Despite the presence of several different calculations of leg stiffness during hopping, little is known about how the methodologies produce differences in the leg stiffness. The purpose of this study was to directly compare K leg during hopping as calculated from three previously published computation methods. Ten male subjects hopped in place on two legs, at four frequencies (2.2, 2.6, 3.0, and 3.4 Hz). In this article, leg stiffness was calculated from the natural frequency of oscillation (method A), the ratio of maximal ground reaction force (GRF) to peak center of mass displacement at the middle of the stance phase (method B), and an approximation based on sine-wave GRF modeling (method C). We found that leg stiffness in all methods increased with an increase in hopping frequency, but K leg values using methods A and B were significantly higher than when using method C at all hopping frequencies. Therefore, care should be taken when comparing leg stiffness obtained by method C with those calculated by other methods.
Hiroaki Hobara, Koh Inoue, Yoshiyuki Kobayashi, and Toru Ogata
C. Mark Woodard, Margaret K. James, and Stephen P. Messier
Our purpose was to compare methods of calculating loading rate to the first peak vertical ground reaction force during walking and provide a rationale for the selection of a loading rate algorithm in the analysis of gait in clinical and research environments. Using vertical ground reaction force data collected from 15 older adults with symptomatic knee osteoarthritis and 15 healthy controls, we: (a) calculated loading rate as the first peak vertical force divided by the time from touchdown until the first peak; (b) calculated loading rate as the slope of the least squares regression line using vertical force and time as the dependent and independent variables, respectively; (c) calculated loading rate over discrete intervals using the Central Difference method; and (d) calculated loading rate using vertical force and lime data representing 20% and 90% of the first peak vertical force. The largest loading rate, which may be of greatest clinical importance, occurred when loading rates were calculated using the fewest number of data points. The Central Difference method appeared to maximize our ability to detect differences between healthy and pathologic cohorts. Finally, there was a strong correlation between methods, suggesting that all four methods are acceptable. However, if maximizing the chances of detecting differences between groups is of primary importance, the Central Difference method appears superior.
Fahim A. Salim, Fasih Haider, Dees Postma, Robby van Delden, Dennis Reidsma, Saturnino Luz, and Bert-Jan van Beijnum
Joshua Twaites, Richard Everson, Joss Langford, and Melvyn Hillsdon
Introduction: Data from wrist-worn accelerometers often has an inherent natural segmentation that reflects transitioning from one activity to another. The aim of this study was to develop an activity transition detection method to realize this natural segmentation. Methods: Data was gathered from 16 participants who wore triaxial wrist accelerometers in a lab-based protocol and 47 participants in a free-living protocol. Change point detection was used to create a method for detecting activity transitions. The agreement between observed and predicted transitions was assessed by the Matthews Correlation Coefficient (MCC), Root Mean Squared Error (RMSE), and two additional metrics created for this task; the Ratio of Minimum Mean Distance (RMMD) and the Ratio of Sensitivity (RoS). The effects of varying combinations of acceleration axes were also investigated to determine the most effective set of axes. A novel post-processing technique was developed to mitigate a major limitation identified in current transition detection methods. Results: The developed transition detection method achieved a MCC of 0.763, a RMSE of 3.17, a RoS of 2.40, and a RMMD of 3.21, outperforming existing techniques. The post-processing technique developed improved the performance of all methods when identifying transitions. It was found that using solely the y-axis (vertical acceleration) allowed for optimal performance. Conclusion: Change point detection is a valid method for identifying transitions in activity using wrist-worn accelerometer data. The new post processing technique developed improves the performance of transition detection methods.
Alberto Flórez-Pregonero, Matthew S. Buman, and Barbara E. Ainsworth
Background: Published accelerometer cut-points have limited accuracy in measuring sedentary (SED) and stationary time (STA) despite hip or wrist placement. Few studies have evaluated established cut-points to measure SED and STA in free-living settings. Methods: This study evaluated published uniaxial and triaxial cut-points of accelerometers and identified optimal cut-points to measure SED and STA. Twenty participants, ages 18–65, wore three ActiGraph GT3X+ (one on each wrist and the waist) and two GENEActiv accelerometers (one on each wrist) for one weekday and one weekend day during simultaneous direct observation of movement. ActiGraph uniaxial cut-points (50, 100, 150, and 500 counts per minute [cpm]) and GENEActiv vector magnitude cut-points (VMCP; 217 and 386 cpm) were compared against the criterion measure of direct observation. As compared to the criterion, accuracy was determined with mean percent error, Bland-Altman plots, kappa coefficient, sensitivity, and specificity. Receiver operating characteristic curves identified cut-points with greatest discrimination to detect SED and STA. Results: For the GENEActiv, the 217 VMCP was most accurate for measuring SED and STA regardless of which arm wore the monitor. The ActiGraph was most accurate worn on the right hip using 100 and 150 uniaxial cpm to measure STA and 50 cpm to measure SED. Optimal ActiGraph VMCP cut-points to classify SED and STA were ActiGraph 2,000 cpm (left-wrist) and 63 cpm (right hip), respectively. Conclusion: Accuracy of ActiGraph uniaxial cut-points and GENEActiv VMCP is limited in assessing SED in free-living settings. Newer cut-points may increase the accuracy of measuring SED and STA from monitors in free-living settings.
Abderrahmane Rahmani, Pierre Samozino, Jean-Benoit Morin, and Baptiste Morel
barbell flight). The purpose of the present study was thus to assess the reliability and validity of a simple computation method 18 based on easy-to-measure data and the fundamental law of dynamics to determine the force and velocity produced during a ballistic bench-press exercise and, in turn
Pedro Jiménez-Reyes, Pierre Samozino, Fernando Pareja-Blanco, Filipe Conceição, Víctor Cuadrado-Peñafiel, Juan José González-Badillo, and Jean-Benoît Morin
To analyze the reliability and validity of a simple computation method to evaluate force (F), velocity (v), and power (P) output during a countermovement jump (CMJ) suitable for use in field conditions and to verify the validity of this computation method to compute the CMJ force–velocity (F–v) profile (including unloaded and loaded jumps) in trained athletes.
Sixteen high-level male sprinters and jumpers performed maximal CMJs under 6 different load conditions (0–87 kg). A force plate sampling at 1000 Hz was used to record vertical ground-reaction force and derive vertical-displacement data during CMJ trials. For each condition, mean F, v, and P of the push-off phase were determined from both force-plate data (reference method) and simple computation measures based on body mass, jump height (from flight time), and push-off distance and used to establish the linear F–v relationship for each individual.
Mean absolute bias values were 0.9% (± 1.6%), 4.7% (± 6.2%), 3.7% (± 4.8%), and 5% (± 6.8%) for F, v, P, and slope of the F–v relationship (SFv), respectively. Both methods showed high correlations for F–v-profile-related variables (r = .985–.991). Finally, all variables computed from the simple method showed high reliability, with ICC >.980 and CV <1.0%.
These results suggest that the simple method presented here is valid and reliable for computing CMJ force, velocity, power, and F–v profiles in athletes and could be used in practice under field conditions when body mass, push-off distance, and jump height are known.
Dieter Heinrich, Martin Mössner, Peter Kaps, and Werner Nachbauer
The deformation of skis and the contact pressure between skis and snow are crucial factors for carved turns in alpine skiing. The purpose of the current study was to develop and to evaluate an optimization method to determine the bending and torsional stiffness that lead to a given bending and torsional deflection of the ski. Euler-Bernoulli beam theory and classical torsion theory were applied to model the deformation of the ski. Bending and torsional stiffness were approximated as linear combinations of B-splines. To compute the unknown coefficients, a parameter optimization problem was formulated and successfully solved by multiple shooting and least squares data fitting. The proposed optimization method was evaluated based on ski stiffness data and ski deformation data taken from a recently published simulation study. The ski deformation data were used as input data to the optimization method. The optimization method was capable of successfully reproducing the shape of the original bending and torsional stiffness data of the ski with a root mean square error below 1 N m2. In conclusion, the proposed computational method offers the possibility to calculate ski stiffness properties with respect to a given ski deformation.
Glenn Street, Scott McMillan, Wayne Board, Mike Rasmussen, and J. Michael Heneghan
A comprehensive error analysis was performed on the impulse method. To evaluate the potential errors, jump height was recalculated after altering one of the measurement or calculation techniquaes while leaving the others unchanged, and then comparing it to the reference jump height (best estimate of true jump height). Measurement techniques introduced the greatest error. Low-pass filters with cutoff frequencies < 580 Hz led to systematic underestimations of jump height, ≤26%. Low sampling frequencies (<1,080 Hz) caused jump height to be underestimated by ≤4.4%. Computational methods introduced less error. Selecting takeoff too early by using an elevated threshold caused jump height to be overestimated by ≤1.5%. Other potential sources of computational error: (a) duration of body weight averaging period; (b) method of integration; (c) gravity constant; (d) start of integration; (e) duration of offset averaging period; and (f) sample duration, introduced < 1% error to the calculated jump height. Employing the recommended guidelines presented in this study reduces total error to ≤ ±0.76%. Failing to follow the guidelines can lead to average errors as large as 26%.
Movement Bruce C. Elliott * Kevin G. Baxter * Thor F. Besier * 11 1999 15 4 381 395 10.1123/jab.15.4.381 Technical Notes Object Plane Deformation Due to Refraction in Two-Dimensional Underwater Motion Analysis Young-Hoo Kwon * 11 1999 15 4 396 403 10.1123/jab.15.4.396 Research Computational Methods Used in