Fuzzy filtering and fuzzy K-means clustering on biomedical sample characterization

In this article, fuzzy logic approach is proposed for sample differentiation using Raman spectroscopy in order to characterize various biomedical samples for decision-making and medical diagnosis. Raman spectra are relatively weak signals whose features are inevitably affected by various types of noises during its calibration process. These noises must be eliminated to an acceptable level. Fuzzy logic method has been widely used to solve uncertainty, imprecision and vague phenomena. As a result, fuzzy filtering is employed for noise filtering so as to enhance the signal to noise ratio. Any raw Raman spectrum has to be preprocessed and normalized prior to further analysis. The resulting intrinsic Raman spectra can be classified into different categories via fuzzy K-means clustering, which is applicable for decision making. A complete fuzzy logic approach is then formulated to characterize several biomedical samples. The long-term research objective is to create a realtime approach for sample analysis using a Raman spectrometer directly mounted at the end-effector of medical robots.