Martin Lewis

Movement velocity as a measure of exercise intensity in three lower limb exercises

ABSTRACT The purpose of this study was to investigate the relationship between movement velocity and relative load in three lower limbs exercises commonly used to develop strength: leg press, full squat and half squat. The percentage of one repetition maximum (%1RM) has typically been used as the main parameter to control resistance training; however, more recent research has proposed movement velocity as an alternative. Fifteen participants performed a load progression with a range of loads until they reached their 1RM. Maximum instantaneous velocity (Vmax) and mean propulsive velocity (MPV) of the knee extension phase of each exercise were assessed. For all exercises, a strong relationship between Vmax and the %1RM was found: leg press (r2adj = 0.96; 95% CI for slope is [−0.0244, −0.0258], P < 0.0001), full squat (r2adj = 0.94; 95% CI for slope is [−0.0144, −0.0139], P < 0.0001) and half squat (r2adj = 0.97; 95% CI for slope is [−0.0135, −0.00143], P < 0.0001); for MPV, leg press (r2adj = 0.96; 95% CI for slope is [−0.0169, −0.0175], P < 0.0001, full squat (r2adj = 0.95; 95% CI for slope is [−0.0136, −0.0128], P < 0.0001) and half squat (r2adj = 0.96; 95% CI for slope is [−0.0116, 0.0124], P < 0.0001). The 1RM was attained with a MPV and Vmax of 0.21 ± 0.06 m s−1 and 0.63 ± 0.15 m s−1, 0.29 ± 0.05 m s−1 and 0.89 ± 0.17 m s−1, 0.33 ± 0.05 m s−1 and 0.95 ± 0.13 m s−1 for leg press, full squat and half squat, respectively. Results indicate that it is possible to determine an exercise-specific %1RM by measuring movement velocity for that exercise.

Optimisation of a machine learning algorithm in human locomotion using principal component and discriminant function analyses

Assessment methods in human locomotion often involve the description of normalised graphical profiles and/or the extraction of discrete variables. Whilst useful, these approaches may not represent the full complexity of gait data. Multivariate statistical methods, such as Principal Component Analysis (PCA) and Discriminant Function Analysis (DFA), have been adopted since they have the potential to overcome these data handling issues. The aim of the current study was to develop and optimise a specific machine learning algorithm for processing human locomotion data. Twenty participants ran at a self-selected speed across a 15m runway in barefoot and shod conditions. Ground reaction forces (BW) and kinematics were measured at 1000 Hz and 100 Hz, respectively from which joint angles (°), joint moments (N.m.kg-1) and joint powers (W.kg-1) for the hip, knee and ankle joints were calculated in all three anatomical planes. Using PCA and DFA, power spectra of the kinematic and kinetic variables were used as a training database for the development of a machine learning algorithm. All possible combinations of 10 out of 20 participants were explored to find the iteration of individuals that would optimise the machine learning algorithm. The results showed that the algorithm was able to successfully predict whether a participant ran shod or barefoot in 93.5% of cases. To the authors’ knowledge, this is the first study to optimise the development of a machine learning algorithm.

The effect of accounting for biarticularity in hip flexor and hip extensor joint torque representations

Subject-specific torque-driven models have ignored biarticular effects at the hip. The aim of this study was to establish the contribution of monoarticular hip flexors and hip extensors to total hip flexor and total hip extensor joint torques for an individual and to investigate whether torque-driven simulation models should consider incorporating biarticular effects at the hip joint. Maximum voluntary isometric and isovelocity hip flexion and hip extension joint torques were measured for a single participant together with surface electromyography. Single-joint and two-joint representations were fitted to the collected torque data and used to determine the maximum voluntary joint torque capacity. When comparing two-joint and single-joint representations, the single-joint representation had the capacity to produce larger maximum voluntary hip flexion torque (larger by around 9% of maximum torque) and smaller maximum voluntary hip extension torque (smaller by around 33% of maximum torque) with the knee extended. Considering the range of kinematics found for jumping movements, the single-joint hip flexors had the capacity to produce around 10% additional torque, while the single joint hip extensors had about 70% of the capacity of the two-joint representation. Two-joint representations may overcome an over-simplification of single-joint representations by accounting for biarticular effects, while building on the strength of determining subject-specific parameters from measurements on the participant.

The Importance of Biarticular Torque Generators in Torque-Driven Simulation Models

Forward-dynamics computer simulation models of human movement are typically driven by individual muscle models, or torque generators. In muscle-driven models, muscle parameters are typically determined from experimental data in the literature. While in torque-driven models, subject-specific torque parameters can be determined from torque measurements collected on an isovelocity dynamometer. Such a method avoids some of the errors encountered with individual muscle models by determining strength parameters directly from torque measurements. The disadvantage of existing subject-specific torque generator models over individual muscle models has been that the torque exerted at a joint has been represented by a function of the kinematics of the primary joint. As such torque generator models may not accurately represent the torques exerted by biarticular muscles where the kinematics of a primary and secondary joint may be important.Copyright