Radial basis function classifiers to help in the diagnosis of the obstructive sleep apnoea syndrome from nocturnal oximetry

• Published in: Medical & biological engineering & computing Vol. 46; no. 4; pp. 323 - 332
• Main Authors: Marcos, J. Víctor, Hornero, Roberto, Álvarez, Daniel, del Campo, Félix, López, Miguel, Zamarrón, Carlos
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer-Verlag 01.04.2008; Springer Nature B.V
• Subjects: Aged; Algorithms; Biomedical and Life Sciences; Biomedical Engineering and Bioengineering; Biomedicine; Body mass index; Computer Applications; Diagnostics; Discriminant analysis; Female; Hospitals; Human Physiology; Humans; Imaging; Male; Medical diagnosis; Methods; Middle Aged; Models, Statistical; Neural networks; Neural Networks (Computer); Original Article; Oximetry; Oxygen - blood; Oxygen saturation; Pattern Recognition, Automated; Radiology; ROC Curve; Sensitivity and Specificity; Signal processing; Sleep apnea; Sleep Apnea, Obstructive - diagnosis; Studies; Radial basis function neural network; Obstructive sleep apnoea syndrome; Data clustering; Nonlinear analysis; Nocturnal oximetry
• ISSN: 0140-0118; 1741-0444
• DOI: 10.1007/s11517-007-0280-0

The aim of this study is to assess the ability of radial basis function (RBF) classifiers as an assistant tool for the diagnosis of the obstructive sleep apnoea syndrome (OSAS). A total of 187 subjects suspected of suffering from OSAS were available for our research. The initial population was divided into training, validation and test sets for deriving and testing our neural classifiers. We used nonlinear features from nocturnal oxygen saturation (SaO 2 ) to perform patients’ classification. We evaluated three different RBF construction techniques based on the following algorithms: k -means (KM), fuzzy c -means (FCM) and orthogonal least squares (OLS). A diagnostic accuracy of 86.1, 84.7 and 85.5% was provided by the networks developed with KM, FCM and OLS, respectively. The three proposed networks achieved an area under the receiver operating characteristic (ROC) curve over 0.90. Our results showed that a useful non-invasive method could be applied to diagnose OSAS from nonlinear features of SaO 2 with RBF classifiers.