This paper introduces an approach to classifying swallowing sound signals to detect those patients at risk of aspiration, or choking using rough set methods. An important contribution of a recent study of segmenting the waveform of swallowing sound signals has been the use of the waveform dimension (WD) to describe signal complexity and major changes in signal variance. Prior swallowing sound classification studies have not considered discretization in the extraction of features from swallow sound data tables. In addition, derivation of decision rules for classifying swallowing sounds have not been considered. In the study reported in this paper, both discretization (quantization of real-valued attributes) and non-discretization have been used to achieve attribute reduction and decision rule derivation.
CITATION STYLE
Lazareck, L., & Ramanna, S. (2004). Classification of swallowing sound signals: A rough set approach. In Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science) (Vol. 3066, pp. 679–684). Springer Verlag. https://doi.org/10.1007/978-3-540-25929-9_85
Mendeley helps you to discover research relevant for your work.