Classifying features of the Intrinsic Mode Functions generated by Empirical Mode Decomposition of isometric force response using a fuzzy classifier

• Published in: 2011 Annual International Conference of the IEEE Engineering in Medicine and Biology Society Vol. 2011; pp. 7849 - 7852
• Main Authors: Stitt, J. P., Newell, K. M.
• Format: Conference Proceeding Journal Article
• Language: English
• Published: United States IEEE 01.01.2011
• Subjects: Adult; Algorithms; Biomechanical Phenomena - physiology; Educational institutions; Feature extraction; Female; Force; Fuzzy Logic; Humans; Indexes; Isometric Contraction - physiology; Male; Power capacitors; Steady-state; Time series analysis; Young Adult
• ISBN: 9781424441211; 1424441218
• ISSN: 1094-687X; 1557-170X; 1558-4615
• DOI: 10.1109/IEMBS.2011.6091934

The analysis of isometric force may provide early detection of certain types of neuropathology such as Parkinson's disease. Our long term goal is to determine if there are detectable differences between model parameters of healthy verses unhealthy individuals. As a first step toward our long-term goal, we studied 24 healthy young adults ages 18 through 24 years, both male and female. The experiments involved the participants exerting isometric force over a range from 5% to 65% of maximal voluntary contraction. The analysis involved the steady-state portion of the recorded time series. Each times-series was decomposed into a set of Intrinsic Mode Functions using Empirical Mode Decomposition. Next, eight features were extracted and used to train a Fuzzy Set Classifier. The participants in this study were assigned to two categories: (1) high strength; and (2) low strength based upon the values of the eight extracted features. Even though the participants were all healthy and young, the features exhibited enough differences to successfully classify 99% of the participants. This finding suggests that, when clinical data become available, the features extracted from the Intrinsic Mode Functions and input into the Fuzzy Set Classifier may be capable of discriminating between healthy individuals and those who are in an early stage of neurodegenerative disease.