Modified sequential floating selection for blood glucose monitoring using near infrared spectral data

To enable non-invasive blood glucose monitoring using near infrared spectroscopy, a modified sequential floating selection method is proposed to remove uninformative data from the spectrum. A linear discriminant function is then used for classification based on selected features. Experiments show that this approach is able to give promising prediction results by classifying near infrared spectroscopic data of blood glucose with good accuracy. Copyright © 2013 Springer Science+Business Media New York.