Date of Award
Master of Science
As technology continues to advance, there continues to be an interest in applying engineering theories to solve medical problems. Technology has allowed researchers to investigate the properties and characteristics of the human body in hopes of emulating their complexities. The rising interest in the study of biopotentials for use in prosthetic limbs, exoskeletons and machine interfacing has led to a necessity to validate that the understanding of these signals is accurate. The study of biopotentials generated by muscles via a method known as electromyography (EMG) has improved the understanding of effects of rehabilitative treatments, assisted in diagnosing diseases, control of prosthetic limbs and much more. Because of its diagnostic capabilities, it is increasingly necessary to verify that the information extracted from EMG signals is valid. The aim of our work is to subjectively quantify human muscle fatigue using instantaneous frequency and instantaneous amplitude, extracted from EMG signals, using a cosine modulated filter bank (CMFB). Compared to the conventional method of signal analysis techniques, the filter bank allows for study of the frequency components of a signal with regards to time. Surface EMG signals were collected by the Air Force Research Lab (AFRL), from 26 healthy individuals (15 Male and 11 Female) ranging in age from 23 to 27 years. The subjects participated in five 8-hour sessions whilst wearing different helmet load configurations ranging from 3-6 lbs. The participants completed a 100% MVC, pre-and post each 8 hour session. Throughout the 8 hour session, the subjects were instructed to perform an isometric voluntary contraction consisting of neck extension at 70% MVC for a maximum of 3 minutes or until the subject could no longer hold the contraction, whichever occurred first. This movement was performed for every hour up until conclusion of the session. During each session volunteers completed a perceived level of exertion questionnaire prior to recordings and at hours two, four, and six. In our study, we compare the CMFB features IF and IA to the classic EMG parameters MDF, MNF, and RMS to determine if IF and IA can subjectively quantify sensitively local muscle fatigue by using the perceived level of exertion results obtained from subjects. As EMG literature has established, the analysis of sEMG data ranges anywhere from 10s to 30s; because of this, we will be analyzing the last 30 s of each voluntary contraction performed by our subjects. The use of Receiver Operating Characteristic (ROC) curves are implemented to observe the specificity and sensitivity of IF and IA to objectively quantify the development of muscle fatigue.
Received from ProQuest
Liza Stephanie Rodriguez
Rodriguez, Liza Stephanie, "Use of Cosine Modulated Filter Bank to Quantify Human Muscle Fatigue Using Electromyography Signals Obtained During Isometric Voluntary Contractions" (2014). Open Access Theses & Dissertations. 1720.