Although the scientific field known as infant cry analysis is close to celebrate its 50 anniversary, considering the Scandinavian experience as the starting point, until now none reliable cry-based clinical routines for diagnosis has been successfully achieved. Nevertheless in support of that goal some expectations are appearing when new automatic infant cry classification approaches displaying potentialities for diagnosis purposes are added to the traditional perceptive approach and direct spectrogram observation practice. In this paper we present some of those classification approaches and analyze their potentials for newborn pathologies diagnosis as well. Here we describe some classifiers based on soft computing methodologies, among them; one following the genetic-neural approach, an experimental essay with a hybrid classifier combining the traditional approach based on threshold classification and the classification approach with ANN, one more applying type-2 fuzzy sets for pattern matching, and one using fuzzy relational products to compress the crying patterns before classification. Experiments and some results are also presented. © 2010 Springer-Verlag Berlin Heidelberg.
CITATION STYLE
Reyes-Garcia, C. A., Reyes-Galaviz, O. F., Cano-Ortiz, S. D., Escobedo-Becerro, D. I., Zatarain, R., & Barrón-Estrada, L. (2010). Soft computing approaches to the problem of infant cry classification with diagnostic purposes. Studies in Computational Intelligence, 312, 3–18. https://doi.org/10.1007/978-3-642-15111-8_1
Mendeley helps you to discover research relevant for your work.