Prediction of Achilles Tendon Force During Common Motor Tasks From Markerless Video.

Remodeling of the Achilles tendon (AT) is partly driven by its mechanical environment. AT force can be estimated with neuromusculoskeletal (NMSK) modeling; however, the complex experimental setup required to perform the analyses confines use to the laboratory. We developed task-specific long short-term memory (LSTM) neural networks that employ markerless video data to predict the AT force during walking, running, countermovement jump, single-leg landing, and single-leg heel rise. The task-specific LSTM models were trained on pose estimation keypoints and corresponding AT force data from 16 subjects, calculated via an established NMSK modeling pipeline, and cross-validated using a leave-one-subject-out approach. As proof-of-concept, new motion data of one participant was collected with two smartphones and used to predict AT forces. The task-specific LSTM models predicted the time-series AT force using synthesized pose estimation data with root mean square error (RMSE) ≤ 526 N, normalized RMSE (nRMSE) ≤ 0.21 , R 2 ≥ 0.81 . Walking task resulted the most accurate with RMSE = 189±62 N; nRMSE = 0.11±0.03 , R 2 = 0.92±0.04 . AT force predicted with smartphones video data was physiologically plausible, agreeing in timing and magnitude with established force profiles. This study demonstrated the feasibility of using low-cost solutions to deploy complex biomechanical analyses outside the laboratory.

Biceps femoris long head muscle and aponeurosis geometry in males with and without a history of hamstring strain injury.

OBJECTIVES: Hamstring strain injuries (HSIs) commonly affect the proximal biceps femoris long head (BFlh) musculotendinous junction. Biomechanical modeling suggests narrow proximal BFlh aponeuroses and large muscle-to-aponeurosis width ratios increase localized tissue strains and presumably risk of HSI. This study aimed to determine if BFlh muscle and proximal aponeurosis geometry differed between limbs with and without a history of HSI. METHODS: Twenty-six recreationally active males with (n = 13) and without (n = 13) a history of unilateral HSI in the last 24 months underwent magnetic resonance imaging of both thighs. BFlh muscle and proximal aponeurosis cross-sectional areas, length, volume, and interface area between muscle and aponeurosis were extracted. Previously injured limbs were compared to uninjured contralateral and control limbs for discrete variables and ratios, and along the relative length of tissues using statistical parametric mapping. RESULTS: Previously injured limbs displayed significantly smaller muscle-to-aponeurosis volume ratios (p = 0.029, Wilcoxon effect size (ES) = 0.43) and larger proximal BFlh aponeurosis volumes (p = 0.019, ES = 0.46) than control limbs with no history of HSI. No significant differences were found between previously injured and uninjured contralateral limbs for any outcome measure (p = 0.216-1.000, ES = 0.01-0.36). CONCLUSIONS: Aponeurosis geometry differed between limbs with and without a history of HSI. The significantly larger BFlh proximal aponeuroses and smaller muscle-to-aponeurosis volume ratios in previously injured limbs could alter the strain experienced in muscle adjacent to the musculotendinous junction during active lengthening. Future research is required to determine if geometric differences influence the risk of re-injury and whether they can be altered via targeted training.

Predicting Free Achilles Tendon Strain From Motion Capture Data Using Artificial Intelligence.

The Achilles tendon (AT) is sensitive to mechanical loading, with appropriate strain improving tissue mechanical and material properties. Estimating free AT strain is currently possible through personalized neuromusculoskeletal (NMSK) modeling; however, this approach is time-consuming and requires extensive laboratory data. To enable in-field assessments, we developed an artificial intelligence (AI) workflow to predict free AT strain during running from motion capture data. Ten keypoints commonly used in pose estimation algorithms (e.g., OpenPose) were synthesized from motion capture data and noise was added to represent real-world data obtained using video cameras. Two AI workflows were compared: (1) a Long Short-Term Memory (LSTM) neural network that predicted free AT strain directly (called LSTM only workflow); and (2) an LSTM neural network that predicted AT force which was subsequently converted to free AT strain using a personalized force-strain curve (called LSTM+ workflow). AI models were trained and evaluated using estimates of free AT strain obtained from a validated NMSK model with personalized AT force-strain curve. The effect of using different input features (position, velocity, and acceleration of keypoints, height and mass) on free AT strain predictions was also assessed. The LSTM+ workflow significantly improved the predictions of free AT strain compared to the LSTM only workflow (p < 0.001). The best free AT strain predictions were obtained using positions and velocities of keypoints as well as the height and mass of the participants as input, with average time-series root mean square error (RMSE) of 1.72±0.95% strain and r2 of 0.92±0.10, and peak strain RMSE of 2.20% and r2 of 0.54. In conclusion, we showed feasibility of predicting accurate free AT strain during running using low fidelity pose estimation data.

Throwing performance in water polo is related to in-water shoulder proprioception.

Water polo players require a high level of upper-extremity strength, flexibility and coordination to achieve a peak level of throwing performance. Increased levels of shoulder proprioceptive acuity, strength and range of motion (ROM) have been previously associated with higher sporting performance. A coach-rating scale, used to quantify an athlete's kicking proficiency in soccer; was adapted in the current study to measure each coach's subjective expert opinion regarding athletes' throwing mechanics, velocity, and accuracy. To examine this hypothesis shoulder proprioception acuity of 18 water polo players was measured both in-water and on-land using an AMEDA apparatus and correlated with coach-rated throwing performance and clinical measures of shoulder strength and ROM. There was a moderate positive correlation between the in-water and the on-land proprioception acuity (r = 0.47, p < 0.05). The in-water score showing a strong positive correlation with coach rated throwing mechanics (r = 0.68, p < 0.05) and velocity (r = 0.75, p = 0.02), suggesting that superior proprioception acuity contributed to fast, mechanically-efficient throwing. These findings support the notion that in-water proprioceptive acuity is an important determinant of the throwing performance achieved by water polo athletes and its measurement may be a valuable adjunct to current athlete screening.

Shoulder internal and external rotation strength and prediction of subsequent injury in water-polo players.

Water-polo players have greater isokinetic shoulder strength than age-matched controls. Due to the repetitive demands of throwing, however, water-polo players demonstrate an altered strength ratio, with greater internal rotation (IR) strength relative to external rotation (ER). The relationship between shoulder strength and risk of shoulder injury is unknown. In addition, the effect on test position for strength testing on the reliability of handheld dynamometry (HHD) in this population is not known. The aims were to determine the: (a) Inter-rater reliability of HHD testing of IR and ER strength in two positions: neutral and 90°abduction-90°ER (90-90) and (b) relationship between preseason shoulder strength and occurrence of future injury in sub-elite water-polo players. Two assessors measured shoulder IR and ER strength using HHD in 15 water-polo players across two testing days. Athletes were followed over a 6-month period, and injury was assessed and recorded by the team physiotherapist. Measurement of water-polo players' isometric IR and ER strength in the clinical setting had good to excellent inter-rater reliability; however, systematic error was observed in the neutral position but not the 90-90 position. Irrespective of testing position, the neutral and 90-90 test position showed a significant difference (P = 0.01) in absolute preseason IR and ER mean strength between prospectively injured and non-injured players. There was no significant difference in strength ratio or strength normalized for body mass index. These results suggest that preseason strength testing may help identify players at risk of in-season shoulder injury.