While shoulder injuries resulting from the bench press exercise are commonly reported, no biomechanical evidence for lowering injury risk is currently available. Therefore, the aim of the present study was to compare musculoskeletal shoulder loads and potential injury risk during several bench press variations. Ten experienced strength athletes performed 21 technical variations of the barbell bench press, including variations in grip width of 1,1.5 and 2 bi-acromial widths (BAW), shoulder abduction angles of 45°, 70° and 90°, and scapula poses including neutral, retracted, and released conditions. Motions and forces were recorded by an opto-electronic measurement system and an instrumented barbell. An OpenSim musculoskeletal shoulder model was employed to estimate joint reaction forces in the glenohumeral and acromioclavicular joints. Time-series of joint reaction forces were compared between techniques by statistical non-parametric mapping. Results showed that narrower grip widths of <1.5 BAW decreased acromioclavicular compression (p < 0.05), which may decrease the risk for distal clavicular osteolysis. Moreover, scapula retraction, as well as a grip width of <1.5 BAW (p < 0.05), decreased glenohumeral posterior shear force components and rotator cuff activity and may decrease the risk for glenohumeral instability and rotator cuff injuries. Furthermore, results showed that mediolaterally exerted barbell force components varied considerably between athletes and largely affected shoulder reaction forces. It can be concluded that the grip width, scapula pose and mediolateral exerted barbell forces during the bench press influence musculoskeletal shoulder loads and the potential injury risk. Results of this study can contribute to safer bench press training guidelines.@en

Muscle overload injuries in strength training might be prevented by providing personalized feedback about muscle load during a workout. In the present study, a new muscle load feedback application, which monitors and visualizes the loading of specific muscle groups, was developed in collaboration with the fitness company Gymstory. The aim of the present study was to examine the effectiveness of this feedback application in managing muscle load balance, muscle load level, and muscle soreness, and to evaluate how its actual use was experienced. Thirty participants were randomly distributed into ‘control’, ‘partial feedback’, and ‘complete feedback’ groups and monitored for eight workouts using the automatic exercise tracking system of Gymstory. The control group received no feedback, while the partial feedback group received a visualization of their estimated cumulative muscle load after each exercise, and the participants in the complete feedback group received this visualization together with suggestions for the next exercise to target muscle groups that had not been loaded yet. Generalized estimation equations (GEEs) were used to compare muscle load balance and soreness, and a one-way ANOVA was used to compare user experience scores between groups. The complete feedback group showed a significantly better muscle load balance (β = −18.9; 95% CI [−29.3, −8.6]), adhered better to the load suggestion provided by the application (significant interactions), and had higher user experience scores for Attractiveness (p = 0.036), Stimulation (p = 0.031), and Novelty (p = 0.019) than the control group. No significant group differences were found for muscle soreness. Based on these results, it was concluded that personal feedback about muscle load in the form of a muscle body map in combination with exercise suggestions can effectively guide strength training practitioners towards certain load levels and more balanced cumulative muscle loads. This application has potential to be applied in strength training practice as a training tool and may help in preventing muscle overload.@en

Muscle force analysis can be essential for injury risk estimation and performance enhancement in sports like strength training. However, current methods to record muscle forces including electromyography or marker-based measurements combined with a musculoskeletal model are time-consuming and restrict the athlete's natural movement due to equipment attachment. Therefore, the feasibility and validity of a more applicable method, requiring only a single standard camera for the recordings, combined with a deep-learning model and musculoskeletal model is evaluated in the present study during upper-body strength exercises performed by five athletes. Comparison of muscle forces obtained by the single camera driven model against those obtained from a state-of-the art marker-based driven musculoskeletal model revealed strong to excellent correlations and reasonable RMSD's of 0.4–2.1% of the maximum force (Fmax) for prime movers, and weak to strong correlations with RMSD's of 0.4–0.7% Fmax for stabilizing and secondary muscles. In conclusion, a single camera deep-learning driven model is a feasible method for muscle force analysis in a strength training environment, and first validity results show reasonable accuracies, especially for prime mover muscle forces. However, it is evident that future research should investigate this method for a larger sample size and for multiple exercises.

@en