CONTROL OF NEURALPROSTHESES II: EVENT DETECTION USING MACHINE LEARNING
By Williamson, Richard Andrews, Brian Au, Raymond Department of Biomedical Engineering, University of Alberta, Edmonton, Canada; Proceedings of the RESNA '96 Annual Conference, Vol. 16Publication Date: 1996
Machine Learning has been explored as a control method for FES, and modern accelerometers provide a potential sensor that can be implanted or conveniently mounted on an orthotic brace or belt without the necessity for precise anatomical alignment. The theory here under study is that classifying accelerometer pattern with Rough Sets will allow for new control paths in both implant and surface FES. This study attempted to answer the following questions: 1) Are Adaptive Logic Networks (ALN) and/or Rough Sets capable of determining gate phases as from an accelerometer data; 2) Which accelerometer attributes are important for gait phase determination. Tables show recorded data.
Published by: Rehabilitation Engineering & Assistive Technology Society of North America (RESNA) (Website:http://www.resna.org)
Link to text: http://www.resna.org
