Multiple band-pass filtering method for improvement on prediction accuracyof linear multivariate analysis

An approach coupling signal processing and partial least-squares regression analysis (PLS) is described in which raw spectral data are processed with a multiple band-pass filter and the filtered spectra are used in a PLS to b uild a calibration model for the analyte of interest. The multiple band-pas s filter is specifically designed for a desired analyte based on the Fourie r frequency characteristics of the pure spectrum of the desired analyte and the spectra of the interference background. It maximizes the ratio of sign al to background. This combined multiple band-pass filtering and PLS method (MFPLS) was evaluated by determining clinically relevant levels of glucose , urea, ethanol, and acetaminophen in simulated human sera, in which trigly ceride was simulated with triacetin; bovine serum albumin and globulin were used to model protein molecules in the serum. The results demonstrate that MFPLS produces better accuracy of prediction than PLS in all instances.