Publication Date

12-2009

Type of Culminating Activity

Thesis

Degree Title

Master of Science in Mechanical Engineering

Department

Mechanical and Biomechanical Engineering

Supervisory Committee Chair

Joseph Guarino, Ph.D

Abstract

A study done for the Centers for Disease Control (CDC) found that over 50% of sports related injuries are to the lower extremities [1] with additional studies indicating that females, adolescent through collegiate, have a higher rate of lower extremity sports related injuries than males [2-4]. Conditions surrounding non-contact injuries can be analyzed using ground reaction force (GRF) data from force plates during unanticipated single leg drops, however, the expected gender differences in GRF may not be apparent when viewing data independently in the time or frequency domains.

Graphing the results of a wavelet transform, currently used by scientific, medical, and financial communities [5], allows simultaneous viewing of time, frequency, and magnitude data. In this case, the wavelet transform was chosen over the Short Time Fourier Transform as it is better suited for analyzing transient or aperiodic signals.

The differences in the transformed signals were subtle therefore, further steps were taken to further elucidate the differences. In previous studies coherence between the two matrices (in this case, male and female) was calculated and graphed [6, 7]. The method proposed in this thesis utilizes the coherence method to determine the regions of greatest difference then, a slicing technique is used to view differences at specific frequencies.

Use of the slicing technique illustrates gender differences in the frequencies between 10 and 30 Hz in the X (side to side), Y (front to back), and Z (up and down) axes forces and Z axis moments (about the vertical axis). The greatest differences are shown in the X and Y forces and the moments about the Z axis. In previous studies, due to the magnitude of the forces, the focus was on the Z axis reaction forces. Results of this thesis imply that future studies should also consider on the moments and the reaction forces of the X and Y axes. The results of this thesis combined with studies showing that the frequencies of maximum transmissibility of the lower extremity muscles are less than 50 Hz [8] and the resonant frequencies of the patellar tendon are between 22 and 25 Hz [9], imply that the gender differences seen at low frequencies may correlate with the difference in injury rates.

The second method explored in this thesis utilizes a high quality sound system to listen to the signals after they are converted to a sound file then modified to amplify the areas of difference. The frequencies of difference for this data were less than 40 Hz therefore, software was used to pitch shift the signals making the frequencies of difference more audible. After the pitch shift, audible differences were noted which confirmed the findings of the wavelet transforms.

Download

Share

COinS