Article Title



K. DeRosia, M. Norcross, S. Johnson

Oregon State University, Corvallis, OR

Rate of torque development (RTD) is commonly used to measure explosive muscle performance over time intervals associated with injury (e.g., 0-50 ms). Over short time intervals, onset detection may have a large influence on the resulting RTD computations. While manual identification of contraction onset has been claimed as the “gold standard”, its subjective nature brings its reliability into question compared to automated methods. PURPOSE: To determine the within-session reliability and precision of early phase RTD measurements using both automated and manual onset detection methods. METHODS: 33 subjects performed 3 explosive voluntary isometric contractions of the knee extensors on a dynamometer. Trials were sampled at 1000 Hz and filtered using a low-pass, 4th-order Butterworth filter (10 Hz cutoff frequency). Automated onset identification was completed using custom computer software and was defined as the first instance that torque exceeded 3 standard deviations from baseline for at least 200ms. Manual identification involved rescaling the torque-time curve’s y-axis to 1 Nm and x-axis to 500 ms. A vertical cursor was placed at the minima of the last trough before the signal deflected from baseline. Next, the y-axis was rescaled to 0.5 Nm and the x-axis to 25 ms. The vertical cursor was then moved to the point just before the signal deflected from baseline, and that time point was identified as contraction onset. RTD was computed as the slope of the line of best fit of the first 50 ms of the contraction, using the onsets for each method. The intraclass correlation coefficient (ICC2,1) and standard error of measurement (SEM) were calculated for RTD normalized to body mass for both onset methods. RESULTS: RTD calculated using automated onsets (1.838±1.035 Nm • kg-1 • s-1) had an ICC of 0.249 and an SEM of 0.897 Nm • kg-1 • s-1. The RTD calculated using manually identified onsets (0.386±0.360 Nm • kg-1 • s-1) resulted in an ICC of 0.139 and an SEM of 0.334 Nm • kg-1 • s-1. CONCLUSION: Both methods demonstrated poor consistency. The automated onset detection resulted in more reliable but less precise measures of early phase RTD when compared to the manually identified onsets. Further investigation is needed to determine the cause of the observed poor measurement consistency.

Funded through the John C. Erkkila, M.D., Endowment for Health and Human Performance.

This document is currently not available here.