Maximum velocity and leg-specific ground reaction force production change with radius during flat curve sprinting.

Humans attain slower maximum velocity (vmax) on curves versus straight paths, potentially due to centripetal ground reaction force (GRF) production, and this depends on curve radius. Previous studies found GRF production differences between an athlete's inside versus outside leg relative to the center of the curve. Further, sprinting clockwise (CW) versus counterclockwise (CCW) slows vmax. We determined vmax, step kinematics and individual leg GRF on a straight path and on curves with 17.2 and 36.5 m radii for nine (8 male, 1 female) competitive sprinters running CW and CCW and compared vmax with three predictive models. We combined CW and CCW directions and found that vmax slowed by 10.0±2.4% and 4.1±1.6% (P<0.001) for the 17.2 and 36.5 m radius curves versus the straight path, respectively. vmax values from the predictive models were up to 3.5% faster than the experimental data. Contact length was 0.02 m shorter and stance average resultant GRF was 0.10 body weights (BW) greater for the 36.5 versus 17.2 m radius curves (P<0.001). Stance average centripetal GRF was 0.10 BW greater for the inside versus outside leg (P<0.001) on the 36.5 m radius curve. Stance average vertical GRF was 0.21 BW (P<0.001) and 0.10 BW (P=0.001) lower for the inside versus outside leg for the 17.2 and 36.5 m radius curves, respectively. For a given curve radius, vmax was 1.6% faster in the CCW compared with CW direction (P=0.003). Overall, we found that sprinters change contact length and modulate GRFs produced by their inside and outside legs as curve radius decreases, potentially limiting vmax.

Predicting continuous ground reaction forces from accelerometers during uphill and downhill running: a recurrent neural network solution.

BACKGROUND: Ground reaction forces (GRFs) are important for understanding human movement, but their measurement is generally limited to a laboratory environment. Previous studies have used neural networks to predict GRF waveforms during running from wearable device data, but these predictions are limited to the stance phase of level-ground running. A method of predicting the normal (perpendicular to running surface) GRF waveform using wearable devices across a range of running speeds and slopes could allow researchers and clinicians to predict kinetic and kinematic variables outside the laboratory environment. PURPOSE: We sought to develop a recurrent neural network capable of predicting continuous normal (perpendicular to surface) GRFs across a range of running speeds and slopes from accelerometer data. METHODS: Nineteen subjects ran on a force-measuring treadmill at five slopes (0°, ±5°, ±10°) and three speeds (2.5, 3.33, 4.17 m/s) per slope with sacral- and shoe-mounted accelerometers. We then trained a recurrent neural network to predict normal GRF waveforms frame-by-frame. The predicted versus measured GRF waveforms had an average ± SD RMSE of 0.16 ± 0.04 BW and relative RMSE of 6.4 ± 1.5% across all conditions and subjects. RESULTS: The recurrent neural network predicted continuous normal GRF waveforms across a range of running speeds and slopes with greater accuracy than neural networks implemented in previous studies. This approach may facilitate predictions of biomechanical variables outside the laboratory in near real-time and improves the accuracy of quantifying and monitoring external forces experienced by the body when running.

Sacral acceleration can predict whole-body kinetics and stride kinematics across running speeds.

BACKGROUND: Stress fractures are injuries caused by repetitive loading during activities such as running. The application of advanced analytical methods such as machine learning to data from multiple wearable sensors has allowed for predictions of biomechanical variables associated with running-related injuries like stress fractures. However, it is unclear if data from a single wearable sensor can accurately estimate variables that characterize external loading during running such as peak vertical ground reaction force (vGRF), vertical impulse, and ground contact time. Predicting these biomechanical variables with a single wearable sensor could allow researchers, clinicians, and coaches to longitudinally monitor biomechanical running-related injury risk factors without expensive force-measuring equipment. PURPOSE: We quantified the accuracy of applying quantile regression forest (QRF) and linear regression (LR) models to sacral-mounted accelerometer data to predict peak vGRF, vertical impulse, and ground contact time across a range of running speeds. METHODS: Thirty-seven collegiate cross country runners (24 females, 13 males) ran on a force-measuring treadmill at 3.8-5.4 m/s while wearing an accelerometer clipped posteriorly to the waistband of their running shorts. We cross-validated QRF and LR models by training them on acceleration data, running speed, step frequency, and body mass as predictor variables. Trained models were then used to predict peak vGRF, vertical impulse, and contact time. We compared predicted values to those calculated from a force-measuring treadmill on a subset of data (n = 9) withheld during model training. We quantified prediction accuracy by calculating the root mean square error (RMSE) and mean absolute percentage error (MAPE). RESULTS: The QRF model predicted peak vGRF with a RMSE of 0.150 body weights (BW) and MAPE of 4.27  ±  2.85%, predicted vertical impulse with a RMSE of 0.004 BW*s and MAPE of 0.80  ±  0.91%, and predicted contact time with a RMSE of 0.011 s and MAPE of 4.68  ±  3.00%. The LR model predicted peak vGRF with a RMSE of 0.139 BW and MAPE of 4.04  ±  2.57%, predicted vertical impulse with a RMSE of 0.002 BW*s and MAPE of 0.50  ±  0.42%, and predicted contact time with a RMSE of 0.008 s and MAPE of 3.50  ±  2.27%. There were no statistically significant differences between QRF and LR model prediction MAPE for peak vGRF (p = 0.549) or vertical impulse (p = 0.073), but the LR model's MAPE for contact time was significantly lower than the QRF model's MAPE (p = 0.0497). CONCLUSIONS: Our findings indicate that the QRF and LR models can accurately predict peak vGRF, vertical impulse, and contact time (MAPE < 5%) from a single sacral-mounted accelerometer across a range of running speeds. These findings may be beneficial for researchers, clinicians, or coaches seeking to monitor running-related injury risk factors without force-measuring equipment.

Low-pass filter cutoff frequency affects sacral-mounted inertial measurement unit estimations of peak vertical ground reaction force and contact time during treadmill running.

Inertial measurement units (IMUs) are popular tools for estimating biomechanical variables such as peak vertical ground reaction force (GRFv) and foot-ground contact time (tc), often by using multiple sensors or predictive models. Despite their growing use, little is known about the effects of varying low-pass filter cutoff frequency, which can affect the magnitude of force-related dependent variables, the accuracy of IMU-derived metrics, or if simpler methods for such estimations exist. The purpose of this study was to investigate the effects of varying low-pass filter cutoff frequency on the correlation of IMU-derived peak GRFv and tc to gold-standard lab-based measurements. Thirty National Collegiate Athletics Association Division 1 cross country runners ran on an instrumented treadmill at a range of speeds while outfitted with a sacral-mounted IMU. A simple method for estimating peak GRFv from the IMU was implemented by multiplying the IMU's vertical acceleration by the runner's body mass. Data from the IMU were low-pass filtered with 5, 10, and 30 Hz cutoffs. Pearson correlation coefficients were used to determine how well the IMU-derived estimates matched gold-standard biomechanical estimations. Correlations ranged from very weak to moderate for peak GRFv and tc. For peak GRFv, the 10 Hz low-pass filter cutoff performed best (r = 0.638), while for tc the 5 Hz cut-off performed best (r = 0.656). These results suggest that IMU-derived estimates of force and contact time are influenced by the low-pass filter cutoff frequency. Further investigations are needed to determine the optimal low-pass filter cutoff frequency or a different method to accurately estimate force and contact time is suggested.

Added lower limb mass does not affect biomechanical asymmetry but increases metabolic power in runners with a unilateral transtibial amputation.

PURPOSE: we determined the metabolic and biomechanical effects of adding mass to the running-specific prosthesis (RSP) and biological foot of individuals with a unilateral transtibial amputation (TTA) during running. METHODS: 10 individuals (8 males, 2 females) with a TTA ran on a force-measuring treadmill at 2.5 m/s with 100 g and 300 g added to their RSP alone or to their RSP and biological foot while we measured their metabolic rates and calculated peak vertical ground reaction force (vGRF), stance-average vGRF, and step time symmetry indices. RESULTS: for every 100 g added to the RSP alone, metabolic power increased by 0.86% (p = 0.007) and for every 100 g added to the RSP and biological foot, metabolic power increased by 1.74% ([Formula: see text] 0.001) during running. Adding mass had no effect on peak vGRF (p = 0.102), stance-average vGRF (p = 0.675), or step time (p = 0.413) symmetry indices. We also found that the swing time of the affected leg was shorter than the unaffected leg across conditions ([Formula: see text] 0.007). CONCLUSIONS: adding mass to the lower limbs of runners with a TTA increased metabolic power by more than what has been reported for those without an amputation. We found no effect of added mass on biomechanical asymmetry, but the affected leg had consistently shorter swing times than the unaffected leg. This suggests that individuals with a TTA maintain asymmetries despite changes in RSP mass and that lightweight prostheses could improve performance by minimizing metabolic power without affecting asymmetry.

Calcaneus range of motion underestimated by markers on running shoe heel.

BACKGROUND: The measurement of rearfoot kinematics by placing reflective markers on the shoe heel assumes its motion is identical to the foot's motion. Studies have compared foot and shoe kinematics during running but with conflicting results. The primary purpose of this study was to compare shoe and calcaneus three-dimensional range of motion during running. A secondary purpose was to determine the effect of a less rigid heel counter on tibia motion. RESEARCH QUESTION: Do markers placed on the shoe heel accurately represent calcaneus kinematics during running? METHODS: Three-dimensional coordinate data were collected on 14 subjects (M/F: 9/5) who ran on an instrumented treadmill at 3.35 m/s under four conditions: modified/intact neutral shoes, and modified/intact support shoes. Shoes were modified by placing holes through the heel to allow for shoe heel and calcaneus coordinate data to be collected simultaneously via reflective markers on the shoe and on the skin of the heel within the shoe. Calcaneus, shoe heel, and tibia ROM were calculated from 0 to 50% stance phase and compared across shoe conditions. RESULTS: Calcaneal frontal plane ROM was significantly greater than neutral and support shoe heel ROM (p < 0.001). Calcaneus ROM was also significantly greater than shoe heel ROM in the transverse (p < 0.001) and sagittal (p < 0.001) planes. No change in tibial transverse plane ROM was observed (p = 0.346) across shoe heel conditions. SIGNIFICANCE: Shoe markers significantly underestimated calcaneus ROM across all planes of motion. These findings suggest calcaneus kinematics cannot be accurately measured with markers placed solely on the shoe heel. Additionally, the required modifications to the shoe's heel had no effect on tibia ROM in the transverse plane.

Stroller running: Energetic and kinematic changes across pushing methods.

OBJECTIVE: Running with a stroller provides an opportunity for parents to exercise near their child and counteract health declines experienced during early parenthood. Understanding biomechanical and physiological changes that occur when stroller running is needed to evaluate its health impact, yet the effects of stroller running have not been clearly presented. Here, three commonly used stroller pushing methods were investigated to detect potential changes in energetic cost and lower-limb kinematics. METHODS: Sixteen individuals (M/F: 10/6) ran at self-selected speeds for 800m under three stroller conditions (2-Hands, 1-Hand, and Push/Chase) and an independent running control. RESULTS: A significant decrease in speed (p = 0.001) and stride length (p<0.001) was observed between the control and stroller conditions, however no significant change in energetic cost (p = 0.080) or heart rate (p = 0.393) was observed. Additionally, pushing method had a significant effect on speed (p = 0.001) and stride length (p<0.001). CONCLUSIONS: These findings suggest that pushing technique influences stroller running speed and kinematics. These findings suggest specific fitness effects may be achieved through the implementation of different pushing methods.