Machine learning and statistical prediction of fastball velocity with biomechanical predictors.

In recent years, one of the most important factors for success among baseball pitchers is fastball velocity. The purpose of this study was to (1) to develop statistical and machine learning models of fastball velocity, (2) to identify the strongest predictors of fastball velocity, and (3) to compare the models' prediction performances. Three dimensional biomechanical analyses were performed on high school (n = 165) and college (n = 62) baseball pitchers. A total of 16 kinetic and kinematic predictors from the entire pitching sequence were included in regression and machine learning models. All models were internally validated through ten-fold cross-validation. Model performance was evaluated through root mean square error (RMSE) and calibration with 95% confidence intervals. Gradient boosting machines demonstrated the best prediction performance [RMSE: 0.34; Calibration: 1.00 (95% CI: 0.999, 1.001)], while regression demonstrated the greatest prediction error [RMSE: 2.49; Calibration: 1.00 (95% CI: 0.85, 1.15)]. Maximum elbow extension velocity (relative influence: 19.3%), maximum humeral rotation velocity (9.6%), maximum lead leg ground reaction force resultant (9.1%), trunk forward flexion at release (7.9%), time difference of maximum pelvis rotation velocity and maximum trunk rotation velocity (7.8%) demonstrated the greatest influence on pitch velocity. Gradient boosting machines demonstrated better calibration and reduced RMSE compared to regression. The influence of lead leg ground reaction force resultant and trunk and arm kinematics on pitch velocity demonstrates the interdependent relationship of the entire kinetic chain during the pitching motion. Coaches, players, and performance professionals should focus on the identified metrics when designing pitch velocity improvement programs.

Verification of High-Rate Vertical Loading Laboratory Skeletal Fractures by Comparison with Theater Injury Patterns.

For the current study, an existing theater injury data set was compared to component and whole body experiments meant to replicate the theater high rate vertical loading environment. The theater injury data set was derived from real world events that were within the design range of the Warrior Injury Assessment Manikin. A qualitative and quantitative assessment of the whole body fracture patterns was developed to determine whether the laboratory loading was correctly representing the resulting injuries seen in theater Underbody Blast (UBB) events. Results indicated that most of the experimental test fracture patterns were similar to the theater injuries for Abbreviated Injury Scale body regions of interest (lower extremities, pelvis, and spine); however, some of the body regions had higher similarity scores compared to others. Whole body fracture distribution was less similar than the component tests because of differences in injury distributions. The lower extremity whole body similarity was lower than spine and pelvis similarity. This analysis was able to identify some experimental tests that might not represent theater loading. In conclusion, this analysis confirmed that some laboratory testing produced skeletal injury patterns that are seen in comparable theater UBB events.

Epidemiological profile of pain and non-steroid anti-inflammatory drug use in collegiate athletes in the United States.

BACKGROUND: Although athletic endeavours are associated with a high amount of physical stress and injury, the prevalence of pain is underreported in the sports medicine literature with only a few studies reporting pain on collegiate athletes or exploring sex difference of pain. Impact of pain on athlete availability, training and performance can be mitigated when key epidemiological information is used to inform adequate pain management strategies. This study aims to 1) provide an epidemiological profile of self-reported pain experienced by the National Collegiate Athletic Association (NCAA) athletes by sex during the first half of the 2019 season, 2) describe their self-reported non-steroidal anti-inflammatory drug (NSAID) use. METHODS: Online survey was completed by athletes at three NCAA institutions from 1 August to 30 September 2019. Descriptive statistics were used to describe player demographic data, self-reported pain and self-reported NSAID use. Pain incidence proportion were calculated. RESULTS: Two hundred thirty female athletes and 83 male athletes completed the survey. Self-reported pain incidence proportion for female athletes was 45.0 (95% CI 41.5-48.5) vs 34.9 (95% CI 29.4-40.4) for male athletes. Majority of the athletes did not report pain (55% female vs 62% male) during the first half of the 2019 season. Female athletes reported pain in their back (35%), knee (26%), and ankle/foot (23%) whilst male athletes reported pain in their knee (35%), back (28%), and shoulder (24%). Of all athletes, 28% female vs 20% male athletes reported currently taking NSAIDs. Of athletes that reported pain, 46% female vs 38% male athletes currently took NSAIDs. 70% female vs 61% male athletes self-purchased NSAIDs, and 40% female vs 55% male athletes consumed alcohol. CONCLUSIONS: Half of female athletes and one in three male athletes reported pain. Most commonly back, knee and foot/ankle pain and knee, back and shoulder pain was reported in female and male athletes respectively. One in four female athletes and one in five male athletes use NSAIDs for pain or prophylactic purpose. Majority self-purchase these medications indicating need for health literacy interventions to mitigate potential adverse effects.