Fifth International Workshop on Models and Analysis of Vocal Emissions for Biomedical Applications (MAVEBA 2007)

Florence, Italy
December 13-15, 2007

Detection of Pathological Diseases Using a Parametric Model of Vocal Folds and Neural Networks

P. Chytil (1,2), Cheolwoo Jo (3), K. Drake (4), D. Graville (4), M. Wax (4), M. Pavel (1)

(1) Biomedical Engineering Department, Oregon Health & Science University, Portland, OR, USA
(2) Faculty of Electrical Engineering and Communications, Brno University of Technology, Czech Republic
(3) School of Mechatronics, Changwon National University, Changwon, Gyeongnam, Korea
(4) Department of Otolaryngology, Oregon Health & Science University, Portland, OR, USA

There are a number of clinical conditions that affect directly or indirectly the function of the vocal folds and thereby the pressure waveforms of elicited sounds. If the relationships between the clinical conditions and the voice quality are sufficiently reliable, it should be possible to detect these diseases or disorders. The focus of this paper is to determine the set of features and their values that would characterize the speaker’s state of vocal folds. To the extent that these features can capture the anatomical, physiological, and neurological aspects of the speaker they can be potentially used to mediate an unobtrusive approach to diagnosis. We will show a new approach to this problem, supported with results obtained from two disordered voice corpora.
Index Terms. Model, glottal pulse, pathological voice

Full Paper (reprinted with permission from Firenze University Press)

Bibliographic reference.  Chytil, P. / Jo, Cheolwoo / Drake, K. / Graville, D. / Wax, M. / Pavel, M. (2007): "Detection of pathological diseases using a parametric model of vocal folds and neural networks", In MAVEBA-2007, 71-74.