Zurück zur Liste

Discrimination of Tremor Diseases by Interactive Sonification

ID 54731
Abstract We introduce an interactive sonification approach for the discrimination of tremor diseases. Following up to our previous research, we developed two new sonification methods of measured 3-axes acceleration data of patient’s hands. Prior to sonification, the data is conditioned by Principal Component Analysis (PCA) in order to separate translational and rotational components of the movement. The first sonification implements a vocoder-based method in which energies of relevant frequency bands are used to control individual amplitudes of harmonically tuned oscillators. The second sonification approach is based on Empirical Mode Decomposition (EMD). The input signal is decomposed into Intrinsic Mode Functions (IMFs) whose frequencies and amplitudes control an oscillator bank. In order to enhance the distinct rhythmic qualities of tremor signals in the output, additional amplitude modulation based on various instantaneous energy measures has been applied for both sonifications. An intuitive interface allows to switch interactively between both sonifications and control critical parameters in order to listen to specific aspects of the observed tremor. The results of a pilot study indicate that both sonification methods are able to provide relevant information on tremor data and represent a useful and complementary addition to already available diagnostic tools.
Edition Proceedings of ISon 2016, 5th Interactive Sonification Workshop
Monat 12
Name CITEC, Bielefeld University, Germany
Ort CITEC, Bielefeld University, Bielefeld (Deutschland)
Anmerkung procedure: peer-reviewed
Status veröffentlicht
Art der Arbeit full paper
Publikationsart Konferenzband-Beitrag
Jahr 2016
AutorInnen Weger, M., Pirrò, D., Wankhammer, A., Höldrich, R.
EditorInnen Bresin, R., Hermann, T., Yang, J.