of three different pitch types was 61.3%. The outcomes of this study demonstrated the potential for the utilization of wearables in baseball pitching. The automatic detection of pitch types based on pelvis and trunk kinematics may provide actionable insight into pitching performance during training for pitchers of various levels of play.

Cross-Cultural Adaption and Validation of the Dutch Version of the Kerlan-Jobe Orthopaedic Clinic Questionnaire in Juvenile Baseball Pitchers.

Monitoring the performance and functional status of baseball pitchers' upper extremity is important in maintaining the athlete's health and performance. This study validated a Dutch translation of the original English Kerlan-Jobe Orthopaedic Clinic (KJOC) against the previously validated Disabilities of the Arm, Shoulder and Hand (DASH) and Western Ontario Shoulder Instability Index (WOSI) questionnaires in a group of talented juvenile Dutch baseball pitchers. Three times, from 2014-2016, 107 pitchers completed the Dutch KJOC, DASH and WOSI questionnaires. Participants' questionnaire scores were analysed for the whole group and the symptomatic player subgroup separately. Internal consistency, construct validity and ceiling and floor effects were examined. Cronbach's alpha was consistently above 0.8 for the three time periods for the whole group, and ranged between 0.62 and 0.86 for the symptomatic subgroup. Spearman's rank correlation coefficients ranged from 0.47 to 0.67 for the whole group and 0.32 to 0.99 for the symptomatic subgroup. No floor effects were observed in the scores of the KJOC and only a ceiling effect for the whole group (15.2%) at one time period. The Dutch version of the KJOC has shown acceptable internal consistency and construct validity and can be used to assess overhead athletes' shoulder and elbow functionality.

Individualised Ball Speed Prediction in Baseball Pitching Based on IMU Data.

Ball velocity is considered an important performance measure in baseball pitching. Proper pitching mechanics play an important role in both maximising ball velocity and injury-free participation of baseball pitchers. However, an individual pitcher's characteristics display individuality and may contribute to velocity imparted to the ball. The aim of this study is to predict ball velocity in baseball pitching, such that prediction is tailored to the individual pitcher, and to investigate the added value of the individuality to predictive performance. Twenty-five youth baseball pitchers, members of a national youth baseball team and six baseball academies in The Netherlands, performed ten baseball pitches with maximal effort. The angular velocity of pelvis and trunk were measured with IMU sensors placed on pelvis and sternum, while the ball velocity was measured with a radar gun. We develop three Bayesian regression models with different predictors which were subsequently evaluated based on predictive performance. We found that pitcher's height adds value to ball velocity prediction based on body segment rotation. The developed method provides a feasible and affordable method for ball velocity prediction in baseball pitching.

Influence of biomechanical models on joint kinematics and kinetics in baseball pitching.

In baseball pitching, biomechanical parameters have been linked to ball velocity and potential injury risk. However, although the features of a biomechanical model have a significant influence on the kinematics and kinetics of a motion, this influence have not been assessed for pitching. The aim of this study was to evaluate the choice of the trunk and shoulder features, by comparing two models using the same input. The models differed in thoraco-humeral joint definition (moving or fixed with the thorax), joint centre estimation, values of the inertial parameters and computational framework. One professional pitcher participated in the study. We found that the different features of the biomechanical models have a substantial influence on the kinematics and kinetics of the pitchers. With a fixed thoraco-humeral joint the peak average thorax angular velocity was delayed and underestimated by 17% and the shoulder internal rotation velocity was overestimated by 7%. The use of a thoraco-humeral joint fixed to the thorax will lead to an overestimation of the rotational power at the shoulder and will neglect the power produced by the forward and upward translation of the shoulder girdle. These findings have direct implications for the interpretation of shoulder muscle contributions to the pitch.

Knee Angle and Stride Length in Association with Ball Speed in Youth Baseball Pitchers.

The purpose of this study was to determine whether stride length and knee angle of the leading leg at foot contact, at the instant of maximal external rotation of the shoulder, and at ball release are associated with ball speed in elite youth baseball pitchers. In this study, fifty-two elite youth baseball pitchers (mean age 15.2 SD (standard deviation) 1.7 years) pitched ten fastballs. Data were collected with three high-speed video cameras at a frequency of 240 Hz. Stride length and knee angle of the leading leg were calculated at foot contact, maximal external rotation, and ball release. The associations between these kinematic variables and ball speed were separately determined using generalized estimating equations. Stride length as percentage of body height and knee angle at foot contact were not significantly associated with ball speed. However, knee angles at maximal external rotation and ball release were significantly associated with ball speed. Ball speed increased by 0.45 m/s (1 mph) with an increase in knee extension of 18 degrees at maximal external rotation and 19.5 degrees at ball release. In conclusion, more knee extension of the leading leg at maximal external rotation and ball release is associated with higher ball speeds in elite youth baseball pitchers.