Effects of two neuromuscular training programs on running biomechanics with load carriage: a study protocol for a randomised controlled trial

Comparing shallow, deep, and transfer learning in predicting joint moments in running

Joint moments are commonly calculated in biomechanics research and provide an indirect measure of muscular behaviors and joint loads. However, joint moments cannot be easily quantified clinically or in the field, primarily due to challenges measuring ground reaction forces outside the laboratory. The present study aimed to compare the accuracy of three different machine learning (ML) techniques - functional regression [ ML_fregress ], a deep neural network (DNN) built from scratch [ ML_DNN ], and transfer learning [ ML_TL ], in predicting joint moments during running. Data for this study came from an open-source dataset and two studies on running with and without external loads. Three-dimensional (3D) joint moments of the hip, knee, and ankle, were derived using inverse dynamics. 3D joint angle, velocity, and acceleration of the three joints served as predictors for each of the three ML techniques. Prediction performance was generally the best using ML_DNN, and the worse using ML_fregress. Absolute predictive performance was the best for sagittal plane moments, which ranged from a RMSE of 0.16 Nm/kg at the ankle using ML_DNN, to a RMSE of 0.49Nm/kg at the knee using ML_fregress. ML_DNN resulted in the greatest improvement in relative prediction performance (relRMSE) by 20% compared to ML_fregress for the ankle adduction-abduction moment. DNN with or without transfer learning was superior in predicting joint moments using kinematic inputs compared to functional regression. Synergizing ML with kinematic inputs has the potential to solve the constraints of obtaining high fidelity biomechanics data normally only possible during laboratory studies.

Optimization of the Critical Speed Concept for Tactical Professionals: A Brief Review

Tactical professionals often depend on their physical ability and fitness to perform and complete occupational tasks to successfully provide public services or survive on the battlefield. Critical speed (CS), or maximal aerobic steady-state, is a purported measure that predicts performance, prescribes exercise, and detects training adaptions with application to tactical professionals. The CS concept has the versatility to adapt to training with load carriage as an integrated bioenergetic system approach for assessment. The aims of this review are to: (1) provide an overview of tactical populations and the CS concept; (2) describe the different methods and equipment used in CS testing; (3) review the literature on CS associated with tactical occupational tasks; and (4) demonstrate the use of CS-derived exercise prescriptions for tactical populations.

Effect of Plyometric Training on Speed and Change of Direction Ability in Elite Field Hockey Players

This study investigated the effects of two plyometric training protocols on sprint and change of direction (COD) performance in elite hockey players. A parallel-group randomized controlled trial design was used and seventeen elite male and female field hockey players were randomly allocated into either low-to-high (L-H, n = 8) or high-to-low (H-L, n = 9) training groups. Each group performed separate variations of the drop jump exercise twice weekly for six weeks, with an emphasis on either jump height (L-H) or drop height (H-L). Performance variables assessed included sprint times over 10 m and 20 m, as well as 505 time. A two-way repeated measures analysis of variance was performed and Cohen's d effect sizes (ESs) were calculated. The H-L group displayed a significant small ES improvement from baseline to post-training in the 10 m sprint (1.893 ± 0.08 s pre vs. 1.851 ± 0.06 s post) (ES = -0.44) (p < 0.05). Differences between groups for 10 m and 20 m sprint performance failed to reach statistical significance, and no significant differences were observed within or between groups for 505 time. These findings highlight the difficulty in substantially enhancing speed and COD ability in highly trained athletic populations through the addition of a low volume, short duration plyometric training protocol.

Validity of Critical Velocity Concept for Weighted Sprinting Performance

We investigated the validity of a recently developed equation for predicting sprinting times of various tactical loads based upon the performance of a running 3-min all-out exercise test (3MT). Thirteen recreationally trained participants completed the running 3MT to determine critical velocity (CV) and finite running capacity for running velocities exceeding CV (D'). Two subsequent counterbalanced loaded sprints of 800 and 1000 m distances with 20 and 15% of their body mass, respectively, were evaluated. Estimated times (t, sec) for running 800 and 1000 m with a tactical load was derived using t = (D - D')/CV. Critical velocity adjusted for an added load using the following regression equation: original CV + (-0.0638 × %load) + 0.6982, D was 800 or 1000 m, and whole percentage load was ~15 or 20% of the participant's body mass. From the 3MT, CV (3.80 ± 0.5 m·s^-1) and D' (200 ± 49.88 m) values were determined. The typical error of predicting actual times for the 800 and 1000 m loaded sprints were 5.6 and 10.1 s, with corresponding ICCs of 0.95 and 0.87, and coefficient of variations of 2.9 and 4.3%. The effect size differences between estimated and actual sprint times were small (0.27) and moderate (0.60) for 800 and 1000 m, respectively. The adjustment to CV through the regression equation yields small to moderate overestimates of maximally loaded sprint times for distances of 800 and 1000 m. Whether such errors remain pervasive for prescribing high-intensity interval training is unclear and requires further investigation.

Joint-level energetics differentiate isoinertial from speed-power resistance training-a Bayesian analysis

Background

There is convincing evidence for the benefits of resistance training on vertical jump improvements, but little evidence to guide optimal training prescription. The inability to detect small between modality effects may partially reflect the use of ANOVA statistics. This study represents the results of a sub-study from a larger project investigating the effects of two resistance training methods on load carriage running energetics. Bayesian statistics were used to compare the effectiveness of isoinertial resistance against speed-power training to change countermovement jump (CMJ) and squat jump (SJ) height, and joint energetics.

Methods

Active adults were randomly allocated to either a six-week isoinertial (n = 16; calf raises, leg press, and lunge), or a speed-power training program (n = 14; countermovement jumps, hopping, with hip flexor training to target pre-swing running energetics). Primary outcome variables included jump height and joint power. Bayesian mixed modelling and Functional Data Analysis were used, where significance was determined by a non-zero crossing of the 95% Bayesian Credible Interval (CrI).

Results

The gain in CMJ height after isoinertial training was 1.95 cm (95% CrI [0.85–3.04] cm) greater than the gain after speed-power training, but the gain in SJ height was similar between groups. In the CMJ, isoinertial training produced a larger increase in power absorption at the hip by a mean 0.018% (equivalent to 35 W) (95% CrI [0.007–0.03]), knee by 0.014% (equivalent to 27 W) (95% CrI [0.006–0.02]) and foot by 0.011% (equivalent to 21 W) (95% CrI [0.005–0.02]) compared to speed-power training.

Discussion

Short-term isoinertial training improved CMJ height more than speed-power training. The principle adaptive difference between training modalities was at the level of hip, knee and foot power absorption.