Nyquist-Shannon theorem application for Savitzky-Golay smoothing window size parameter determination in bio-optical signals

Abstract

The amount of data generated in the laboratory increases constantly, where a common problem needed to be addressed during the data preparation stage is related to noise. The Savitzky-Golay method has been established as the common smoothing technique, where the process is based on two parameters: polynomial order and window size, which are commonly empirically determined. This work shows how the application of the Nyquist theorem allows to calculate the window size parameter of the Savitzky-Golay smoothing method, in order to process bio-optical signals. Like in the case of mid-IR glucose absorption and cervical spine X-ray images, from which a derivative based analysis allows to either confirm or determine regions of interest for further study.