MURAL - Maynooth University Research Archive Library



    Intelligent Artifact Classification for Ambulatory Physiological Signals


    Sweeney, Kevin and Leamy, Darren J. and Ward, Tomas E. and McLoone, Sean F. (2010) Intelligent Artifact Classification for Ambulatory Physiological Signals. In: 32nd Annual International Conference of the IEEE EMBS, August 31 - September 4, 2010, Buenos Aires, Argentina..

    [img] Download (537kB)


    Share your research

    Twitter Facebook LinkedIn GooglePlus Email more...



    Add this article to your Mendeley library


    Abstract

    Connected health represents an increasingly important model for health-care delivery. The concept is heavily reliant on technology and in particular remote physiological monitoring. One of the principal challenges is the maintenance of high quality data streams which must be collected with minimally intrusive, inexpensive sensor systems operating in difficult conditions. Ambulatory monitoring represents one of the most challenging signal acquisition challenges of all in that data is collected as the patient engages in normal activities of everyday living. Data thus collected suffers from considerable corruption as a result of artifact, much of it induced by motion and this has a bearing on its utility for diagnostic purposes. We propose a model for ambulatory signal recording in which the data collected is accompanied by labeling indicating the quality of the collected signal. As motion is such an important source of artifact we demonstrate the concept in this case with a quality of signal measure derived from motion sensing technology viz. accelerometers. We further demonstrate how different types of artifact might be tagged to inform artifact reduction signal processing elements during subsequent signal analysis. This is demonstrated through the use of multiple accelerometers which allow the algorithm to distinguish between disturbance of the sensor relative to the underlying tissue and movement of this tissue. A brain monitoring experiment utilizing EEG and fNIRS is used to illustrate the concept.

    Item Type: Conference or Workshop Item (Paper)
    Keywords: Intelligent Artifact Classification; Ambulatory Physiological Signals;
    Academic Unit: Faculty of Science and Engineering > Electronic Engineering
    Item ID: 3857
    Depositing User: Dr Tomas Ward
    Date Deposited: 14 Sep 2012 09:31
    Refereed: Yes
    Funders: Irish Research Council for Engineering, Science and Technology (IRCSET), Science Foundation Ireland: Research Frontiers Program 2009, Grant No. 09/RFP/ECE2376
    URI:
      Use Licence: This item is available under a Creative Commons Attribution Non Commercial Share Alike Licence (CC BY-NC-SA). Details of this licence are available here

      Repository Staff Only(login required)

      View Item Item control page

      Downloads

      Downloads per month over past year

      Origin of downloads