Fusion of wavelet packet-fuzzy contraction and whale optimization algorithm for hemoglobin spectral feature extraction

Near-infrared spectroscopy (NIRS) technology has a wide range of potential applications in hemoglobin detection, but its accuracy is susceptible to noise and background interference. In order to improve the accuracy of quantitative analysis of hemoglobin in blood NIRS spectra, this study introduces a hybrid method (WPT-FS-WOA), which integrates wavelet packet transform (WPT), fuzzy shrinkage (FS), and Whale Optimization Algorithm (WOA) for hemoglobin feature band extraction. The method uses WPT to decompose the blood near-infrared spectrum at multiple scales, applies FS to denoise the wavelet packet node coefficients for data affiliation assessment, and combines WOA to optimize the wavelet packet nodes at different frequencies, and finally reconstructs the hemoglobin feature spectrum. The analysis of real blood data shows that this method can effectively capture the hemoglobin spectral features compared with the traditional preprocessing techniques.